1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.504 2007/04/16 11:17:12 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, but only in the
207 case where the client has not sent further data following the '.'
208 (unfortunately, this is allowed). However, in many cases there won't be
209 any further data because there won't be any more messages to send. A call
210 to select() can be used: if it shows that the input is "ready", there is
211 either input waiting, or the socket has been closed. An attempt to read
212 the next input character can distinguish the two cases. Previously, Exim
213 would have sent an OK response which the client would never have see.
214 This could lead to message repetition. This fix should cure that, at
215 least in a lot of common cases.
217 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
218 advertised in response to EHLO.
224 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
225 fixed by 4.65/MH/01 (is this a record?) are fixed:
227 (i) An empty string was always treated as zero by the numeric comparison
228 operators. This behaviour has been restored.
230 (ii) It is documented that the numeric comparison operators always treat
231 their arguments as decimal numbers. This was broken in that numbers
232 starting with 0 were being interpreted as octal.
234 While fixing these problems I realized that there was another issue that
235 hadn't been noticed. Values of message_size_limit (both the global option
236 and the transport option) were treated as octal if they started with 0.
237 The documentation was vague. These values are now always treated as
238 decimal, and I will make that clear in the documentation.
244 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
245 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
248 MH/01 Don't check that the operands of numeric comparison operators are
249 integers when their expansion is in "skipping" mode (fixes bug
250 introduced by 4.64-PH/07).
252 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
253 child addresses, Exim now panics and dies. Previously, because the count
254 is held in a short int, deliveries were likely to be lost. As such a
255 large number of recipients for a single message is ridiculous
256 (performance will be very, very poor), I have chosen to impose a limit
257 rather than extend the field.
263 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
264 leftover -K file (the existence of which was triggered by #402).
265 While we were at it, introduced process PID as part of the -K
266 filename. This should rule out race conditions when creating
269 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
270 processing considerably. Previous code took too long for large mails,
271 triggering a timeout which in turn triggers #401.
273 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
274 in the DK code in transports.c. sendfile() is not really portable,
275 hence the _LINUX specificness.
277 TF/01 In the add_headers option to the mail command in an Exim filter,
278 there was a bug that Exim would claim a syntax error in any
279 header after the first one which had an odd number of characters
282 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
283 callout verification, Exim cached a "reject" for the entire domain. This
284 is correct for most verifications, but it is not correct for a recipient
285 verification with use_sender or use_postmaster set, because in that case
286 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
287 case of MAIL FROM:<> rejection from other early rejections (e.g.
288 rejection of HELO). When verifying a recipient using a non-null MAIL
289 address, the cache is ignored if it shows MAIL FROM:<> rejection.
290 Whatever the result of the callout, the value of the domain cache is
291 left unchanged (for any other kind of callout, getting as far as trying
292 RCPT means that the domain itself is ok).
294 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
297 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
298 manifest itself as EPIPE rather than ECONNECT. When tidying away a
299 session, the daemon ignores ECONNECT errors and logs others; it now
300 ignores EPIPE as well.
302 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
303 (quoted-printable decoding).
305 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
306 later the small subsequent patch to fix an introduced bug.
308 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
310 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
312 PH/08 An error is now given if message_size_limit is specified negative.
314 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
315 to be given (somewhat) arbitrary names.
317 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
320 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
321 miscellaneous code fixes
323 PH/10 Added the log_reject_target ACL modifier to specify where to log
326 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
327 hostname. This is wrong, because it relates to the incoming message (and
328 probably the interface on which it is arriving) and not to the outgoing
329 callout (which could be using a different interface). This has been
330 changed to use the value of the helo_data option from the smtp transport
331 instead - this is what is used when a message is actually being sent. If
332 there is no remote transport (possible with a router that sets up host
333 addresses), $smtp_active_hostname is used.
335 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
336 tweaks were necessary in order to get it to work (see also 21 below):
337 (a) The code assumed that strncpy() returns a negative number on buffer
338 overflow, which isn't the case. Replaced with Exim's string_format()
340 (b) There were several signed/unsigned issues. I just did the minimum
341 hacking in of casts. There is scope for a larger refactoring.
342 (c) The code used strcasecmp() which is not a standard C function.
343 Replaced with Exim's strcmpic() function.
344 (d) The code set only $1; it now sets $auth1 as well.
345 (e) A simple test gave the error "authentication client didn't specify
346 service in request". It would seem that Dovecot has changed its
347 interface. Fortunately there's a specification; I followed it and
348 changed what the client sends and it appears to be working now.
350 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
353 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
354 address A is aliased to B and C, where B exists and C does not. Without
355 -v the output is "A verified" because verification stops after a
356 successful redirection if more than one address is generated. However,
357 with -v the child addresses are also verified. Exim was outputting "A
358 failed to verify" and then showing the successful verification for C,
359 with its parentage. It now outputs "B failed to verify", showing B's
360 parentage before showing the successful verification of C.
362 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
363 look up a TXT record in a specific list after matching in a combined
366 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
367 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
368 they consult the DNS. I had assumed they would set it the way they
369 wanted; and indeed my experiments on Linux seem to show that in some
370 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
371 To be on the safe side, however, I have now made the interface to
372 host_find_byname() similar to host_find_bydns(), with an argument
373 containing the DNS resolver options. The host_find_byname() function now
374 sets these options at its start, just as host_find_bydns() does. The smtp
375 transport options dns_qualify_single and dns_search_parents are passed to
376 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
377 of host_find_byname() use the default settings of RES_DEFNAMES
378 (qualify_single) but not RES_DNSRCH (search_parents).
380 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
381 spool_read_header() do less string testing, by means of a preliminary
382 switch on the second character of optional "-foo" lines. (This is
383 overdue, caused by the large number of possibilities that now exist.
384 Originally there were few.) While I was there, I also converted the
385 str(n)cmp tests so they don't re-test the leading "-" and the first
386 character, in the hope this might squeeze out yet more improvement.
388 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
389 flag allowing group syntax was set by the header_syntax check but not
390 turned off, possible causing trouble later; (2) The flag was not being
391 set at all for the header_verify test, causing "group"-style headers to
392 be rejected. I have now set it in this case, and also caused header_
393 verify to ignore an empty address taken from a group. While doing this, I
394 came across some other cases where the code for allowing group syntax
395 while scanning a header line wasn't quite right (mostly, not resetting
396 the flag correctly in the right place). These bugs could have caused
397 trouble for malformed header lines. I hope it is now all correct.
399 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
400 with the "reply" argument non-NULL. The code, however (which originally
401 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
402 but it didn't always do it. This confused somebody who was copying the
403 code for some other use. I have removed all the tests.
405 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
406 feature that was used to support insecure browsers during the U.S. crypto
407 embargo. It requires special client support, and Exim is probably the
408 only MTA that supported it -- and would never use it because real RSA is
409 always available. This code has been removed, because it had the bad
410 effect of slowing Exim down by computing (never used) parameters for the
411 RSA_EXPORT functionality.
413 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
414 authenticator to fail if there's a tab character in the incoming data
415 (there should never be unless someone is messing about, as it's supposed
416 to be base64-encoded). Also added, on Timo's advice, the "secured" option
417 if the connection is using TLS or if the remote IP is the same as the
418 local IP, and the "valid-client-cert option" if a client certificate has
421 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
422 authenticators. This can be used for authorization after authentication
423 succeeds. (In the case of plaintext, it servers for both authentication
426 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
427 if any retry times were supplied.
429 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
430 connection, where there is no client IP address to check. In this
431 situation, the verify now always succeeds.
433 PH/25 Applied John Jetmore's -Mset patch.
435 PH/26 Added -bem to be like -Mset, but loading a message from a file.
437 PH/27 In a string expansion for a processed (not raw) header when multiple
438 headers of the same name were present, leading whitespace was being
439 removed from all of them, but trailing whitespace was being removed only
440 from the last one. Now trailing whitespace is removed from each header
441 before concatenation. Completely empty headers in a concatenation (as
444 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
445 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
447 PH/29 [Removed. This was a change that I later backed out, and forgot to
448 correct the ChangeLog entry (that I had efficiently created) before
449 committing the later change.]
451 PH/30 Exim was sometimes attempting to deliver messages that had suffered
452 address errors (4xx response to RCPT) over the same connection as other
453 messages routed to the same hosts. Such deliveries are always "forced",
454 so retry times are not inspected. This resulted in far too many retries
455 for the affected addresses. The effect occurred only when there were more
456 hosts than the hosts_max_try setting in the smtp transport when it had
457 the 4xx errors. Those hosts that it had tried were not added to the list
458 of hosts for which the message was waiting, so if all were tried, there
459 was no problem. Two fixes have been applied:
461 (i) If there are any address or message errors in an SMTP delivery, none
462 of the hosts (tried or untried) are now added to the list of hosts
463 for which the message is waiting, so the message should not be a
464 candidate for sending over the same connection that was used for a
465 successful delivery of some other message. This seems entirely
466 reasonable: after all the message is NOT "waiting for some host".
467 This is so "obvious" that I'm not sure why it wasn't done
468 previously. Hope I haven't missed anything, but it can't do any
469 harm, as the worst effect is to miss an optimization.
471 (ii) If, despite (i), such a delivery is accidentally attempted, the
472 routing retry time is respected, so at least it doesn't keep
473 hammering the server.
475 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
476 in ${readsocket because some servers need this prod.
478 PH/32 Added some extra debug output when updating a wait-xxx database.
480 PH/33 The hint "could be header name not terminated by colon", which has been
481 given for certain expansion errors for a long time, was not being given
482 for the ${if def:h_colon_omitted{... case.
484 PH/34 The spec says: "With one important exception, whenever a domain list is
485 being scanned, $domain contains the subject domain." There was at least
486 one case where this was not true.
488 PH/35 The error "getsockname() failed: connection reset by peer" was being
489 written to the panic log as well as the main log, but it isn't really
490 panic-worthy as it just means the connection died rather early on. I have
491 removed the panic log writing for the ECONNRESET error when getsockname()
494 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
495 runs only) independently of the message's sender address. This meant
496 that, if the 4xx error was in fact related to the sender, a different
497 message to the same recipient with a different sender could confuse
498 things. In particualar, this can happen when sending to a greylisting
499 server, but other circumstances could also provoke similar problems.
500 I have changed the default so that the retry time for these errors is now
501 based a combination of the sender and recipient addresses. This change
502 can be overridden by setting address_retry_include_sender=false in the
505 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
506 remote server are returned as part of bounce messages. This was not
507 happening for LMTP over a pipe (the lmtp transport), but now it is the
508 same for both kinds of LMTP.
510 PH/38 Despite being documented as not happening, Exim was rewriting addresses
511 in header lines that were in fact CNAMEs. This is no longer the case.
513 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
514 and queue runs started by the daemon processed all messages. This has
515 been fixed so that -R and -S can now usefully be given with -q<time>.
517 PH/40 Import PCRE release 6.7 (fixes some bugs).
519 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
521 PH/42 Give an error if -q is specified more than once.
523 PH/43 Renamed the variables $interface_address and $interface_port as
524 $received_ip_address and $received_port, to make it clear that these
525 values apply to message reception, and not to the outgoing interface when
526 a message is delivered. (The old names remain recognized, of course.)
528 PH/44 There was no timeout on the connect() call when using a Unix domain
529 socket in the ${readsocket expansion. There now is.
531 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
532 be meaningful with "accept".
534 SC/01 Eximstats V1.43
535 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
537 SC/02 Eximstats V1.44
538 Use a glob alias rather than an array ref in the generated
539 parser. This improves both readability and performance.
541 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
542 Collect SpamAssassin and rejection statistics.
543 Don't display local sender or destination tables unless
544 there is data to show.
545 Added average volumes into the top table text output.
547 SC/04 Eximstats V1.46
548 Collect data on the number of addresses (recipients)
549 as well as the number of messages.
551 SC/05 Eximstats V1.47
552 Added 'Message too big' to the list of mail rejection
553 reasons (thanks to Marco Gaiarin).
555 SC/06 Eximstats V1.48
556 Mainlog lines which have GMT offsets and are too short to
557 have a flag are now skipped.
559 SC/07 Eximstats V1.49 (Alain Williams)
560 Added the -emptyok flag.
562 SC/08 Eximstats V1.50
563 Fixes for obtaining the IP address from reject messages.
565 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
566 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
567 whitesspace changes from 4.64-PH/27
569 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
572 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
573 are found, allow negative numbers in numeric criteria)
575 JJ/06 exipick.20061117.2, added new $message_body_missing variable
577 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
578 to match changes made in 4.64-PH/43
580 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
582 PH/47 Put in an explicit test for a DNS lookup of an address record where the
583 "domain" is actually an IP address, and force a failure. This locks out
584 those revolvers/nameservers that support "A-for-A" lookups, in
585 contravention of the specifications.
587 PH/48 When a host name was looked up from an IP address, and the subsequent
588 forward lookup of the name timed out, the host name was left in
589 $sender_host_name, contrary to the specification.
591 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
592 restricted to single-key lookups, Exim was not diagnosing an error if
593 * or *@ was used with a query-style lookup.
595 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
597 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
598 long ago, but noone interested enough thought of it. Let's just say that
599 the "1.1" means that there are some new functions that weren't there at
600 some point in the past.
602 PH/51 Error processing for expansion failure of helo_data from an smtp
603 transport during callout processing was broken.
605 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
606 tested/used via the -bh/-bhc/-bs options.
608 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
609 bug, fixed in subsequent PCRE releases).
611 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
612 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
614 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
620 SC/01 Use a glob alias rather than an array ref in eximstats generated
621 parser. This improves both readability and performance.
623 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
624 Don't display local sender or destination tables in eximstats unless
625 there is data to show.
626 Added average volumes into the eximstats top table text output.
628 SC/03 Collect data on the number of addresses (recipients) as well
629 as the number of messages in eximstats.
631 TF/01 Correct an error in the documentation for the redirect router. Exim
632 does (usually) call initgroups() when daemonizing.
634 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
635 with consistent privilege compared to when running as a daemon.
637 TF/03 Note in the spec that $authenticated_id is not set for local
638 submissions from trusted users.
640 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
641 Thanks to Dean Brooks <dean@iglou.com> for the patch.
643 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
644 by adding some example configuration directives to the default
645 configuration file. A little bit of work is required to uncomment the
646 directives and define how usernames and passwords are checked, but
647 there is now a framework to start from.
649 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
650 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
651 without this. I don't know how relevant this is to other LDAP libraries.
653 PH/02 Add the verb name to the "unknown ACL verb" error.
655 PH/03 Magnus Holmgren's patch for filter_prepend_home.
657 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
659 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
660 directory not expanded when it should be if an expanded home directory
661 was set for the address (which is overridden by the transport).
663 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
666 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
667 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
668 because it is too late at that time, and has no effect.
670 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
671 security issue with \' (bugzilla #107). I could not use the
672 PQescapeStringConn() function, because it needs a PGconn value as one of
675 PH/08 When testing addresses using -bt, indicate those final addresses that
676 are duplicates that would not cause an additional delivery. At least one
677 person was confused, thinking that -bt output corresponded to deliveries.
678 (Suppressing duplicates isn't a good idea as you lose the information
679 about possibly different redirections that led to the duplicates.)
681 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
682 systems where poll() doesn't work, in particular OS X.
684 PH/10 Added more information to debugging output for retry time not reached.
686 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
687 operations in malware.c.
689 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
692 PH/13 If write_rejectlog was set false when logging was sent to syslog with
693 syslog_duplication set false, log lines that would normally be written
694 both the the main log and to the reject log were not written to syslog at
697 PH/14 In the default configuration, change the use of "message" in ACL warn
698 statements to "add_header".
700 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
701 not followed by a command (e.g. "seen endif").
703 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
704 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
707 PH/17 Added extra conditions to the default value of delay_warning_condition
711 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
712 { match{$h_precedence:}{(?i)bulk|list|junk} } \
713 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
716 The Auto-Submitted: and various List- headers are standardised, whereas I
717 don't think Precedence: ever was.
719 PH/18 Refactored debugging code in route_finduser() to show more information,
720 in particular, the error code if getpwnam() issues one.
722 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
723 This is apparently needed in addition to the PH/07 change above to avoid
724 any possible encoding problems.
726 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
727 but not after initializing Perl.
729 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
730 output them only if debugging. By default they are written stderr,
731 apparently, which is not desirable.
733 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
736 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
739 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
741 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
742 authenticated or an ident call has been made. Suppress the default
743 values for $authenticated_id and $authenticated_sender (but permit -oMai
744 and -oMas) when testing with -bh.
746 PH/24 Re-jigged the order of the tests in the default configuration so that the
747 tests for valid domains and recipients precede the DNS black list and CSA
748 tests, on the grounds that those ones are more expensive.
750 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
751 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
752 This bug exists in every version of Exim that I still have, right back to
755 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
756 However, an attempt to turn on case-sensitivity in a regex key by
757 including (?-i) didn't work because the subject string was already
758 lowercased, and the effects were non-intuitive. It turns out that a
759 one-line patch can be used to allow (?-i) to work as expected.
765 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
766 other effects) broke the use of negated acl sub-conditions.
768 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
771 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
772 "Deny" causes Exim to reject the incoming connection with a 554 error.
773 Unfortunately, if there is a major crisis, such as a disk failure,
774 tcp-wrappers gives "deny", whereas what one would like would be some
775 kind of temporary error. A kludge has been added to help with this.
776 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
777 554 error is used if errno is still zero or contains ENOENT (which occurs
778 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
781 PH/03 Add -lutil to the default FreeBSD LIBS setting.
783 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
784 errors. Otherwise a message that provokes a temporary error (when other
785 messages do not) can cause a whole host to time out.
787 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
788 addresses were routed directly to files or pipes from a redirect router.
789 File deliveries just didn't batch; pipe deliveries might have suffered
792 PH/06 A failure to get a lock for a hints database would erroneously always say
793 "Failed to get write lock", even when it was really a read lock.
795 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
796 of 0600. This has been changed to use the value of the lockfile_mode
797 option (which defaults to 0600).
799 PH/08 Applied small patch from the Sieve maintainer.
801 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
802 folder from quota calculations, a direct delivery into this folder messed
803 up the contents of the maildirsize file. This was because the regex was
804 used only to exclude .Trash (or whatever) when the size of the mailbox
805 was calculated. There was no check that a delivery was happening into an
806 excluded directory. This bug has been fixed by ignoring all quota
807 processing for deliveries into excluded directories.
809 PH/10 Added the maildirfolder_create_regex option to appendfile.
815 PH/01 The code for finding all the local interface addresses on a FreeBSD
816 system running IPv6 was broken. This may well have applied to all BSD
817 systems, as well as to others that have similar system calls. The broken
818 code found IPv4 interfaces correctly, but gave incorrect values for the
819 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
820 that it would not match correctly against @[] and not recognize the IPv6
823 PH/02 The ipliteral router was not recognizing addresses of the form user@
824 [ipv6:....] because it didn't know about the "ipv6:" prefix.
826 PH/03 Added disable_ipv6.
828 PH/04 Changed $reply_address to use the raw form of the headers instead of the
829 decoded form, because it is most often used to construct To: headers
830 lines in autoreplies, and the decoded form may well be syntactically
831 invalid. However, $reply_address has leading white space removed, and all
832 newlines turned into spaces so that the autoreply transport does not
835 PH/05 If group was specified without a user on a router, and no group or user
836 was specified on a transport, the group from the router was ignored.
838 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
839 for visible compile-time settings that can be used to change these
840 numbers, for those that want even more. Backwards compatibility with old
841 spool files has been maintained. However, going back to a previous Exim
842 release will lost any variables that are in spool files.
844 PH/07 Two small changes when running in the test harness: increase delay when
845 passing a TCP/IP connection to a new process, in case the original
846 process has to generate a bounce, and remove special handling of
847 127.0.0.2 (sic), which is no longer necessary.
849 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
850 be the same on different OS.
852 PH/09 Moved a debug statement in filter processing to avoid a race problem when
855 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
856 whether --show-vars was specified or not
858 JJ/02 exipick: Added support for new ACL variable spool format introduced
861 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
862 syntactically invalid From: or Reply-to: line, and a filter used this to
863 generate an autoreply, and therefore failed to obtain an address for the
864 autoreply, Exim could try to deliver to a non-existent relative file
865 name, causing unrelated and misleading errors. What now happens is that
866 it logs this as a hard delivery error, but does not attempt to create a
869 PH/11 The exinext utility has a -C option for testing purposes, but although
870 the given file was scanned by exinext itself; it wasn't being passed on
871 when Exim was called.
873 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
874 an end-of-file indication when reading a command response.
876 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
877 compiled. In many other places in Exim, IPv6 addresses are always
878 recognized, so I have changed this. It also means that IPv4 domain
879 literals of the form [IPV4:n.n.n.n] are now always recognized.
881 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
882 used if the router is not running as root, for example, when verifying at
883 ACL time, or when using -bh. The debugging output from this situation was
884 non-existent - all you got was a failure to exec. I have made two
887 (a) Failures to set uid/gid, the current directory, or a process leader
888 in a subprocess such as that created by queryprogram now generate
889 suitable debugging ouput when -d is set.
891 (b) The queryprogram router detects when it is not running as root,
892 outputs suitable debugging information if -d is set, and then runs
893 the subprocess without attempting to change uid/gid.
895 PH/15 Minor change to Makefile for building test_host (undocumented testing
898 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
899 additional section of a DNS packet that returns MX or SRV records.
900 Instead, it always explicitly searches for A/AAAA records. This avoids
901 major problems that occur when a DNS server includes only records of one
902 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
903 fixed another bug: if SRV records were looked up and the corresponding
904 address records were *not* found in the additional section, the port
905 values from the SRV records were lost.
907 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
908 using the correct key (the original address) when searching the retry
909 rules in order to find which one to use for generating the retry hint.
911 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
912 adding the default one. Similarly, if it contains a Reply-To: header, the
913 errors_reply_to option, if set, is not used.
915 PH/19 When calculating a retry time, Exim used to measure the "time since
916 failure" by looking at the "first failed" field in the retry record. Now
917 it does not use this if it is later than than the arrival time of the
918 message. Instead it uses the arrival time. This makes for better
919 behaviour in cases where some deliveries succeed, thus re-setting the
920 "first failed" field. An example is a quota failure for a huge message
921 when small messages continue to be delivered. Without this change, the
922 "time since failure" will always be short, possible causing more frequent
923 delivery attempts for the huge message than are intended.
924 [Note: This change was subsequently modified - see PH/04 for 4.62.]
926 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
927 $1, $2, $3) because the numerical variables can be reset during some
928 expansion items (e.g. "match"), thereby losing the authentication data.
930 PH/21 Make -bV show the size of off_t variables so that the test suite can
931 decide whether to run tests for quotas > 2G.
933 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
934 mailbox_size, and mailbox_filecount in the appendfile transport. If a
935 filecount value is greater than 2G or if a quota value is greater than 2G
936 on a system where the size of off_t is not greater than 4, a panic error
939 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
940 never match. The debug and -bh output now contains an explicit error
941 message indicating a malformed IPv4 address or mask.
943 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
944 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
947 PH/25 Do not write to syslog when running in the test harness. The only
948 occasion when this arises is a failure to open the main or panic logs
949 (for which there is an explicit test).
951 PH/26 Added the /no_tell option to "control=freeze".
953 PH/27 If a host name lookup failed very early in a connection, for example, if
954 the IP address matched host_lookup and the reverse lookup yielded a name
955 that did not have a forward lookup, an error message of the form "no IP
956 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
957 could be logged. Now it outputs the IP address instead of "NULL".
959 PH/28 An enabling patch from MH: add new function child_open_exim2() which
960 allows the sender and the authenticated sender to be set when
961 submitting a message from within Exim. Since child_open_exim() is
962 documented for local_scan(), the new function should be too.
964 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
965 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
966 results in an empty string is now treated as unset.
968 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
970 PH/31 Added sender_verify_fail logging option.
972 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
973 needed by RFC 822 but not by RFC 2822 was commented out. I have now
974 tidied the source and removed it altogether.
976 PH/33 When a queue run was abandoned because the load average was too high, a
977 log line was always written; now it is written only if the queue_run log
978 selector is set. In addition, the log line for abandonment now contains
979 information about the queue run such as the pid. This is always present
980 in "start" and "stop" lines but was omitted from the "abandon" line.
982 PH/34 Omit spaces between a header name and the colon in the error message that
983 is given when verify = headers_syntax fails (if there are lots of them,
984 the message gets confusing).
986 PH/35 Change the default for dns_check_names_pattern to allow slashes within
987 names, as there are now some PTR records that contain slashes. This check
988 is only to protect against broken name servers that fall over on strange
989 characters, so the fact that it applies to all lookups doesn't matter.
991 PH/36 Now that the new test suite is complete, we can remove some of the
992 special code in Exim that was needed for the old test suite. For example,
993 sorting DNS records because real resolvers return them in an arbitrary
994 order. The new test suite's fake resolver always returns records in the
997 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
998 bounces) except when queue_only is set, to avoid logging races between
999 the different processes.
1001 PH/38 Panic-die if .include specifies a non-absolute path.
1003 PH/39 A tweak to the "H" retry rule from its user.
1005 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1006 a label. They prevented compilation on older perls.
1008 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1009 a warning to be raised on newish perls.
1011 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1012 on queue. Changes to match documented behaviour of showing count of
1013 messages matching specified criteria.
1015 PH/40 Changed the default ident timeout from 30s to 5s.
1017 PH/41 Added support for the use of login_cap features, on those BSD systems
1018 that have them, for controlling the resources used by pipe deliveries.
1020 PH/42 The content-scanning code uses fopen() to create files in which to put
1021 message data. Previously it was not paying any attention to the mode of
1022 the files. Exim runs with umask(0) because the rest of the code creates
1023 files with open(), and sets the required mode explicitly. Thus, these
1024 files were ending up world-writeable. This was not a big issue, because,
1025 being within the spool directory, they were not world-accessible. I have
1026 created a function called modefopen, which takes an additional mode
1027 argument. It sets umask(777), creates the file, chmods it to the required
1028 mode, then resets the umask. All the relevant calls to fopen() in the
1029 content scanning code have been changed to use this function.
1031 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1032 to 24 hours. This avoids potential overflow problems when processing G
1033 and H retry rules. I suspect nobody ever tinkers with this value.
1035 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1037 PH/45 When the plaintext authenticator is running as a client, the server's
1038 challenges are checked to ensure they are valid base64 strings. By
1039 default, the authentication attempt is cancelled if an invalid string is
1040 received. Setting client_ignore_invalid_base64 true ignores these errors.
1041 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1042 they are received. Thus, the responses can be made to depend on the
1043 challenges. If an invalid string is ignored, an empty string is placed in
1046 PH/46 Messages that are created by the autoreply transport now contains a
1047 References: header, in accordance with RFCs 2822 and 3834.
1049 PH/47 Added authenticated_sender_force to the smtp transport.
1051 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1053 PH/49 Installed latest patch from the Sieve maintainer.
1055 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1056 was also set, the appendfile transport was unnecessarily scanning a
1057 directory of message files (e.g. for maildir delivery) to find the count
1058 of files (along with the size), even though it did not need this
1059 information. It now does the scan only if it needs to find either the
1060 size of the count of files.
1062 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1064 PH/52 Two bugs concerned with error handling when the smtp transport is
1067 (i) Exim was not creating retry information for temporary errors given
1068 for individual recipients after the DATA command when the smtp transport
1069 was used in LMTP mode. This meant that they could be retried too
1070 frequently, and not timed out correctly.
1072 (ii) Exim was setting the flag that allows error details to be returned
1073 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1074 recipients that were returned after the DATA command.
1076 PH/53 This is related to PH/52, but is more general: for any failing address,
1077 when detailed error information was permitted to be returned to the
1078 sender, but the error was temporary, then after the final timeout, only
1079 "retry timeout exceeded" was returned. Now it returns the full error as
1080 well as "retry timeout exceeded".
1082 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1083 do this, and (what is worse) MTAs that accept it.
1085 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1086 will now be deprecated.
1088 PH/56 New os.c-cygwin from the Cygwin maintainer.
1090 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1091 formats (previously only available in exim formats via -bpr, -bpru,
1092 and -bpra. Now also available in native and exiqgrep formats)
1094 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1095 with very large, slow to parse queues
1097 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1099 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1101 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1102 responses to authentication challenges, though it was showing the
1103 challenges; (ii) I've removed the CR characters from the debug output for
1106 PH/58 Allow for the insertion of a newline as well as a space when a string
1107 is turned into more than one encoded-word during RFC 2047 encoding. The
1108 Sieve code now uses this.
1110 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1111 data_4xx, lost_connection, tls_required.
1113 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1114 message was being sent as an SMTP response.
1116 PH/61 Add -l and -k options to exicyclog.
1118 PH/62 When verifying, if an address was redirected to one new address, so that
1119 verification continued, and the new address failed or deferred after
1120 having set something in $address_data, the value of $address_data was not
1121 passed back to the ACL. This was different to the case when no
1122 redirection occurred. The value is now passed back in both cases.
1124 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1125 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1126 use login_cap.h, so on its own it isn't the distinguishing feature. The
1127 new name refers directly to the setclassresources() function.
1129 PH/65 Added configuration files for NetBSD3.
1131 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1133 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1134 is preferred over IPv4.
1136 PH/68 The bounce_return_message and bounce_return_body options were not being
1137 honoured for bounces generated during the reception of non-SMTP messages.
1138 In particular, this applied to messages rejected by the ACL. This bug has
1139 been fixed. However, if bounce_return_message is true and bounce_return_
1140 body is false, the headers that are returned for a non-SMTP message
1141 include only those that have been read before the error was detected.
1142 (In the case of an ACL rejection, they have all been read.)
1144 PH/69 The HTML version of the specification is now built in a directory called
1145 spec_html instead of spec.html, because the latter looks like a path with
1146 a MIME-type, and this confuses some software.
1148 PH/70 Catch two compiler warnings in sieve.c.
1150 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1151 function verify_get_ident() calls ip_connect() to connect a socket, but
1152 if the "connect()" function timed out, ip_connect() used to close the
1153 socket. However, verify_get_ident() also closes the socket later, and in
1154 between Exim writes to the log, which may get opened at this point. When
1155 the socket was closed in ip_connect(), the log could get the same file
1156 descriptor number as the socket. This naturally causes chaos. The fix is
1157 not to close the socket in ip_connect(); the socket should be closed by
1158 the function that creates it. There was only one place in the code where
1159 this was missing, in the iplookup router, which I don't think anybody now
1160 uses, but I've fixed it anyway.
1162 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1163 well as to direct DNS lookups. Otherwise the handling of names in host
1164 lists is inconsistent and therefore confusing.
1170 PH/01 Two changes to the default runtime configuration:
1172 (1) Move the checks for relay_from_hosts and authenticated clients from
1173 after to before the (commented out) DNS black list checks.
1175 (2) Add control=submission to the relay_from_hosts and authenticated
1176 clients checks, on the grounds that messages accepted by these
1177 statements are most likely to be submissions.
1179 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1181 (1) Generate an error if the third argument for the ${prvs expansion is
1184 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1187 (3) Reset the variables that are obtained from the first argument of
1188 ${prvscheck and used in the second argument before leaving the code,
1189 because their memory is reclaimed, so using them afterwards may do
1192 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1193 one (it's much easier than Tom thought :-).
1195 (5) Because of (4), we can now allow for the use of $prvscheck_result
1196 inside the third argument.
1198 PH/03 For some reason, the default setting of PATH when running a command from
1199 a pipe transport was just "/usr/bin". I have changed it to
1202 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1203 anything to be listed in the output from -bV.
1205 PH/05 When a filter generated an autoreply, the entire To: header line was
1206 quoted in the delivery log line, like this:
1208 => >A.N.Other <ano@some.domain> <original@ddress> ...
1210 This has been changed so that it extracts the operative address. There
1211 may be more than one such address. If so, they are comma-separated, like
1214 => >ano@some.domain,ona@other.domain <original@ddress> ...
1216 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1217 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1218 not being looked up in the rDNS to get a host name, Exim was showing the
1219 IP address twice in Received: lines, even though the IP addresses were
1220 identical. For example:
1222 Received: from [1.2.3.4] (helo=[1.2.3.4])
1224 However, if the real host name was known, it was omitting the HELO data
1225 if it matched the actual IP address. This has been tidied up so that it
1226 doesn't show the same IP address twice.
1228 PH/07 When both +timestamp and +memory debugging was on, the value given by
1229 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1230 called by the debug printing, thereby overwriting the timestamp buffer.
1231 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1233 PH/08 When the original message was included in an autoreply transport, it
1234 always said "this is a copy of the message, including all the headers",
1235 even if body_only or headers_only was set. It now gives an appropriate
1238 PH/09 Applied a patch from the Sieve maintainer which:
1240 o fixes some comments
1241 o adds the (disabled) notify extension core
1242 o adds some debug output for the result of if/elsif tests
1243 o points to the current vacation draft in the documentation
1244 and documents the missing references header update
1248 o fixes a bug in processing the envelope test (when testing
1249 multiple envelope elements, the last element determinted the
1252 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1253 Electronic Mail") by including:
1255 Auto-submitted: auto-generated
1257 in the messages that it generates (bounce messages and others, such as
1258 warnings). In the case of bounce messages for non-SMTP mesages, there was
1259 also a typo: it was using "Auto_submitted" (underscore instead of
1260 hyphen). Since every message generated by Exim is necessarily in response
1261 to another message, thes have all been changed to:
1263 Auto-Submitted: auto-replied
1265 in accordance with these statements in the RFC:
1267 The auto-replied keyword:
1269 - SHOULD be used on messages sent in direct response to another
1270 message by an automatic process,
1272 - MUST NOT be used on manually-generated messages,
1274 - MAY be used on Delivery Status Notifications (DSNs) and Message
1275 Disposition Notifications (MDNs),
1277 - MUST NOT be used on messages generated by automatic or periodic
1278 processes, except for messages which are automatic responses to
1281 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1282 to the default Received: header definition.
1284 PH/12 Added log selector acl_warn_skipped (default on).
1286 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1287 variables because (a) they are in the wrong storage pool and (b) even if
1288 they were copied, it wouldn't work properly because of the caching.
1290 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1291 checking when decoding. Apparently there are clients that generate
1292 overlong encoded strings. Why am I not surprised?
1294 PH/15 If the first argument of "${if match_address" was not empty, but did not
1295 contain an "@" character, Exim crashed. Now it writes a panic log message
1296 and treats the condition as false.
1298 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1300 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1301 extension "envelope-auth". The code is finished and in agreement with
1302 other implementations, but there is no documentation so far and in fact,
1303 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1304 not changing the active code.
1306 Print executed "if" and "elsif" statements when debugging is used. This
1307 helps a great deal to understand what a filter does.
1309 Document more things not specified clearly in RFC3028. I had all this
1310 sorted out, when out of a sudden new issues came to my mind. Oops."
1312 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1315 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1316 canonical form (as documented). However, after a host name lookup from
1317 the IP address, check_host() was doing a simple string comparison with
1318 addresses acquired from the DNS when checking that the found name did
1319 have the original IP as one of its addresses. Since any found IPv6
1320 addresses are likely to be in abbreviated form, the comparison could
1321 fail. Luckily, there already exists a function for doing the comparison
1322 by converting both addresses to binary, so now that is used instead of
1323 the text comparison.
1325 PH/20 There was another similar case to PH/19, when a complete host name was
1326 given in a host list; looking up its IP address could give an abbreviated
1327 form, whereas the current host's name might or might not be abbreviated.
1328 The same fix has been applied.
1334 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1335 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1338 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1339 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1341 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1343 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1344 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1345 Also, for "Auto-Submitted", treat anything other than "no" as
1346 non-personal, in accordance with RFC 3834. (Previously it treated
1347 anything starting "auto-" as non-personal.)
1349 TF/01 The control=submission/name=... option had a problem with syntax
1350 errors if the name included a slash character. The /name= option
1351 now slurps the rest of the string, so it can include any characters
1352 but it must come last in the list of options (after /sender_retain
1355 PH/05 Some modifications to the interface to the fake nameserver for the new
1363 TK/01 Added the "success_on_redirect" address verification option. See
1364 NewStuff for rationale and an example.
1366 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1368 PH/02 Patch to exigrep to allow it to work on syslog lines.
1370 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1371 fread() to skip over the body file's header line, because in Cygwin the
1372 header line is locked and is inaccessible.
1374 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1375 co-exist for some time) to make it clear that it is the Exim ID that is
1376 referenced, not the Message-ID: header line.
1378 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1379 string_format() function, because snprintf() does not exist on all
1382 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1383 ${stat: expansion item.
1385 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1386 protocol synchronization error", to keep the pedants happy.
1388 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1389 well as for IRIX systems, when gcc is being used. See the host.c source
1392 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1394 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1397 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1398 transport and to the smtp transport in LMTP mode.
1400 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1402 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1403 run a filter in a subprocess. This could lead to confusion in subsequent
1404 lookups in the parent process. There should also be a search_tidyup() at
1405 the end of the subprocess.
1407 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1408 only if the host matched helo_try_verify_hosts, which caused the
1409 verification to occur when the EHLO/HELO command was issued. The ACL just
1410 tested the remembered result. Now, if a previous verification attempt has
1411 not happened, "verify = helo" does it there and then.
1413 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1415 TK/03 Fix log output including CR from clamd.
1417 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1418 exist provoked a memory error which could cause a segfault.
1420 PH/15 Installed PCRE 6.2
1422 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1424 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1425 of the problem. Specifically, suggested +O2 rather than +O1 for the
1428 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1430 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1431 record, the port was not correctly logged when the outgoing_port log
1432 selector was set (it logged the transort's default port).
1434 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1435 fallback_hosts, and "hosts" in the smtp transport.
1437 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1438 host errors such as "Connection refused".
1440 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1441 authentication with radiusclient 0.4.9:
1443 - Error returned from rc_read_config was caught wrongly
1444 - Username/password not passed on to radius server due to wrong length.
1446 The presumption is that some radiusclient API changes for 4.51/PH/17
1447 were not taken care of correctly. The code is still untested by me (my
1448 Linux distribution still has 0.3.2 of radiusclient), but it was
1449 contributed by a Radius user.
1451 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1452 expanding the "port" option of the smtp transport.
1454 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1455 while reading a MIME header. Thanks to Tom Hughes for a patch.
1457 PH/24 Include config.h inside local_scan.h so that configuration settings are
1460 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1461 that in an ACL for RCPT (for example), you can examine exactly what was
1464 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1465 commands, but it was not correctly comparing the address with the actual
1466 client host address. Thus, it would show the EHLO address in Received:
1467 header lines when this was not necessary.
1469 PH/27 Added the % operator to ${eval:}.
1471 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1472 it should be ignoring failures (because with -C, for example, it has lost
1473 privilege). It wasn't ignoring creation failures other than "already
1476 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1479 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1480 because an input line was too long, either on its own, or by virtue of
1481 too many continuations, the temporary file was not being removed, and the
1482 return code was incorrect.
1484 PH/31 Missing "BOOL" in function definition in filtertest.c.
1486 PH/32 Applied Sieve patches from the maintainer.
1488 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1490 PH/33 Added "verify = not_blind".
1492 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1493 Local/Makefile (with some defaults set). These are used in built scripts
1494 such as exicyclog, but they have never been used in the exim_install
1495 script (though there are many overriding facilities there). I have
1496 arranged that the exim_install script now takes note of these two
1499 PH/35 Installed configuration files for Dragonfly.
1501 PH/36 When a locally submitted message by a trusted user did not contain a
1502 From: header, and the sender address was obtained from -f or from an SMTP
1503 MAIL command, and the trusted user did not use -F to supply a sender
1504 name, $originator_name was incorrectly used when constructing a From:
1505 header. Furthermore, $originator_name was used for submission mode
1506 messages from external hosts without From: headers in a similar way,
1507 which is clearly wrong.
1509 PH/37 Added control=suppress_local_fixups.
1511 PH/38 When log_selector = +received_sender was set, and the addition of the
1512 sender made the log line's construction buffer exactly full, or one byte
1513 less than full, an overflow happened when the terminating "\n" was
1516 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1517 when the result of a list match is failure because a DNS lookup failed.
1519 PH/40 RM_COMMAND is now used in the building process.
1521 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1522 the "build-* directories that it finds.
1524 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1525 address in a domain literal was a prefix of an interface address.
1527 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1528 when verifying a sender address, unless rewrite_headers is false.
1530 PH/44 Wrote a long comment about why errors_to addresses are verified as
1531 recipients, not senders.
1533 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1534 the ratelimit ACL was added.
1536 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1538 PH/46 Added extra information about PostgreSQL errors to the error string.
1540 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1541 avoiding the need to install special zones in a real server. This is
1542 backwards compatible; if it can't find the fake resolver, it drops back.
1543 Thus, both old and new test suites can be run.
1545 TF/02 Added util/ratelimit.pl
1547 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1548 clock is set back in time.
1550 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1551 Candler <B.Candler@pobox.com>.
1553 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1554 OK for addresses that are the result of redirections.
1556 PH/48 A number of further additions for the benefit of the new test suite,
1557 including a fake gethostbyname() that interfaces to the fake DNS resolver
1560 TF/06 The fix for widen_domains has also been applied to qualify_single and
1561 search_parents which are the other dnslookup options that can cause
1564 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1567 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1569 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1570 service" if it encountered an MX record with an empty target hostname.
1571 The message is now "an MX or SRV record indicated no SMTP service".
1573 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1574 if the DNS of the sending site is misconfigured. This is quite a
1575 common situation. This change restores the behaviour of treating a
1576 helo verification defer as a failure.
1578 PH/51 If self=fail was set on a router, the bounce message did not include the
1579 actual error message.
1585 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1587 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1588 command itself ended in error, the underlying message about the transport
1589 filter timeout was being overwritten with the pipe command error. Now the
1590 underlying error message should be appended to the second error message.
1592 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1594 PH/02 Reduce the amount of output that "make" produces by default. Full output
1595 can still be requested.
1597 PH/03 The warning log line about a condition test deferring for a "warn" verb
1598 was being output only once per connection, rather than after each
1599 occurrence (because it was using the same function as for successful
1600 "warn" verbs). This seems wrong, so I have changed it.
1602 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1603 it should not have, which might have caused a crash in the right
1604 circumstances, but probably never did.
1606 PH/04 Installed a modified version of Tony Finch's patch to make submission
1607 mode fix the return path as well as the Sender: header line, and to
1608 add a /name= option so that you can make the user's friendly name appear
1611 TF/03 Added the control = fakedefer ACL modifier.
1613 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1614 Mark Lowes for thorough testing.
1616 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1618 TK/03 Merged latest SRS patch from Miles Wilton.
1620 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1621 with the definition in sysexits.h (which is #included earlier).
1622 Fortunately, Exim does not actually use EX_OK. The code used to try to
1623 preserve the sysexits.h value, by assumimg that macro definitions were
1624 scanned for macro replacements. I have been disabused of this notion,
1625 so now the code just undefines EX_OK before #including unistd.h.
1627 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1628 in the smtp transport. When a block could not be written in a single
1629 write() function, the timeout was being re-applied to each part-write.
1630 This seems wrong - if the receiver was accepting one byte at a time it
1631 would take for ever. The timeout is now adjusted when this happens. It
1632 doesn't have to be particularly precise.
1634 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1635 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1637 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1638 without a domain if the check to <postmaster@domain> fails.
1640 SC/01 Eximstats: added -xls and the ability to specify output files
1641 (patch written by Frank Heydlauf).
1643 SC/02 Eximstats: use FileHandles for outputing results.
1645 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1647 SC/04 Eximstats: fixed display of large numbers with -nvr option
1649 SC/05 Eximstats: fixed merging of reports with empty tables.
1651 SC/06 Eximstats: added the -include_original_destination flag
1653 SC/07 Eximstats: removed tabs and trailing whitespace.
1655 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1657 TK/06 MBOX spool code: Add real "From " MBOX separator line
1658 so the .eml file is really in mbox format (even though
1659 most programs do not really care). Patch from Alex Miller.
1661 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1662 The latter is generated from $received_to and is only set if the
1663 message has one envelope recipient. SA can use these headers,
1664 obviously out-of-the-box. Patch from Alex Miller.
1666 PH/08 The ${def test on a variable was returning false if the variable's
1667 value was "0", contrary to what the specification has always said!
1668 The result should be true unless the variable is empty.
1670 PH/09 The syntax error of a character other than { following "${if
1671 def:variable_name" (after optional whitespace) was not being diagnosed.
1672 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1673 accidental colon was present, for example, could give incorrect results.
1675 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1676 result is used (not including appendfile, where other changes are about
1679 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1680 This involved changing a lot of size variables from int to off_t. It
1681 should work with maildirs and everything.
1683 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1684 spamd dying while we are connected to it.
1686 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1689 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1690 function for BDB 4.3.
1692 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1694 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1695 the sources for PCRE have been re-arranged and more files are now
1698 PH/15 The code I had for printing potentially long long variables in PH/11
1699 above was not the best (it lost precision). The length of off_t variables
1700 is now inspected at build time, and an appropriate printing format (%ld
1701 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1702 to be "long long int" or "long int". This is needed for the internal
1703 formatting function string_vformat().
1705 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1706 the configuration file to be ":syslog", then the script "guesses" where
1707 the logs files are, rather than using the compiled in default. In our
1708 case the guess is not the same as the compiled default, so the script
1709 suddenly stopped working when I started to use syslog. The patch checks
1710 to see if log_file_path is "". If so, it attempts to read it from exim
1711 with no configuration file to get the compiled in version, before it
1712 falls back to the previous guessing code."
1714 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1715 implementing BATV in an Exim configuration. See NewStuff for the gory
1718 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1719 Makefile that are specific to HP-UX.
1721 PH/18 If the "use_postmaster" option was set for a recipient callout together
1722 with the "random" option, the postmaster address was used as the MAIL
1723 FROM address for the random test, but not for the subsequent recipient
1724 test. It is now used for both.
1726 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1727 patch removes a few documentation additions to RFC 3028, because the
1728 latest draft now contains them. It adds the new en;ascii-case comparator
1729 and a new error check for 8bit text in MIME parts. Comparator and
1730 require names are now matched exactly. I enabled the subaddress
1731 extension, but it is not well tested yet (read: it works for me)."
1733 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1734 rework some of the code of TK/09 above to avoid the hardwired use of
1735 "%lld" and "long long". Replaced the call to snprintf() with a call to
1738 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1739 records point to non-existent hosts", "retry timeout exceeded", and
1740 "retry time not reached for any host after a long failure period".
1742 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1743 experimental DomainKeys support:
1745 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1746 (2) On an error such as an illegally used "control", the wrong name for
1747 the control was given.
1749 These problems did NOT occur unless DomainKeys support was compiled.
1751 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1753 PH/24 Added ${if match_ip condition.
1755 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1756 databases so that it will be absolutely obvious if a crash occurs in the
1757 DB library. This is a regular occurrence (often caused by mis-matched
1760 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1761 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1762 picked up on a user's system that detects such things. There doesn't seem
1763 to be a gcc warning option for this - only an attribute that has to be
1764 put on the function's prototype. It seems that in Fedora Core 4 they have
1765 set this on a number of new functions. No doubt there will be more in due
1768 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1769 specify a transport. However, if an address that was verified by such a
1770 router was the subject of a callout, Exim crashed because it tried to
1771 read the rcpt_include_affixes from the non-existent transport. Now it
1772 just assumes that the setting of that option is false. This bug was
1773 introduced by 4.51/PH/31.
1775 PH/28 Changed -d+all to exclude +memory, because that information is very
1776 rarely of interest, but it makes the output a lot bigger. People tend to
1777 do -d+all out of habit.
1779 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1780 code in os-type was giving problems when libc.so lives in lib64, like on
1783 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1784 aren't the modern standard, and it seems that some systems' include files
1785 don't always have them. Exim was already checking for some of the newer
1786 ones like T_AAAA, and defining it itself. I've added checks for all the
1787 record types that Exim uses.
1789 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1790 not automatically generating a new one, as it is supposed to. This
1791 prevented TLS from working. If the file did exist, but contained invalid
1792 data, a new version was generated, as expected. It was only the case of a
1793 non-existent file that was broken.
1795 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1796 with a change in libdomainkeys > 0.64.
1798 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1799 from DNS. If the selector record carries the flag, it now has
1800 precedence over the domain-wide flag.
1802 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1804 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1805 the use of a port name that isn't defined in /etc/services) occurred, the
1806 message was deferred as in a normal delivery, and thus remained on the
1807 spool, instead of being failed because of the mua_wrapper setting. This
1808 is now fixed, and I tidied up some of the mua_wrapper messages at the
1811 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1812 the messages in a hash of arrays rather than using individual hashes.
1813 This is a bit cleaner and results in dramatic memory savings, albeit
1814 at a slight CPU cost.
1816 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1817 as requested by Marc Sherman.
1819 SC/10 Eximstats: added histograms for user specified patterns as requested
1822 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1824 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1825 fopen() in the content-scanning modules that did not already have it.
1831 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1832 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1834 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1836 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1838 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1839 to test Sieve filters that use "vacation".
1841 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1842 that changes the way the GnuTLS parameters are stored in the cache file.
1843 The new format can be generated externally. For backward compatibility,
1844 if the data in the cache doesn't make sense, Exim assumes it has read an
1845 old-format file, and it generates new data and writes a new file. This
1846 means that you can't go back to an older release without removing the
1849 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1850 work if there are any delivery delays because "one_time" forces the
1851 parent to be marked "delivered", so its unseen clone is never tried
1852 again. For this reason, Exim now forbids the simultaneous setting of
1855 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1856 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1857 entry if you want to know the details. The fix, however, neglected to
1858 consider the case when local delivery batching is involved. The test for
1859 "previously delivered" was not happening when checking to see if an
1860 address could be batched with a previous (undelivered) one; under
1861 certain circumstances this could lead to multiple deliveries to the same
1864 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1865 in its include files, and this causes problems building Exim.
1867 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1868 header_syntax) but Exim was just ignoring anything given after a slash.
1869 In particular, this caused confusion with an attempt to use "verify =
1870 reverse_host_lookup/defer_ok". An error is now given when options are
1871 supplied for verify items that do not have them. (Maybe reverse_host_
1872 lookup should have a defer_ok option, but that's a different point.)
1874 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1875 defined by RFC 821) to 2048, because there were problems with some AUTH
1876 commands, and RFC 1869 says the size should be increased for extended
1877 SMTP commands that take arguments.
1879 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1882 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1883 "unknown" error; now it says that the functionality isn't in the binary.
1885 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1886 an address' error message when a string expansion fails (syntax or
1887 whatever). Otherwise the password may appear in the log. Following change
1888 PH/42 below, there is no longer a chance of it appearing in a bounce
1891 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1893 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1894 of its addresses was ever tried. (Bugzilla bug #2.)
1896 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1897 the result incorrectly in the debug output. (It correctly added a newline
1898 to what was transported.)
1900 TF/01 Added $received_time.
1902 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1903 commented out examples of how to interface to a virus scanner and to
1904 SpamAssassin. Also added commented examples of av_scanner and
1905 spamd_address settings.
1907 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1908 and controls are allowed in which ACLs. There were a couple of minor
1909 errors. Some of the entries in the conditions table (which is a table of
1910 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1911 negation of where the condition IS allowed.
1913 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1915 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1916 header file does not have a version number, so I've had to invent a new
1917 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1918 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1919 radiusclient), but it was contributed by a Radius user.
1921 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1922 files or directories, for OpenSSL.
1924 PH/19 When an Exim process that is running as root has to create an Exim log
1925 file, it does so in a subprocess that runs as exim:exim so as to get the
1926 ownership right at creation (otherwise, other Exim processes might see
1927 the file with the wrong ownership). There was no test for failure of this
1928 fork() call, which would lead to the process getting stuck as it waited
1929 for a non-existent subprocess. Forks do occasionally fail when resources
1930 run out. I reviewed all the other calls to fork(); they all seem to check
1933 PH/20 When checking for unexpected SMTP input at connect time (before writing
1934 the banner), Exim was not dealing correctly with a non-positive return
1935 from the read() function. If the client had disconnected by this time,
1936 the result was a log entry for a synchronization error with an empty
1937 string after "input=" when read() returned zero. If read() returned -1
1938 (an event I could not check), uninitialized data bytes were printed.
1939 There were reports of junk text (parts of files, etc) appearing after
1942 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1944 PH/22 Added support for macro redefinition, and (re)definition in between
1945 driver and ACL definitions.
1947 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1948 forgetting to use the resulting value; it was using the unexpanded value.
1950 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1951 hadn't been configured. The fix is from Juergen Kreileder, who
1952 understands it better than I do:
1954 "Here's what I see happening with three configured cyrus_sasl
1955 authenticators configured (plain, login, cram-md5):
1957 On startup auth_cyrus_sasl_init() gets called for each of these.
1958 This means three calls to sasl_listmech() without a specified mech_list.
1959 => SASL tests which mechs of all available mechs actually work
1960 => three warnings about OTP not working
1961 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1963 With the patch, sasl_listmech() also gets called three times. But now
1964 SASL's mech_list option is set to the server_mech specified in the the
1965 authenticator. Or in other words, the answer from sasl_listmech()
1966 gets limited to just the mech you're testing for (which is different
1968 => the return list contains just 'plain' or 'login', 'cram-md5' or
1969 nothing depending on the value of ob->server_mech.
1971 I've just tested the patch: Authentication still works fine,
1972 unavailable mechs specified in the exim configuration are still
1973 caught, and the auth.log warnings about OTP are gone."
1975 PH/25 When debugging is enabled, the contents of the command line are added
1976 to the debugging output, even when log_selector=+arguments is not
1979 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1980 answer is "GNU", and only if the return is "GNU/something" is the answer
1983 PH/27 $acl_verify_message is now set immediately after the failure of a
1984 verification in an ACL, and so is available in subsequent modifiers. In
1985 particular, the message can be preserved by coding like this:
1987 warn !verify = sender
1988 set acl_m0 = $acl_verify_message
1990 Previously, $acl_verify_message was set only while expanding "message"
1991 and "log_message" when a very denied access.
1993 PH/28 Modified OS/os.c-Linux with
1995 -#ifndef OS_LOAD_AVERAGE
1996 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1998 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1999 nomenclature these days.)
2001 PH/29 Installed patch from the Sieve maintainer that adds the options
2002 sieve_useraddress and sieve_subaddress to the redirect router.
2004 PH/30 In these circumstances:
2005 . Two addresses routed to the same list of hosts;
2006 . First host does not offer TLS;
2007 . First host accepts first address;
2008 . First host gives temporary error to second address;
2009 . Second host offers TLS and a TLS session is established;
2010 . Second host accepts second address.
2011 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2012 and peerdn, if requested) that were in fact used only for the second
2015 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2016 attention to any local part prefix or suffix that was matched by the
2017 router that accepted the address. It now behaves in the same way as it
2018 does for delivery: the affixes are removed from the local part unless
2019 rcpt_include_affixes is set on the transport.
2021 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2022 timeout during the DATA phase of an incoming message.
2024 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2025 applied a patch sanctioned by the Sieve maintainer.
2027 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2028 the uid or gid is negative. A case of a negative gid caused this to be
2029 noticed. The fix allows for either to be negative.
2031 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2032 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2035 PH/36 The change PH/12 above was broken. Fixed it.
2037 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2038 the grounds that routing the same address twice would always produce the
2039 same answer. This might have been true once, but it is certainly no
2040 longer true now. Routing a child address may depend on the previous
2041 routing that produced that child. Some complicated redirection strategies
2042 went wrong when messages had multiple recipients, and made Exim's
2043 behaviour dependent on the order in which the addresses were given.
2045 I have moved the duplicate checking until after the routing is complete.
2046 Exim scans the addresses that are assigned to local and remote
2047 transports, and removes any duplicates. This means that more work will be
2048 done, as duplicates will always all be routed, but duplicates are
2049 presumably rare, so I don't expect this is of any significance.
2051 For deliveries to pipes, files, and autoreplies, the duplicate checking
2052 still happens during the routing process, since they are not going to be
2055 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2056 It corrects a timeout issue with spamd. This is Ian's comment: "The
2057 background is that sometimes spamd either never reads data from a
2058 connection it has accepted, or it never writes response data. The exiscan
2059 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2060 blindly assumes that writes won't block so it may never time out."
2062 PH/39 Allow G after quota size as well as K and M.
2064 PH/40 The value set for $authenticated_id in an authenticator may not contain
2065 binary zeroes or newlines because the value is written to log lines and
2066 to spool files. There was no check on this. Now the value is run through
2067 the string_printing() function so that such characters are converted to
2068 printable escape sequences.
2070 PH/41 $message_linecount is a new variable that contains the total number of
2071 lines in the message. Compare $body_linecount, which is the count for the
2074 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2075 bounce and delay warning messages, except in certain special cases, which
2078 (a) An SMTP error message from a remote host;
2079 (b) A message specified in a :fail: redirection;
2080 (c) A message specified in a "fail" command in a system filter;
2081 (d) A message specified in a FAIL return from the queryprogram router;
2082 (e) A message specified by the cannot_route_message router option.
2084 In these cases only, Exim does include the error details in bounce and
2085 warning messages. There are also a few cases where bland messages such
2086 as "unrouteable address" or "local delivery error" are given.
2088 PH/43 $value is now also set for the "else" part of a ${run expansion.
2090 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2091 being worked on, but at least Exim now implements the latest version to
2094 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2095 process to complete was treated as a delivery failure, a timeout while
2096 writing the message to the pipe was logged, but erroneously treated as a
2097 successful delivery. Such timeouts include transport filter timeouts. For
2098 consistency with the overall process timeout, these timeouts are now
2099 treated as errors, giving rise to delivery failures by default. However,
2100 there is now a new Boolean option for the pipe transport called
2101 timeout_defer, which, if set TRUE, converts the failures into defers for
2102 both kinds of timeout. A transport filter timeout is now identified in
2105 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2106 systems where "make" and "gmake" are different, calling "gmake" at top
2107 level broke things. I've arranged for the value of $(MAKE) to be passed
2108 from the Makefile to this script so that it can call the same version of
2112 A note about Exim versions 4.44 and 4.50
2113 ----------------------------------------
2115 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2116 changes of various kinds. As a consequence, a big documentation update was
2117 needed. This delayed the release for rather longer than seemed good, especially
2118 in the light of a couple of (minor) security issues. Therefore, the changes
2119 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2120 release. So 4.44 and 4.50 are in effect two different branches that both start
2123 I have left the 4.50 change log unchanged; it contains all the changes since
2124 4.43. The change log for 4.44 is below; many of its items are identical to
2125 those for 4.50. This seems to be the most sensible way to preserve the
2126 historical information.
2132 1. Minor wording change to the doc/README.SIEVE file.
2134 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2135 computation of the current number of files was incorrect.
2137 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2138 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2141 4. Give more explanation in the error message when the command for a transport
2142 filter fails to execute.
2144 5. There are several places where Exim runs a non-Exim command in a
2145 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2146 was being done only for the command run by the queryprogram router. It is
2147 now done for all such subprocesses. The other cases are: ${run, transport
2148 filters, and the commands run by the lmtp and pipe transports.
2150 6. Added CONFIGURE_GROUP build-time option.
2152 7. Some older OS have a limit of 256 on the maximum number of file
2153 descriptors. Exim was using setrlimit() to set 1000 as a large value
2154 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2155 systems. I've change it so that if it can't get 1000, it tries for 256.
2157 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2158 was an oversight, and furthermore, ever since the addition of extra
2159 controls (e.g. 4.43/32), the checks on when to allow different forms of
2160 "control" were broken. There should now be diagnostics for all cases when a
2161 control that does not make sense is encountered.
2163 9. Added the /retain_sender option to "control=submission".
2165 10. $recipients is now available in the predata ACL (oversight).
2167 11. Tidy the search cache before the fork to do a delivery from a message
2168 received from the command line. Otherwise the child will trigger a lookup
2169 failure and thereby defer the delivery if it tries to use (for example) a
2170 cached ldap connection that the parent has called unbind on.
2172 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2173 of $address_data from the recipient verification was clobbered by the
2174 sender verification.
2176 13. The value of address_data from a sender verification is now available in
2177 $sender_address_data in subsequent conditions in the ACL statement.
2179 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2181 15. Added a new option "connect=<time>" to callout options, to set a different
2184 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2185 was its contents. (It was OK if the option was not defined at all.)
2187 17. A "Completed" log line is now written for messages that are removed from
2188 the spool by the -Mrm option.
2190 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2191 information about exactly what failed.
2193 19. Added -dd to debug only the daemon process.
2195 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2196 handles timeouts, both on the server side and network timeouts. Renamed the
2197 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2199 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2200 It is now set to "smtps".
2202 22. $host_address is now set to the target address during the checking of
2203 ignore_target_hosts.
2205 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2206 being passed; this would have caused $sender_host_name to have been used if
2207 matching the list had actually called for a host name (not very likely,
2208 since this list is usually IP addresses). A host name is now passed as
2211 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2212 code that specifies a non-restarting handler (typically sigaction() in
2213 modern systems) in an attempt to fix a rare and obscure crash bug.
2215 25. Narrowed the window for a race in the daemon that could cause it to ignore
2216 SIGCHLD signals. This is not a major problem, because they are used only to
2217 wake it up if nothing else does.
2219 26. A malformed maildirsize file could cause Exim to calculate negative values
2220 for the mailbox size or file count. Odd effects could occur as a result.
2221 The maildirsize information is now recalculated if the size or filecount
2224 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2225 support for a long time. Removed HAVE_SYS_VFS_H.
2227 28. Installed the latest version of exipick from John Jetmore.
2229 29. In an address list, if the pattern was not a regular expression, an empty
2230 subject address (from a bounce message) matched only if the pattern was an
2231 empty string. Non-empty patterns were not even tested. This was the wrong
2232 because it is perfectly reasonable to use an empty address as part of a
2233 database query. An empty address is now tested by patterns that are
2234 lookups. However, all the other forms of pattern expect the subject to
2235 contain a local part and a domain, and therefore, for them, an empty
2236 address still always fails if the pattern is not itself empty.
2238 30. Exim went into a mad DNS loop when attempting to do a callout where the
2239 host was specified on an smtp transport, and looking it up yielded more
2240 than one IP address.
2242 31. Re-factored the code for checking spool and log partition space into a
2243 function that finds that data and another that does the check. The former
2244 is then used to implement four new variables: $spool_space, $log_space,
2245 $spool_inodes, and $log_inodes.
2247 32. The RFC2047 encoding function was originally intended for short strings
2248 such as real names; it was not keeping to the 75-character limit for
2249 encoded words that the RFC imposes. It now respects the limit, and
2250 generates multiple encoded words if necessary. To be on the safe side, I
2251 have increased the buffer size for the ${rfc2047: expansion operator from
2254 33. It is now permitted to omit both strings after an "if" condition; if the
2255 condition is true, the result is "true". As before, when the second string
2256 is omitted, a false condition yields an empty string. This makes it less
2257 cumbersome to write custom ACL and router conditions.
2259 34. Failure to deliver a bounce message always caused it to be frozen, even if
2260 there was an errors_to setting on the router. The errors_to setting is now
2263 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2264 canonical form (fully expanded) before being placed in
2265 $sender_host_address.
2267 36. The table in the code that translates DNS record types into text (T_A to
2268 "A" for instance) was missing entries for NS and CNAME. It is just possible
2269 that this could have caused confusion if both these types were looked up
2270 for the same domain, because the text type is used as part of Exim's
2271 per-process caching. But the chance of anyone hitting this buglet seems
2274 37. The dnsdb lookup has been extended in a number of ways.
2276 (1) There is a new type, "zns", which walks up the domain tree until it
2277 finds some nameserver records. It should be used with care.
2279 (2) There is a new type, "mxh", which is like "mx" except that it returns
2280 just the host names, not the priorities.
2282 (3) It is now possible to give a list of domains (or IP addresses) to be
2283 looked up. The behaviour when one of the lookups defers can be
2284 controlled by a keyword.
2286 (4) It is now possible to specify the separator character for use when
2287 multiple records are returned.
2289 38. The dnslists ACL condition has been extended: it is now possible to supply
2290 a list of IP addresses and/or domains to be looked up in a particular DNS
2293 39. Added log_selector=+queue_time_overall.
2295 40. When running the queue in the test harness, wait just a tad after forking a
2296 delivery process, to get repeatability of debugging output.
2298 41. Include certificate and key file names in error message when GnuTLS fails
2299 to set them up, because the GnuTLS error message doesn't include the name
2300 of the failing file when there is a problem reading it.
2302 42. Allow both -bf and -bF in the same test run.
2304 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2306 44. The "Exiscan patch" is now merged into the mainline Exim source.
2308 45. Sometimes the final signoff response after QUIT could fail to get
2309 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2310 < 0 before doing a fflush(). This bug looks as though it goes back to the
2311 introduction of TLS in release 3.20, but "sometimes" must have been rare
2312 because the tests only now provoked it.
2314 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2315 (this can affect the format of dates).
2317 47. exim_tidydb, when checking for the continued existence of a message for
2318 which it has found a message-specific retry record, was not finding
2319 messages that were in split spool directories. Consequently, it was
2320 deleting retry records that should have stayed in existence.
2322 48. Steve fixed some bugs in eximstats.
2324 49. The SPA authentication driver was not abandoning authentication and moving
2325 on to the next authenticator when an expansion was forced to fail,
2326 contradicting the general specification for all authenticators. Instead it
2327 was generating a temporary error. It now behaves as specified.
2329 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2330 (the order specifies the preference for clients). The order is now AES256,
2331 AES128, 3DES, ARCFOUR128.
2333 51. Small patch to Sieve code - explicitly set From: when generating an
2336 52. Exim crashed if a remote delivery caused a very long error message to be
2337 recorded - for instance if somebody sent an entire SpamAssassin report back
2338 as a large number of 550 error lines. This bug was coincidentally fixed by
2339 increasing the size of one of Exim's internal buffers (big_buffer) that
2340 happened as part of the Exiscan merge. However, to be on the safe side, I
2341 have made the code more robust (and fixed the comments that describe what
2344 53. Now that there can be additional text after "Completed" in log lines (if
2345 the queue_time_overall log selector is set), a one-byte patch to exigrep
2346 was needed to allow it to recognize "Completed" as not the last thing in
2349 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2350 patch that reportedly fixes this has been added. I am not expert enough to
2351 create a test for it. This is what the patch creator wrote:
2353 "I found a little strange behaviour of ldap code when working with
2354 Windows 2003 AD Domain, where users was placed in more than one
2355 Organization Units. When I tried to give exim partial DN, the exit code
2356 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2357 But simultaneously result of request was absolutely normal ldap result,
2358 so I produce this patch..."
2360 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2361 REFERENCE, so I have modified the code to exclude the patch when that macro
2364 55. Some experimental protocols are using DNS PTR records for new purposes. The
2365 keys for these records are domain names, not reversed IP addresses. The
2366 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2367 leaves it alone. Component reversal etc. now happens only for IP addresses.
2370 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2372 57. Double the size of the debug message buffer (to 2048) so that more of very
2373 long debug lines gets shown.
2375 58. The exicyclog utility now does better if the number of log files to keep
2376 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2378 59. Two changes related to the smtp_active_hostname option:
2380 (1) $smtp_active_hostname is now available as a variable.
2381 (2) The default for smtp_banner uses $smtp_active_hostname instead
2382 of $primary_hostname.
2384 60. The host_aton() function is supposed to be passed a string that is known
2385 to be a valid IP address. However, in the case of IPv6 addresses, it was
2386 not checking this. This is a hostage to fortune. Exim now panics and dies
2387 if the condition is not met. A case was found where this could be provoked
2388 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2389 components; fortuitously, this particular loophole had already been fixed
2390 by change 4.50/55 above.
2392 If there are any other similar loopholes, the new check in host_aton()
2393 itself should stop them being exploited. The report I received stated that
2394 data on the command line could provoke the exploit when Exim was running as
2395 exim, but did not say which command line option was involved. All I could
2396 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2397 running as the user.
2400 61. There was a buffer overflow vulnerability in the SPA authentication code
2401 (which came originally from the Samba project). I have added a test to the
2402 spa_base64_to_bits() function which I hope fixes it.
2405 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2406 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2408 63. The daemon start-up calls getloadavg() while still root for those OS that
2409 need the first call to be done as root, but it missed one case: when
2410 deliver_queue_load_max is set with deliver_drop_privilege. This is
2411 necessary for the benefit of the queue runner, because there is no re-exec
2412 when deliver_drop_privilege is set.
2414 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2415 This has been fixed.
2417 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2418 was in use, was not putting the data itself into the right store pool;
2419 consequently, it could be overwritten for a subsequent message in the same
2420 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2423 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2425 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2426 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2427 the function were treating the return as a boolean value, which happened to
2428 work because 0=false and not-0=true, but is not correct code.
2430 68. The host_aton() function was not handling scoped IPv6 addresses (those
2431 with, for example, "%eth0" on the end) correctly.
2433 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2434 negated items (that is, ~something) in unsigned ints. Some compilers
2435 apparently mutter when there is no cast.
2437 70. If an address verification called from an ACL failed, and did not produce a
2438 user-specific message (i.e. there was only a "system" message), nothing was
2439 put in $acl_verify_message. In this situation, it now puts the system
2442 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2443 session; change 4.31/43 added the unwanted input to the log line - except
2444 that it did not do this in the start of session case. It now does.
2446 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2447 This is wrong and can cause the other end to generate a synchronization
2448 error if it is another Exim or anything else that does the synchronization
2449 check. A QUIT command is no longer sent after a timeout.
2451 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2452 during host lookups.
2454 74. The defer_ok option of callout verification was not working if it was used
2455 when verifying addresses in header lines, that is, for this case:
2457 verify = header_sender/callout=defer_ok
2459 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2460 those file descriptors could be used for SMTP connections. If anything
2461 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2462 could be sent to the SMTP client, causing chaos. The daemon now opens
2463 stdin, stdout, and stderr to /dev/null when it puts itself into the
2466 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2467 log by default. The user can override this with suitable Perl magic.
2469 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2470 the log message when discard triggers, was not working for the DATA ACL or
2471 for the non-SMTP ACL.
2473 78. Error message wording change in sieve.c.
2475 79. If smtp_accept_max_per_host was set, the number of connections could be
2476 restricted to fewer than expected, because the daemon was trying to set up
2477 a new connection before checking whether the processes handling previous
2478 connections had finished. The check for completed processes is now done
2479 earlier. On busy systems, this bug wouldn't be noticed because something
2480 else would have woken the daemon, and it would have reaped the completed
2483 80. If a message was submitted locally by a user whose login name contained one
2484 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2485 It caused a spool format error. I have fixed the spool reading code. A
2486 related problem was that the "from" clause in the Received: line became
2487 illegal because of the space(s). It is now covered by ${quote_local_part.
2489 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2492 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2494 83. Patch from Sieve maintainer to fix unterminated string problem in
2495 "vacation" handling.
2497 84. Some minor changes to the Linux configuration files to help with other
2498 OS variants using glibc.
2500 85. One more patch for Sieve to update vacation handling to latest spec.
2503 ----------------------------------------------------
2504 See the note above about the 4.44 and 4.50 releases.
2505 ----------------------------------------------------
2511 1. Change 4.43/35 introduced a bug that caused file counts to be
2512 incorrectly computed when quota_filecount was set in an appendfile
2515 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2516 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2519 3. Give more explanation in the error message when the command for a transport
2520 filter fails to execute.
2522 4. There are several places where Exim runs a non-Exim command in a
2523 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2524 was being done only for the command run by the queryprogram router. It is
2525 now done for all such subprocesses. The other cases are: ${run, transport
2526 filters, and the commands run by the lmtp and pipe transports.
2528 5. Some older OS have a limit of 256 on the maximum number of file
2529 descriptors. Exim was using setrlimit() to set 1000 as a large value
2530 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2531 systems. I've change it so that if it can't get 1000, it tries for 256.
2533 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2534 was an oversight, and furthermore, ever since the addition of extra
2535 controls (e.g. 4.43/32), the checks on when to allow different forms of
2536 "control" were broken. There should now be diagnostics for all cases when a
2537 control that does not make sense is encountered.
2539 7. $recipients is now available in the predata ACL (oversight).
2541 8. Tidy the search cache before the fork to do a delivery from a message
2542 received from the command line. Otherwise the child will trigger a lookup
2543 failure and thereby defer the delivery if it tries to use (for example) a
2544 cached ldap connection that the parent has called unbind on.
2546 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2547 of $address_data from the recipient verification was clobbered by the
2548 sender verification.
2550 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2551 was its contents. (It was OK if the option was not defined at all.)
2553 11. A "Completed" log line is now written for messages that are removed from
2554 the spool by the -Mrm option.
2556 12. $host_address is now set to the target address during the checking of
2557 ignore_target_hosts.
2559 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2560 being passed; this would have caused $sender_host_name to have been used if
2561 matching the list had actually called for a host name (not very likely,
2562 since this list is usually IP addresses). A host name is now passed as
2565 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2566 code that specifies a non-restarting handler (typically sigaction() in
2567 modern systems) in an attempt to fix a rare and obscure crash bug.
2569 15. Narrowed the window for a race in the daemon that could cause it to ignore
2570 SIGCHLD signals. This is not a major problem, because they are used only to
2571 wake it up if nothing else does.
2573 16. A malformed maildirsize file could cause Exim to calculate negative values
2574 for the mailbox size or file count. Odd effects could occur as a result.
2575 The maildirsize information is now recalculated if the size or filecount
2578 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2579 support for a long time. Removed HAVE_SYS_VFS_H.
2581 18. Updated exipick to current release from John Jetmore.
2583 19. Allow an empty sender to be matched against a lookup in an address list.
2584 Previously the only cases considered were a regular expression, or an
2587 20. Exim went into a mad DNS lookup loop when doing a callout where the
2588 host was specified on the transport, if the DNS lookup yielded more than
2591 21. The RFC2047 encoding function was originally intended for short strings
2592 such as real names; it was not keeping to the 75-character limit for
2593 encoded words that the RFC imposes. It now respects the limit, and
2594 generates multiple encoded words if necessary. To be on the safe side, I
2595 have increased the buffer size for the ${rfc2047: expansion operator from
2598 22. Failure to deliver a bounce message always caused it to be frozen, even if
2599 there was an errors_to setting on the router. The errors_to setting is now
2602 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2603 canonical form (fully expanded) before being placed in
2604 $sender_host_address.
2606 24. Updated eximstats to version 1.33
2608 25. Include certificate and key file names in error message when GnuTLS fails
2609 to set them up, because the GnuTLS error message doesn't include the name
2610 of the failing file when there is a problem reading it.
2612 26. Expand error message when OpenSSL has problems setting up cert/key files.
2615 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2616 (this can affect the format of dates).
2618 28. exim_tidydb, when checking for the continued existence of a message for
2619 which it has found a message-specific retry record, was not finding
2620 messages that were in split spool directories. Consequently, it was
2621 deleting retry records that should have stayed in existence.
2623 29. eximstats updated to version 1.35
2624 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2625 1.35 - bugfix such that pie charts by volume are generated correctly
2627 30. The SPA authentication driver was not abandoning authentication and moving
2628 on to the next authenticator when an expansion was forced to fail,
2629 contradicting the general specification for all authenticators. Instead it
2630 was generating a temporary error. It now behaves as specified.
2632 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2633 (the order specifies the preference for clients). The order is now AES256,
2634 AES128, 3DES, ARCFOUR128.
2636 31. Small patch to Sieve code - explicitly set From: when generating an
2639 32. Exim crashed if a remote delivery caused a very long error message to be
2640 recorded - for instance if somebody sent an entire SpamAssassin report back
2641 as a large number of 550 error lines. This bug was coincidentally fixed by
2642 increasing the size of one of Exim's internal buffers (big_buffer) that
2643 happened as part of the Exiscan merge. However, to be on the safe side, I
2644 have made the code more robust (and fixed the comments that describe what
2647 33. Some experimental protocols are using DNS PTR records for new purposes. The
2648 keys for these records are domain names, not reversed IP addresses. The
2649 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2650 leaves it alone. Component reversal etc. now happens only for IP addresses.
2653 34. The host_aton() function is supposed to be passed a string that is known
2654 to be a valid IP address. However, in the case of IPv6 addresses, it was
2655 not checking this. This is a hostage to fortune. Exim now panics and dies
2656 if the condition is not met. A case was found where this could be provoked
2657 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2658 components; fortuitously, this particular loophole had already been fixed
2659 by change 4.50/55 or 4.44/33 above.
2661 If there are any other similar loopholes, the new check in host_aton()
2662 itself should stop them being exploited. The report I received stated that
2663 data on the command line could provoke the exploit when Exim was running as
2664 exim, but did not say which command line option was involved. All I could
2665 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2666 running as the user.
2669 35. There was a buffer overflow vulnerability in the SPA authentication code
2670 (which came originally from the Samba project). I have added a test to the
2671 spa_base64_to_bits() function which I hope fixes it.
2674 36. The daemon start-up calls getloadavg() while still root for those OS that
2675 need the first call to be done as root, but it missed one case: when
2676 deliver_queue_load_max is set with deliver_drop_privilege. This is
2677 necessary for the benefit of the queue runner, because there is no re-exec
2678 when deliver_drop_privilege is set.
2680 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2681 was in use, was not putting the data itself into the right store pool;
2682 consequently, it could be overwritten for a subsequent message in the same
2683 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2686 38. Sometimes the final signoff response after QUIT could fail to get
2687 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2688 < 0 before doing a fflush(). This bug looks as though it goes back to the
2689 introduction of TLS in release 3.20, but "sometimes" must have been rare
2690 because the tests only now provoked it.
2696 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2697 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2698 Now it returns an integer. A number of calls were still expecting a T/F
2699 return. Fortuitously, in all cases, the tests worked in OK situations,
2700 which is the norm. However, things would have gone wrong on any write
2701 failures on the smtp file descriptor. This function is used when sending
2702 messages over SMTP and also when doing verify callouts.
2704 2. When Exim is called to do synchronous delivery of a locally submitted
2705 message (the -odf or -odi options), it no longer closes stderr before doing
2708 3. Implemented the mua_wrapper option.
2710 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2712 5. Implemented the functions header_remove(), header_testname(),
2713 header_add_at_position(), and receive_remove_recipient(), and exported them
2716 6. If an ACL "warn" statement specified the addition of headers, Exim already
2717 inserted X-ACL-Warn: at the start if there was no header name. However, it
2718 was not making this test for the second and subsequent header lines if
2719 there were newlines in the string. This meant that an invalid header could
2720 be inserted if Exim was badly configured.
2722 7. Allow an ACL "warn" statement to add header lines at the start or after all
2723 the Received: headers, as well as at the end.
2725 8. Added the rcpt_4xx retry error code.
2727 9. Added postmaster_mailfrom=xxx to callout verification option.
2729 10. Added mailfrom=xxxx to the callout verification option, for verify=
2732 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2733 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2735 12. Inserted some casts to stop certain compilers warning when using pointer
2736 differences as field lengths or precisions in printf-type calls (mostly
2737 affecting debugging statements).
2739 13. Added optional readline() support for -be (dynamically loaded).
2741 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2742 same clock tick as a message's arrival, so that its received time was the
2743 same as the "first fail" time on the retry record, and that message
2744 remained on the queue past the ultimate address timeout, every queue runner
2745 would try a delivery (because it was past the ultimate address timeout) but
2746 after another failure, the ultimate address timeout, which should have then
2747 bounced the address, did not kick in. This was a "< instead of <=" error;
2748 in most cases the first failure would have been in the next clock tick
2749 after the received time, and all would be well.
2751 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2752 being recognized when the domain list was tested by the match_domain
2753 condition in an expansion string.
2755 16. Added the ${str2b64: operator.
2757 17. Exim was always calling setrlimit() to set a large limit for the number of
2758 processes, without checking whether the existing limit was already
2759 adequate. (It did check for the limit on file descriptors.) Furthermore,
2760 errors from getrlimit() and setrlimit() were being ignored. Now they are
2761 logged to the main and panic logs, but Exim does carry on, to try to do its
2762 job under whatever limits there are.
2764 18. Imported PCRE 5.0.
2766 19. Trivial typo in log message " temporarily refused connection" (the leading
2769 20. If the log selector return_path_on_delivery was set and an address was
2770 redirected to /dev/null, the delivery process crashed because it assumed
2771 that a return path would always be set for a "successful" delivery. In this
2772 case, the whole delivery is bypassed as an optimization, and therefore no
2775 21. Internal re-arrangement: the function for sending a challenge and reading
2776 a response while authentication was assuming a zero-terminated challenge
2777 string. It's now changed to take a pointer and a length, to allow for
2778 binary data in such strings.
2780 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2782 23. Exim was not respecting finduser_retries when seeking the login of the
2783 uid under which it was called; it was always trying 10 times. (The default
2784 setting of finduser_retries is zero.) Also, it was sleeping after the final
2785 failure, which is pointless.
2787 24. Implemented tls_on_connect_ports.
2789 25. Implemented acl_smtp_predata.
2791 26. If the domain in control=submission is set empty, Exim assumes that the
2792 authenticated id is a complete email address when it generates From: or
2793 Sender: header lines.
2795 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2796 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2797 chown and chgrp in /bin and hostname in /usr/bin.
2799 28. Exim was keeping the "process log" file open after each use, just as it
2800 does for the main log. This opens the possibility of it remaining open for
2801 long periods when the USR1 signal hits a daemon. Occasional processlog
2802 errors were reported, that could have been caused by this. Anyway, it seems
2803 much more sensible not to leave this file open at all, so that is what now
2806 29. The long-running daemon process does not normally write to the log once it
2807 has entered its main loop, and it closes the log before doing so. This is
2808 so that log files can straightforwardly be renamed and moved. However,
2809 there are a couple of unusual error situations where the daemon does write
2810 log entries, and I had neglected to close the log afterwards.
2812 30. The text of an SMTP error response that was received during a remote
2813 delivery was being truncated at 512 bytes. This is too short for some of
2814 the long messages that one sometimes sees. I've increased the limit to
2817 31. It is now possible to make retry rules that apply only when a message has a
2818 specific sender, in particular, an empty sender.
2820 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2821 it possible to be selective about when SMTP synchronization is enforced.
2823 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2825 32. Implemented hosts_connection_nolog.
2827 33. Added an ACL for QUIT.
2829 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2832 35. Added mailbox_size and mailbox_filecount to appendfile.
2834 36. Added control = no_multiline_responses to ACLs.
2836 37. There was a bug in the logic of the code that waits for the clock to tick
2837 in the case where the clock went backwards by a substantial amount such
2838 that the microsecond fraction of "now" was more than the microsecond
2839 fraction of "then" (but the whole seconds number was less).
2841 38. Added support for the libradius Radius client library this is found on
2842 FreeBSD (previously only the radiusclient library was supported).
2848 1. When certain lookups returned multiple values in the form name=value, the
2849 quoting of the values was not always being done properly. Specifically:
2850 (a) If the value started with a double quote, but contained no whitespace,
2852 (b) If the value contained whitespace other than a space character (i.e.
2853 tabs or newlines or carriage returns) it was not quoted.
2854 This fix has been applied to the mysql and pgsql lookups by writing a
2855 separate quoting function and calling it from the lookup code. The fix
2856 should probably also be applied to nisplus, ibase and oracle lookups, but
2857 since I cannot test any of those, I have not disturbed their existing code.
2859 2. A hit in the callout cache for a specific address caused a log line with no
2860 reason for rejecting RCPT. Now it says "Previous (cached) callout
2861 verification failure".
2863 3. There was an off-by-one bug in the queryprogram router. An over-long
2864 return line was truncated at 256 instead of 255 characters, thereby
2865 overflowing its buffer with the terminating zero. As well as fixing this, I
2866 have increased the buffer size to 1024 (and made a note to document this).
2868 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2869 when Exim is waiting for an SMTP response from a remote server, Exim
2870 restarts its select() call on the socket, thereby resetting its timeout.
2871 This is not a problem when such interrupts are rare. Somebody set up a cron
2872 job to run exiwhat every 2 minutes, which is less than the normal select()
2873 timeout (5 or 10 minutes). This meant that the select() timeout never
2874 kicked in because it was always reset. I have fixed this by comparing the
2875 time when an interrupt arrives with the time at the start of the first call
2876 to select(). If more time than the timeout has elapsed, the interrupt is
2877 treated as a timeout.
2879 5. Some internal re-factoring in preparation for the addition of Sieve
2880 extensions (by MH). In particular, the "personal" test is moved to a
2881 separate function, and given an option for scanning Cc: and Bcc: (which is
2882 not set for Exim filters).
2884 6. When Exim created an email address using the login of the caller as the
2885 local part (e.g. when creating a From: or Sender: header line), it was not
2886 quoting the local part when it contained special characters such as @.
2888 7. Installed new OpenBSD configuration files.
2890 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2891 try to make them clearer.
2893 9. Callout options, other than the timeout value, were being ignored when
2894 verifying sender addresses in header lines. For example, when using
2896 verify = header_sender/callout=no_cache
2898 the cache was (incorrectly) being used.
2900 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2901 only the Cygwin environment.
2903 11. When return_path_on_delivery was set as a log selector, if different remote
2904 addresses in the same message used different return paths and parallel
2905 remote delivery occurred, the wrong values would sometimes be logged.
2906 (Whenever a remote delivery process finished, the return path value from
2907 the most recently started remote delivery process was logged.)
2909 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2910 lines when AUTH and/or TLS are in use. This is the "received protocol"
2911 field. Exim used to use "asmtp" for authenticated SMTP, without any
2912 indication (in the protocol name) for TLS use. Now it follows the RFC and
2913 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2914 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2915 names appear in log lines as well as in Received: header lines.
2917 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2918 extensions, and comparison tests, and to fix some bugs.
2920 14. Changes to the "personal" filter test:
2922 (1) The test was buggy in that it was just doing the equivalent of
2923 "contains" tests on header lines. For example, if a user's address was
2924 anne@some.where, the "personal" test would incorrectly be true for
2926 To: susanne@some.where
2928 This test is now done by extracting each address from the header in turn,
2929 and checking the entire address. Other tests that are part of "personal"
2930 are now done using regular expressions (for example, to check local parts
2931 of addresses in From: header lines).
2933 (2) The list of non-personal local parts in From: addresses has been
2934 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2935 taken from the Sieve specification recommendations.
2937 (3) If the message contains any header line starting with "List-" it is
2938 treated as non-personal.
2940 (4) The test for "circular" in the Subject: header line has been removed
2941 because it now seems ill-conceived.
2943 15. Minor typos in src/EDITME comments corrected.
2945 16. Installed latest exipick from John Jetmore.
2947 17. If headers_add on a router specified a text string that was too long for
2948 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2949 of string_sprintf() is now avoided.
2951 18. $message_body_size was not set (it was always zero) when running the DATA
2952 ACL and the local_scan() function.
2954 19. For the "mail" command in an Exim filter, no default was being set for
2955 the once_repeat time, causing a random time value to be used if "once" was
2956 specified. (If the value happened to be <= 0, no repeat happened.) The
2957 default is now 0s, meaning "never repeat". The "vacation" command was OK
2958 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2959 (I found it when inspecting the code).
2961 20. There is now an overall timeout for performing a callout verification. It
2962 defaults to 4 times the callout timeout, which applies to individual SMTP
2963 commands during the callout. The overall timeout applies when there is more
2964 than one host that can be tried. The timeout is checked before trying the
2965 next host. This prevents very long delays if there are a large number of
2966 hosts and all are timing out (e.g. when the network connections are timing
2967 out). The value of the overall timeout can be changed by specifying an
2968 additional sub-option for "callout", called "maxwait". For example:
2970 verify = sender/callout=5s,maxwait=20s
2972 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2973 to the end before writing, but this should make it even safer).
2975 22. Exim was forgetting that it had advertised PIPELINING for the second and
2976 subsequent messages on an SMTP connection. It was also not resetting its
2977 memory on STARTTLS and an internal HELO.
2979 23. When Exim logs an SMTP synchronization error within a session, it now
2980 records whether PIPELINING has been advertised or not.
2982 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2983 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2984 rather than long int.
2986 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2988 26. Added the never_mail option to autoreply.
2994 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2995 crash if the getsockname() call failed; this can happen if a connection is
2996 closed very soon after it is established. The problem was simply in the
2997 order in which certain operations were done, causing Exim to try to write
2998 to the SMTP stream before it had set up the file descriptor. The bug has
2999 been fixed by making things happen in the correct order.
3005 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3006 before the connection was closed, thus losing the rejection response.
3008 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3009 some early Solaris releases, but causes trouble in current releases where
3010 socklen_t is defined.
3012 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3015 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3018 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3019 that releasing the top of it at the end releases what was used for sub-
3020 expansions (unless the block got too big). However, discard this block if
3021 the first thing is a variable or header, so that we can use its block when
3022 it is dynamic (useful for very large $message_headers, for example).
3024 6. Lookups now cache *every* query, not just the most recent. A new, separate
3025 store pool is used for this. It can be recovered when all lookup caches are
3026 flushed. Lookups now release memory at the end of their result strings.
3027 This has involved some general refactoring of the lookup sources.
3029 7. Some code has been added to the store_xxx() functions to reduce the amount
3030 of flapping under certain conditions.
3032 8. log_incoming_interface used to affect only the <= reception log lines. Now
3033 it causes the local interface and port to be added to several more SMTP log
3034 lines, for example "SMTP connection from", and rejection lines.
3036 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3038 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3040 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3041 could be overwritten at the end of the current message (or the start of a
3042 new message if it was set in a HELO ACL). The value is now preserved for
3043 the duration of the SMTP connection.
3045 12. If a transport had a headers_rewrite setting, and a matching header line
3046 contained an unqualified address, that address was qualified, even if it
3047 did not match any rewriting rules. The underlying bug was that the values
3048 of the flags that permit the existence of unqualified sender and recipient
3049 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3050 non-local messages, and by -bnq for local messages) were not being
3051 preserved with the message after it was received.
3053 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3054 "next input=" as part of the text comprising the host identity instead of
3055 the correct text. The code was using the same buffer for two different
3056 strings. However, depending on which order the printing function evaluated
3057 its arguments, the bug did not always show up. Under Linux, for example, my
3058 test suite worked just fine.
3060 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3061 doesn't work with some older versions of Perl. It has been changed to "my",
3062 which in any case is probably the better facility to use.
3064 15. A really picky compiler found some instances of statements for creating
3065 error messages that either had too many or two few arguments for the format
3068 16. The size of the buffer for calls to the DNS resolver has been increased
3069 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3070 for addresses that have a lot of PTR records. This alleviates a problem; it
3071 does not fully solve it.
3073 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3074 buffer now truncates the list and logs the incident, which is the same
3075 action as happens when Exim is looking up a host name and its aliases.
3076 Previously in this situation something unpredictable would happen;
3077 sometimes it was "internal error: store_reset failed".
3079 18. If a server dropped the connection unexpectedly when an Exim client was
3080 using GnuTLS and trying to read a response, the client delivery process
3081 crashed while trying to generate an error log message.
3083 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3084 string, for example:
3086 warn message = H1: something\nH2: something
3088 the text was added as a single header line from Exim's point of view
3089 though it ended up OK in the delivered message. However, searching for the
3090 second and subsequent header lines using $h_h2: did not work. This has been
3091 fixed. Similarly, if a system filter added multiple headers in this way,
3092 the routers could not see them.
3094 20. Expanded the error message when iplsearch is called with an invalid key to
3095 suggest using net-iplsearch in a host list.
3097 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3098 ACLs are no longer actually imposed (and a message to that effect is
3101 22. If a "gecos" field in a passwd entry contained escaped characters, in
3102 particular, if it contained a \" sequence, Exim got it wrong when building
3103 a From: or a Sender: header from that name. A second bug also caused
3104 incorrect handling when an unquoted " was present following a character
3105 that needed quoting.
3107 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3108 was not being matched caselessly.
3110 24. Arranged for all hyphens in the exim.8 source to be escaped with
3113 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3114 properly. Recipient callout cache records were still being keyed to include
3115 the sender, even when use_sender was set false. This led to far more
3116 callouts that were necessary. The sender is no longer included in the key
3117 when use_sender is false.
3119 26. Added "control = submission" modifier to ACLs.
3121 27. Added the ${base62d: operator to decode base 62 numbers.
3123 28. dnsdb lookups can now access SRV records.
3125 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3126 the configuration file.
3128 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3129 (-v) mode. This makes the output for a verbose queue run more intelligible.
3131 31. Added a use_postmaster feature to recipient callouts.
3133 32. Added the $body_zerocount variable, containing the number of binary zero
3134 bytes in the message body.
3136 33. The time of last modification of the "new" subdirectory is now used as the
3137 "mailbox time last read" when there is a quota error for a maildir
3140 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3142 35. Added +ignore_unknown as a special item in host lists.
3144 36. Code for decoding IPv6 addresses in host lists is now included, even if
3145 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3146 address was recognized as an IP address, but was then not correctly decoded
3147 into binary, causing unexpected and incorrect effects when compared with
3154 1. Very minor rewording of debugging text in manualroute to say "list of
3155 hosts" instead of "hostlist".
3157 2. If verify=header_syntax was set, and a header line with an unqualified
3158 address (no domain) and a large number of spaces between the end of the
3159 name and the colon was received, the reception process suffered a buffer
3160 overflow, and (when I tested it) crashed. This was caused by some obsolete
3161 code that should have been removed. The fix is to remove it!
3163 3. When running in the test harness, delay a bit after writing a bounce
3164 message to get a bit more predictability in the log output.
3166 4. Added a call to search_tidyup() just before forking a reception process. In
3167 theory, someone could use a lookup in the expansion of smtp_accept_max_
3168 per_host which, without the tidyup, could leave open a database connection.
3170 5. Added the variables $recipient_data and $sender_data which get set from a
3171 lookup success in an ACL "recipients" or "senders" condition, or a router
3172 "senders" option, similar to $domain_data and $local_part_data.
3174 6. Moved the writing of debug_print from before to after the "senders" test
3177 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3178 problems for message scanning, either using a data ACL, or using
3179 local_scan() because the Received: header was not generated till after they
3180 were called (in order to set the time as the time of reception completion).
3181 I have revised the way this works. The header is now generated after the
3182 body is received, but before the ACL or local_scan() are called. After they
3183 are run, the timestamp in the header is updated.
3189 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3190 before starting a queue runner without re-exec. This happened only when
3191 deliver_drop_privilege was set or when the Exim user was set to root. The
3192 effect of the bug was that timeouts during subsequent deliveries caused
3193 crashes instead of being properly handled. The handler is now left at its
3194 default (and expected) setting.
3196 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3197 message, again when deliver_drop_privilege is set or Exim is run as root.
3198 The bug described in (1) was not present in this case, but the tidying up
3199 of the other signals was missing. I have made the two cases consistent.
3201 3. The ignore_target_hosts setting on a manualroute router was being ignored
3202 for hosts that were looked up using the /MX notation.
3204 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3207 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3208 operated on the sender address. After changing the $sender_address to <>
3209 for the sender address verify, Exim was re-instated it as the original
3210 (before rewriting) address, but remembering that it had rewritten it, so it
3211 wasn't rewriting it again. This bug also had the effect of breaking the
3212 sender address verification caching when the sender address was rewritten.
3214 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3215 This has been changed so that if the ip literal address matches
3216 ignore_target_hosts, the router declines.
3218 7. Added expansion conditions match_domain, match_address, and match_local_
3219 part (NOT match_host).
3221 8. The placeholder for the Received: header didn't have a length field set.
3223 9. Added code to Exim itself and to exim_lock to test for a specific race
3224 condition that could lead to file corruption when using MBX delivery. The
3225 issue is with the lockfile that is created in /tmp. If this file is removed
3226 after a process has opened it but before that process has acquired a lock,
3227 there is the potential for a second process to recreate the file and also
3228 acquire a lock. This could lead to two Exim processes writing to the file
3229 at the same time. The added code performs the same test as UW imapd; it
3230 checks after acquiring the lock that its file descriptor still refers to
3231 the same named file.
3233 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3234 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3237 11. Added the smtp_active_hostname option. If used, this will typically be made
3238 to depend on the incoming interface address. Because $interface_address is
3239 not set up until the daemon has forked a reception process, error responses
3240 that can happen earlier (such as "too many connections") no longer contain
3243 12. If an expansion in a condition on a "warn" statement fails because a lookup
3244 defers, the "warn" statement is abandoned, and the next ACL statement is
3245 processed. Previously this caused the whole ACL to be aborted.
3247 13. Added the iplsearch lookup type.
3249 14. Added ident_timeout as a log selector.
3251 15. Added tls_certificate_verified as a log selector.
3253 16. Added a global option tls_require_ciphers (compare the smtp transport
3254 option of the same name). This controls incoming TLS connections.
3256 17. I finally figured out how to make tls_require_ciphers do a similar thing
3257 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3258 before starting the TLS session.
3260 18. Tabs are now shown as \t in -bP output.
3262 19. If the log selector return_path_on_delivery was set, Exim crashed when
3263 bouncing a message because it had too many Received: header lines.
3265 20. If two routers both had headers_remove settings, and the first one included
3266 a superfluous trailing colon, the final name in the first list and the
3267 first name in the second list were incorrectly joined into one item (with a
3268 colon in the middle).
3274 1. Added -C and -D options to the exinext utility, mainly to make it easier
3275 to include in the automated testing, but these could be helpful when
3276 multiple configurations are in use.
3278 2. The exinext utility was not formatting the output nicely when there was
3279 an alternate port involved in the retry record key, nor when there was a
3280 message id as well (for retries that were specific to a specific message
3281 and a specific host). It was also confused by IPv6 addresses, because of
3282 the additional colons they contain. I have fixed the IPv4 problem, and
3283 patched it up to do a reasonable job for IPv6.
3285 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3286 delivery, the log line now contains "pipelined" if PIPELINING was used.
3288 4. An SMTP transport process used to panic and die if the bind() call to set
3289 an explicit outgoing interface failed. This has been changed; it is now
3290 treated in the same way as a connect() failure.
3292 5. A reference to $sender_host_name in the part of a conditional expansion
3293 that was being skipped was still causing a DNS lookup. This no longer
3296 6. The def: expansion condition was not recognizing references to header lines
3297 that used bh_ and bheader_.
3299 7. Added the _cache feature to named lists.
3301 8. The code for checking quota_filecount in the appendfile transport was
3302 allowing one more file than it should have been.
3304 9. For compatibility with Sendmail, the command line option
3312 and sets the incoming protocol and host name (for trusted callers). The
3313 host name and its colon can be omitted when only the protocol is to be set.
3314 Note the Exim already has two private options, -pd and -ps, that refer to
3315 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3316 "s", but I don't think that's a major issue.
3318 10. A number of refactoring changes to the code, none of which should affect
3321 (a) The number of logging options was getting close to filling up the
3322 32-bit word that was used as a bit map. I have split them into two classes:
3323 those that are passed in the argument to log_write(), and those that are
3324 only ever tested independently outside of that function. These are now in
3325 separate 32-bit words, so there is plenty of room for expansion again.
3326 There is no change in the user interface or the logging behaviour.
3328 (b) When building, for example, log lines, the code previously used a
3329 macro that called string_cat() twice, in order to add two strings. This is
3330 not really sufficiently general. Furthermore, there was one instance where
3331 it was actually wrong because one of the argument was used twice, and in
3332 one call a function was used. (As it happened, calling the function twice
3333 did not affect the overall behaviour.) The macro has been replaced by a
3334 function that can join an arbitrary number of extra strings onto a growing
3337 (c) The code for expansion conditions now uses a table and a binary chop
3338 instead of a serial search (which was left over from when there were very
3339 few conditions). Also, it now recognizes conditions like "pam" even when
3340 the relevant support is not compiled in: a suitably worded error message is
3341 given if an attempt is made to use such a condition.
3343 11. Added ${time_interval:xxxxx}.
3345 12. A bug was causing one of the ddress fields not to be passed back correctly
3346 from remote delivery subprocesses. The field in question was not being
3347 subsequently used, so this caused to problems in practice.
3349 13. Added new log selectors queue_time and deliver_time.
3351 14. Might have fixed a bug in maildirsizefile handling that threw up
3352 "unexpected character" debug warnings, and recalculated the data
3353 unnecessarily. In any case, I expanded the warning message to give more
3356 15. Added the message "Restricted characters in address" to the statements in
3357 the default ACL that block characters like @ and % in local parts.
3359 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3360 Three changes have been made:
3362 (a) There was a serious bug; a negative response to MAIL caused the whole
3363 recipient domain to be cached as invalid, thereby blocking all messages
3364 to all local parts at the same domain, from all senders. This bug has
3365 been fixed. The domain is no longer cached after a negative response to
3366 MAIL if the sender used is not empty.
3368 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3371 (c) A new callout option, "use_sender" has been added for people who want
3372 the modified behaviour.
3378 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3381 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3382 indeed breaks things for older releases.
3384 3. Added additional logging to the case where there is a problem reading data
3385 from a filter that is running in a subprocess using a pipe, in order to
3386 try to track down a specific problem.
3388 4. Testing facility fudge: when running in the test harness and attempting
3389 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3390 getting "No route to host". Convert this to a timeout.
3392 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3395 6. Some OS don't have socklen_t but use size_t instead. This affects the
3396 fifth argument of getsockopt() amongst other things. This is now
3397 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3398 can be set for individual OS. I have set it for SunOS5, OSF1, and
3399 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3400 some earlier ones do not.
3402 7. Change 4.30/15 was not doing the test caselessly.
3404 8. The standard form for an IPv6 address literal was being rejected by address
3405 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3406 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3407 this, as well as the form without the "IPv6" on the front (but only when
3408 address literals are enabled, of course).
3410 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3412 10. Exim crashed if a message with an empty sender address specified by -f
3413 encountered a router with an errors_to setting. This could be provoked only
3414 by a command such as
3418 where an empty string was supplied; "<>" did not hit this bug.
3420 11. Installed PCRE release 4.5.
3422 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3423 remained set. It is now erased.
3425 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3426 times from message ids (which are base 36 rather than the normal 62).
3428 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3429 were being counted as actual protocol errors, and logged if the log
3430 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3431 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3432 and DATA following a set of rejected RCPTs do not count as protocol errors.
3433 In other words, Exim assumes they were pipelined, though this may not
3434 actually be the case. Of course, in all cases the client gets an
3435 appropriate error code.
3437 15. If a lookup fails in an ACL condition, a message about the failure may
3438 be available; it is used if testing the ACL cannot continue, because most
3439 such messages specify what the cause of the deferral is. However, some
3440 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3441 that caused an old message to be retained and used if a later statement
3442 caused a defer, replacing the real cause of the deferral.
3444 16. If an IP address had so many PTR records that the DNS lookup buffer
3445 was not large enough to hold them, Exim could crash while trying to process
3446 the truncated data. It now detects and logs this case.
3448 17. Further to 4.21/58, another change has been made: if (and only if) the
3449 first line of a message (the first header line) ends with CRLF, a bare LF
3450 in a subsequent header line has a space inserted after it, so as not to
3451 terminate the header.
3453 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3454 unnecessarily, used atoi() instead of strtol(), and didn't check the
3455 termination when getting file sizes from file names by regex.
3457 19. Completely re-implemented the support for maildirsize files, in the light
3458 of a number of problems with the previous contributed implementation
3459 (4.30/29). In particular:
3461 . If the quota is zero, the maildirsize file is maintained, but no quota is
3464 . If the maildir directory does not exist, it is created before any attempt
3465 to write a maildirsize file.
3467 . The quota value in the file is just a cache; if the quota is changed in
3468 the transport, the new value overrides.
3470 . A regular expression is available for excluding directories from the
3473 20. The autoreply transport checks the characters in options that define the
3474 message's headers; it allows continued headers, but it was checking with
3475 isspace() after an embedded newline instead of explicitly looking for a
3478 21. If all the "regular" hosts to which an address was routed had passed their
3479 expiry times, and had not reached their retry times, the address was
3480 bounced, even if fallback hosts were defined. Now Exim should go on to try
3483 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3484 equivalent code in the SMTP transport. Some hosts send humungous responses
3485 to HELO/EHLO, more than 1024 it seems.
3487 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3488 gives compiler warnings in some environments. I've now done it "properly",
3491 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3492 (because of problems with the built-in one) was declared to return uschar *
3493 instead of char *, causing compiler failure.
3495 25. Fixed a file descriptor leak when processing alias/forward files.
3497 26. Fixed a minor format string issue in dbfn.c.
3499 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3501 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3502 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3505 29. When a sender address is verified, it is cached, to save repeating the test
3506 when there is more than one recipient in a message. However, when the
3507 verification involves a callout, it is possible for different callout
3508 options to be set for different recipients. It is too complicated to keep
3509 track of this in the cache, so now Exim always runs a verification when a
3510 callout is required, relying on the callout cache for the optimization.
3511 The overhead is duplication of the address routing, but this should not be
3514 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3515 to be verified with callout=postmaster, and the main callout worked but the
3516 postmaster check failed, the verification correctly failed. However, if a
3517 subsequent RCPT command asked for sender verification *without* the
3518 postmaster check, incorrect caching caused this verification also to fail,
3521 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3522 it was not caching the DNS options (qualify_single, search_parents) that
3523 were used when the lookup failed. A subsequent lookup with different
3524 options therefore always gave the same answer, though there were cases
3525 where it should not have. (Example: a "domains = !$mx_any" option on a
3526 dnslookup router: the "domains" option is always processed without any
3527 widening, but the router might have qualify_single set.) Now Exim uses the
3528 cached value only when the same options are set.
3530 32. Added John Jetmore's "exipick" utility to the distribution.
3532 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3533 than a timeout (e.g. a certificate is required, and is not provided), an
3534 Exim server now closes the connection immediately. Previously it waited for
3535 the client to close - but if the client is SSL, it seems that they each
3536 wait for each other, leading to a delay before one of them times out.
3538 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3539 maintained 0.8.x compatibility because I don't think many are using it, and
3540 it is clearly obsolete.
3542 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3545 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3546 client certificate was expired. A simple patch fixes this, though I don't
3547 understand the full logic of why the verify callback is called multiple
3550 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3551 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3552 which causes problems with some clients (such as the Certicom SSL Plus
3553 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3554 disables the coutermeasure allowing Eudora to connect."
3556 38. Exim was not checking that a write() to a log file succeeded. This could
3557 lead to Bad Things if a log got too big, in particular if it hit a file
3558 size limit. Exim now panics and dies if it cannot write to a log file, just
3559 as it does if it cannot open a log file.
3561 39. Modified OS/Makefile-Linux so that it now contains
3563 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3565 The two -D definitions ensure that Exim is compiled with large file
3566 support, which makes it possible to handle log files that are bigger than
3569 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3570 instance) a domain was checked against a named list that involved a lookup,
3571 causing $domain_data to be set, then another domain was checked against the
3572 same list, then the first domain was re-checked, the value of $domain_data
3573 after the final check could be wrong. In particular, if the second check
3574 failed, it could be set empty. This bug probably also applied to
3577 41. The strip_trailing_dot option was not being applied to the address given
3578 with the -f command-line option.
3580 42. The code for reading a message's header from the spool was incrementing
3581 $received_count, but never initializing it. This meant that the value was
3582 incorrect (doubled) while delivering a message in the same process in which
3583 it was received. In the most common configuration of Exim, this never
3584 happens - a fresh exec is done - but it can happen when
3585 deliver_drop_privilege is set.
3587 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3588 it now includes up to 150 characters of the unexpected data in the log
3591 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3592 and building data strings. The size of both of these buffers was 10 000
3593 bytes - far larger than anybody would *ever* want, thought I. Needless to
3594 say, somebody hit the limit. I have increased the maximum line length to
3595 20 000 and the maximum data length of concatenated lines to 100 000. I have
3596 also fixed two bugs, because there was no checking on these buffers. Tsk,
3597 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3598 buffer is too small.
3600 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3601 lsearch lookups. Now it does.
3603 46. When parsing a route_list item in a manualroute router, a fixed-length
3604 buffer was used for the list of hosts. I made this 1024 bytes long,
3605 thinking that nobody would ever have a list of hosts that long. Wrong.
3606 Somebody had a whole pile of complicated expansion conditions, and the
3607 string was silently truncated, leading to an expansion error. It turns out
3608 that it is easier to change to an unlimited length (owing to other changes
3609 that have happened since this code was originally written) than to build
3610 structure for giving a limitation error. The length of the item that
3611 expands into the list of hosts is now unlimited.
3613 47. The lsearch lookup could not handle data where the length of text line was
3614 more than 4095 characters. Such lines were truncated, leading to shortened
3615 data being returned. It should now handle lines of any length.
3617 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3618 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3621 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3622 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3623 no longer happens in comment lines. A list of replacements is now placed
3624 at the head of all of the source files, except those whose only change is
3625 to replace PERL_COMMAND in the very first #! line.
3627 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3628 messages on the queue, with a bottom-up merge sort, using code contributed
3629 by Michael Haardt. This should make operations like -bp somewhat faster on
3630 large queues. It won't affect queue runners, except when queue_run_in_order
3633 51. Installed eximstats 1.31 in the distribution.
3635 52. Added support for SRV lookups to the dnslookup router.
3637 53. If an ACL referred to $message_body or $message_body_end, the value was not
3638 reset for any messages that followed in the same SMTP session.
3640 54. The store-handling optimization for building very long strings was not
3641 differentiating between the different store pools. I don't think this
3642 actually made any difference in practice, but I've tidied it.
3644 55. While running the routers to verify a sender address, $sender_address
3645 was still set to the sender address. This is wrong, because when routing to
3646 send a bounce to the sender, it would be empty. Therefore, I have changed
3647 it so that, while verifying a sender address, $sender_address is set to <>.
3648 (There is no change to what happens when verifying a recipient address.)
3650 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3651 target A or AAAA records (if not already returned) without resetting the
3652 qualify_single or search_parents options of the DNS resolver. These are
3653 inappropriate in this case because the targets of MX and SRV records must
3654 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3655 target that, when qualified, matched something in the local domain. These
3656 two options are now turned off when doing these lookups.
3658 57. It seems that at least some releases of Reiserfs (which does not have the
3659 concept of a fixed number of inodes) returns zero and not -1 for the
3660 number of available inodes. This interacted badly with check_spool_inodes,
3661 which assumed that -1 was the "no such thing" setting. What I have done is
3662 to check that the total number of inodes is greater than zero before doing
3663 the test of how many are available.
3665 58. When a "warn" ACL statement has a log_message modifier, the message is
3666 remembered, and not repeated. This is to avoid a lot of repetition when a
3667 message has many recipients that cause the same warning to be written.
3668 Howewer, Exim was preserving the list of already written lines for an
3669 entire SMTP session, which doesn't seem right. The memory is now reset if a
3670 new message is started.
3672 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3673 debugging output unless log_rewrite was also set.
3675 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3676 of (int)(handle) when we know that handle contains (void *)(-1).
3678 61. The Exim daemon panic-logs an error return when it closes the incoming
3679 connection. However "connection reset by peer" seems to be common, and
3680 isn't really an error worthy of noting specially, so that particular error
3683 62. When Exim is trying to find all the local interfaces, it used to panic and
3684 die if the ioctl to get the interface flags failed. However, it seems that
3685 on at least one OS (Solaris 9) it is possible to have an interface that is
3686 included in the list of interfaces, but for which you get a failure error
3687 for this call. This happens when the interface is not "plumbed" into a
3688 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3689 failure of the "get flags" call assumes that the interface is down.
3691 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3692 makes life easier for people who are doing arithmetic on fields extracted
3693 from dates, where you often get leading zeros that should not be
3694 interpreted as octal.
3696 64. Added qualify_domain to the redirect router, to override the global
3699 65. If a pathologically long header line contained very many addresses (the
3700 report of this problem mentioned 10 000) and each of them was rewritten,
3701 Exim could use up a very large amount of memory. (It kept on making new
3702 copies of the header line as it rewrote, and never released the old ones.)
3703 At the expense of a bit more processing, the header rewriting function has
3704 been changed so that it no longer eats memory in this way.
3706 66. The generation of the Received: header has been moved from the time that a
3707 message starts to be received, to the time that it finishes. The timestamp
3708 in the Received: header should now be very close to that of the <= log
3709 line. There are two side-effects of this change:
3711 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3712 logged header lines no longer include the local Received: line, because
3713 it has not yet been created. The same applies to a copy of the message
3714 that is returned to a non-SMTP sender when a message is rejected.
3716 (b) When a filter file is tested using -bf, no additional Received: header
3717 is added to the test message. After some thought, I decided that this
3720 This change does not affect the value of $received_for. It is still set
3721 after address rewriting, but before local_scan() is called.
3723 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3725 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3726 gave an unhelpful panic error message, and a defer error. I have managed to
3727 change this behaviour so that it now rejects any supplied certificate,
3728 which seems right, as the list of acceptable certificates is empty.
3730 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3731 gave an unhelpful defer error. I have not managed to make this reject any
3732 supplied certificates, but the error message it gives is "no certificate
3733 supplied", which is not helpful.
3735 70. exigrep's output now also includes lines that are not associated with any
3736 message, but which match the given pattern. Implemented by a patch from
3737 Martin Sluka, which also tidied up the Perl a bit.
3739 71. Recipient callout verification, like sender verification, was using <> in
3740 the MAIL FROM command. This isn't really the right thing, since the actual
3741 sender may affect whether the remote host accepts the recipient or not. I
3742 have changed it to use the actual sender in the callout; this means that
3743 the cache record is now keyed on a recipient/sender pair, not just the
3744 recipient address. There doesn't seem to be a real danger of callout loops,
3745 since a callout by the remote host to check the sender would use <>.
3746 [SEE ABOVE: changed after hitting problems.]
3748 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3749 temporary errors. However, in the case of such a code being given after
3750 the end of a data transmission (i.e. after ".") Exim was failing to write
3751 a retry record for the message. (Yes, there was some broken host that was
3752 actually sending 8xx at this point.)
3754 73. An unknown lookup type in a host list could cause Exim to panic-die when
3755 the list was checked. (An example that provoked this was putting <; in the
3756 middle of a list instead of at the start.) If this happened during a DATA
3757 ACL check, a -D file could be left lying around. This kind of configuration
3758 error no longer causes Exim to die; instead it causes a defer errror. The
3759 incident is still logged to the main and panic logs.
3761 74. Buglet left over from Exim 3 conversion. The message "too many messages
3762 in one connection" was written to the rejectlog but not the mainlog, except
3763 when address rewriting (yes!) was being logged.
3765 75. Added write_rejectlog option.
3767 76. When a system filter was run not as root (that is, when system_filter_user
3768 was set), the values of the $n variables were not being returned to the
3769 main process; thus, they were not subsequently available in the $sn
3772 77. Added +return_path_on_delivery log selector.
3774 78. A connection timeout was being treated differently from recipients deferred
3775 when testing hosts_max_try with a message that was older than the host's
3776 retry timeout. (The host should not be counted, thus allowing all hosts to
3777 be tried at least once before bouncing.) This may have been the cause of an
3778 occasionally reported bug whereby a message would remain on the queue
3779 longer than the retry timeout, but would be bounced if a delivery was
3780 forced. I say "may" because I never totally pinned down the problem;
3781 setting up timeout/retry tests is difficult. See also the next item.
3783 79. The ultimate address timeout was not being applied to errors that involved
3784 a combination of host plus message (for example, a timeout on a MAIL
3785 command). When an address resolved to a number of possible hosts, and they
3786 were not all tried for each delivery (e.g. because of hosts_max_try), a
3787 message could remain on the queue longer than the retry timeout.
3789 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3792 81. Fixed an obscure SMTP outgoing bug which required at least the following
3793 conditions: (a) there was another message waiting for the same server;
3794 (b) the server returned 5xx to all RCPT commands in the first message so
3795 that the message was not completed; (c) the server dropped the connection
3796 or gave a negative response to the RSET that Exim sends to abort the
3797 transaction. The observed case was a dropped connection after DATA that had
3798 been sent in pipelining mode. That is, the server had advertised PIPELINING
3799 but was not implementing it correctly. The effect of the bug was incorrect
3800 behaviour, such as trying another host, and this could lead to a crash.
3806 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3807 and daemon.c were passed as pointers to ints; they should have been
3808 pointers to socklen_t variables (which are typically unsigned ints).
3810 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3813 3. Fixed a really odd bug that affected only the testing scheme; patching a
3814 certain fixed string in the binary changed the value of another string that
3815 happened to be identical to the end of the original first string.
3817 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3818 name", it returns that address as the IP address. On some operating
3819 systems (e.g. Solaris), it also passes back the IP address string as the
3820 "host name". However, on others (e.g. Linux), it passes back an empty
3821 string. Exim wasn't checking for this, and was changing the host name to an
3822 empty string, assuming it had been canonicized.
3824 5. Although rare, it is permitted to have more than one PTR record for a given
3825 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3826 all the names associated with an address, because they do in Solaris.
3827 However, it seems that they do not in Linux for data that comes from the
3828 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3829 I found this out when I moved to a new Linux workstation and tried to run
3830 the Exim test suite.
3832 To get round this problem I have changed the code so that it now does its
3833 own call to the DNS to look up PTR records when searching for a host name.
3834 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3835 addresses that are only in /etc/hosts are still found.
3837 This behaviour is, however, controlled by an option called host_lookup_
3838 order, which defaults to "bydns:byaddr". If people want to use the other
3839 order, or indeed, just use one or the other means of lookup, they can
3840 specify it in this variable.
3842 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3843 some operating systems, this comes back from gethostbyaddr() as an empty
3844 string, and this is what Exim used to test for. However, it seems that in
3845 other systems, "." is yielded. Exim now tests for this case too.
3847 7. The values of check_spool_space and check_log_space are now held internally
3848 as a number of kilobytes instead of an absolute number of bytes. If a
3849 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3850 kilobyte. This means that much larger values can be stored.
3852 8. Exim monitor: an attempt to get the action menu when not actually pointing
3853 at a message produces an empty menu entitled "No message selected". This
3854 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3855 no entries in it ("Shell widget menu has zero width and/or height"). So I
3856 have added a single, blank menu entry in this case.
3858 9. Added ${quote_local_part.
3860 10. MIME decoding is now applied to the contents of Subject: header lines when
3863 11. Now that a reference to $sender_host_address automatically causes a reverse
3864 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3865 host lookup before query-style lookups in lists that might use this
3866 variable. This has therefore been abolished, and the "net-" prefix is no
3867 longer necessary for query-style lookups.
3869 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3870 have been a typo for LFLAGS, so it has been changed.
3872 13. The install script calls Exim with "-C /dev/null" in order to find the
3873 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3874 to be output. Howeve, since Exim outputs its version number before the
3875 error, it didn't break the script. It just looked ugly. I fixed this by
3876 always allowing "-C /dev/null" if the caller is root.
3878 14. Ignore overlarge ACL variable number when reading spool file - insurance
3879 against a later release with more variables having written the file.
3881 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3882 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3883 this, as well as the form without the "IPv6" on the front.
3885 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3886 OS/Makefile-Darwin file.
3888 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3889 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3891 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3892 variable. However, some OS use ints for inodes. I've added cast to long int
3893 to get rid of the compiler warning.
3895 19. I had forgotten to lock out "/../" in configuration file names when
3896 ALT_CONFIG_PREFIX was set.
3898 20. Routers used for verification do not need to specify transports. However,
3899 if such a router generated a host list, and callout was configured, Exim
3900 crashed, because it could not find a port number from the (non-existent)
3901 transport. It now assumes port 25 in this circumstance.
3903 21. Added the -t option to exigrep.
3905 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3906 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3907 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3908 them is set, LOOKUP_LSEARCH is forced.
3910 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3912 24. Added sender and host information to the "rejected by local_scan()" log
3913 line; previously there was no indication of these.
3915 25. Added .include_if_exists.
3917 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3918 Linux. It turns out that not all file systems support this. Apparently some
3919 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3920 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3921 not supported on the file descriptor, is now ignored when Exim is trying to
3922 sync a directory. This applies only to Linux.
3924 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3926 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3927 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3930 29. Added support for maildirsize files from supplied patch (modified a bit).
3932 30. The use of :fail: followed by an empty string could lead Exim to respond to
3933 sender verification failures with (e.g.):
3935 550 Verification failed for <xxx>
3936 550 Sender verify failed
3938 where the first response line was missing the '-' that indicates it is not
3939 the final line of the response.
3941 31. The loop for finding the name of the user that called Exim had a hardwired
3942 limit of 10; it now uses the value of finduser_retries, which is used for
3943 all other user lookups.
3945 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3948 33. Exim was neglecting to zero errno before one call of strtol() when
3949 expanding a string and expecting an integer value. On some systems this
3950 resulted in spurious "integer overflow" errors. Also, it was casting the
3951 result into an int without checking.
3953 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3954 did not work. The code looks as if it has *never* worked, though it appears
3955 to have been documented since at least releast 1.62. I have made it work.
3957 35. The "timeout_DNS" error in retry rules, also documented since at least
3958 1.62, also never worked. As it isn't clear exactly what this means, and
3959 clearly it isn't a major issue, I have abolished the feature by treating it
3960 as "timeout", and writing a warning to the main and panic logs.
3962 36. The display of retry rules for -brt wasn't always showing the error code
3965 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3966 timeout_connect_A, timeout_connect_MX.
3968 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3969 to the empty sender.
3971 39. The daemon was not analysing the content of -oX till after it had closed
3972 stderr and disconnected from the controlling terminal. This meant that any
3973 syntax errors were only noted on the panic log, and the return code from
3974 the command was 0. By re-arranging the code a little, I've made the
3975 decoding happen first, so such errors now appear on stderr, and the return
3976 code is 1. However, the actual setting up of the sockets still happens in
3977 the disconnected process, so errors there are still only recorded on the
3980 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3981 connections (as happens on some IP stacks) was logged at start up time as
3982 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3983 it from "IPv6 and IPv4", which means that two separate sockets are being
3986 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3987 says whether AF_INET or AF_INET6 was passed as an argument.
3989 42. Exiwhat output was messed up when time zones were included in log
3992 43. Exiwhat now gives more information about the daemon's listening ports,
3993 and whether -tls-on-connect was used.
3995 44. The "port" option of the smtp transport is now expanded.
3997 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3998 silently ignored. Now an error message is written to the main and panic
4001 46. There's a new ACL modifier called "logwrite" which writes to a log file
4002 as soon as it is encountered.
4004 47. Added $local_user_uid and $local_user_gid at routing time.
4006 48. Exim crashed when trying to verify a sender address that was being
4009 49. Exim was recognizing only a space character after ".include". It now also
4010 recognizes a tab character.
4012 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4013 extracting the relevant information from the specification. The man page no
4014 longer contains scrambled data for the -d option, and I've added a section
4015 at the front about calling Exim under different names.
4017 51. Added "extra_headers" argument to the "mail" command in filter files.
4019 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4022 53. Installed eximstats 1.29.
4024 54. Added transport_filter_timeout as a generic transport option.
4026 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4027 Cc: header lines. This was required by RFC 822, but it not required by RFC
4030 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4031 incoming messages that did not have them. Now it does so only if the
4032 message originates locally, that is, if there is no associated remote host
4033 address. When Resent- header lines are present, this applies to the Resent-
4034 lines rather than the non-Resent- lines.
4036 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4037 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4039 58. Messages for configuration errors now include the name of the main
4040 configuration file - useful now that there may be more than one file in a
4041 list (.included file names were always shown).
4043 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4044 for those rare installations that do not start the daemon as root or run it
4045 setuid root. I've cut out the call to initgroups() if the daemon is not
4048 60. The Exim user and group can now be bound into the binary as text strings
4049 that are looked up at the start of Exim's processing.
4051 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4053 62. Added $mailstore_basename variable.
4055 63. Installed patch to sieve.c from Michael Haardt.
4057 64. When Exim failed to open the panic log after failing to open the main log,
4058 the original message it was trying to log was written to stderr and debug
4059 output, but if they were not available (the usual case in production), it
4060 was lost. Now it is written to syslog before the two lines that record the
4061 failures to open the logs.
4063 65. Users' Exim filters run in subprocesses under the user's uid. It is
4064 possible for a "deliver" command or an alias in a "personal" command to
4065 provoke an address rewrite. If logging of address rewriting is configured,
4066 this fails because the process is not running as root or exim. There may be
4067 a better way of dealing with this, but for the moment (because 4.30 needs
4068 to be released), I have disabled address rewrite logging when running a
4069 filter in a non-root, non-exim process.
4075 1. The buildconfig auxiliary program wasn't quoting the value set for
4076 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4077 not defined. This bug was masked in 4.22 by the effect that was fixed in
4080 2. Some messages that were rejected after a message id was allocated were
4081 shown as "incomplete" by exigrep. It no longer does this for messages that
4082 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4084 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4085 have allow_domain_literals set, the ID did not get logged in the <= line.
4086 Domain literals are now always recognized in Message-ID: header lines.
4088 4. The first argument for a ${extract expansion item is the key name or field
4089 number. Leading and trailing spaces in this item were not being ignored,
4090 causing some misleading effects.
4092 5. When deliver_drop_privilege was set, single queue runner processes started
4093 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4094 same command in the process it spins off) were not dropping privilege.
4096 6. When the daemon running as "exim" started a queue runner, it always
4097 re-executed Exim in the spun-off process. This is a waste of effort when
4098 deliver_drop_privilege is set. The new process now just calls the
4099 queue-runner function directly.
4105 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4108 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4109 ignored. Though the use of -oP was forcing the writing of a pid file, it
4110 was always written to the default place.
4112 3. If the message "no IP address found for host xxxx" is generated during
4113 incoming verification, it is now followed by identification of the incoming
4114 connection (so you can more easily find what provoked it).
4116 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4118 5. Added some features to "harden" Exim a bit more against certain attacks:
4120 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4121 be put in Local/Makefile. This is like the never_users runtime option,
4122 but it cannot be overridden. The default setting is "root".
4124 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4125 prefix string with which any file named in a -C command line option
4128 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4129 is retained for -C and -D only if the caller of Exim is root. Without
4130 it, the exim user may also use -C and -D and retain privilege.
4132 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4133 command line option is disabled.
4135 6. Macro names set by the -D option must start with an upper case letter, just
4136 like macro names defined in the configuration file.
4138 7. Added "dereference=" facility to LDAP.
4140 8. Two instances of the typo "uknown" in the source files are fixed.
4142 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4143 the Configure-Makefile script screwed up while processing it.
4145 10. Incorporated PCRE 4.4.
4147 11. The SMTP synchronization check was not operating right at the start of an
4148 SMTP session. For example, it could not catch a HELO sent before the client
4149 waited for the greeting. There is now a check for outstanding input at the
4150 point when the greeting is written. Because of the duplex, asynchronous
4151 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4152 way, but not yet received, when the check is performed.
4154 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4155 on TCP/IP sockets, because this apparently causes some broken clients to
4158 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4159 unchanged) from the Cygwin maintainer.
4161 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4162 was supported instead of testing for mbx. Effectively a typo.
4164 15. The spa authenticator server code was not checking that the input it
4165 received was valid base64.
4167 16. The debug output line for the "set" modifier in ACLs was not showing the
4168 name of the variable that was being set.
4170 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4172 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4173 lookup on its own. Something else was needed to trigger the lookup. For
4174 example, a match in host_lookup or the need for a host name in a host list.
4175 Now, if $sender_host_name is referenced and the host name has not yet been
4176 looked up, a lookup is performed. If the lookup fails, the variable remains
4177 empty, and $host_lookup_failed is set to "1".
4179 19. Added "eqi" as a case-independent comparison operator.
4181 20. The saslauthd authentication condition could segfault if neither service
4182 nor realm was specified.
4184 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4185 error message that was logged was misleading, and incoming SMTP
4186 connections were dropped. The message is now more accurate, and temporary
4187 errors are given to SMTP connections.
4189 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4190 (except RSET) with a 421 error, until QUIT is received. However, it was
4191 failing to send a response to QUIT.
4193 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4194 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4195 ACL is now run after the helo_try_verify_hosts code.
4197 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4198 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4199 case-independent, so other case variants are also recognized). Apparently
4200 some systems use these upper case variants.
4202 25. If more than two messages were waiting for the same host, and a transport
4203 filter was specified for the transport, Exim sent two messages over the
4204 same TCP/IP connection, and then failed with "socket operation on non-
4205 socket" when it tried to send the third.
4207 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4209 27. The extern definition of crypt16() in expand.c was not being excluded when
4210 the OS had its own crypt16() function.
4212 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4213 as a preferred synonym for return_size_limit, both as an option and as an
4216 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4218 30. Changed the default configuration ACL to relax the local part checking rule
4219 for addresses that are not in any local domains. For these addresses,
4220 slashes and pipe symbols are allowed within local parts, but the sequence
4221 /../ is explicitly forbidden.
4223 31. SPA server authentication was not clearing the challenge buffer before
4226 32. log_message in a "warn" ACL statement was writing to the reject log as
4227 well as to the main log, which contradicts the documentation and doesn't
4228 seem right (because no rejection is happening). So I have stopped it.
4230 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4231 However, I am unable to do any testing of this.
4233 34. Fixed an infelicity in the appendfile transport. When checking directories
4234 for a mailbox, to see if any needed to be created, it was accidentally
4235 using path names with one or more superfluous leading slashes; tracing
4236 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4238 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4239 contents are added to the log line that is written for every discarded
4240 recipient. (Previously a log_message setting was ignored.)
4242 36. The ${quote: operator now quotes the string if it is empty.
4244 37. The install script runs exim in order to find its version number. If for
4245 some reason other than non-existence or emptiness, which it checks, it
4246 could not run './exim', it was installing it with an empty version number,
4247 i.e. as "exim-". This error state is now caught, and the installation is
4250 38. An argument was missing from the function that creates an error message
4251 when Exim fails to connect to the socket for saslauthd authentication.
4252 This could cause Exim to crash, or give a corrupted message.
4254 39. Added isip, isip4, and isip6 to ${if conditions.
4256 40. The ACL variables $acl_xx are now saved with the message, and can be
4257 accessed later in routers, transports, and filters.
4259 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4260 strings in the file are not string-expanded.
4262 42. If a MAIL command specified a SIZE value that was too large to fit into an
4263 int variable, the check against message_size_limit failed. Such values are
4264 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4265 day this will have to be increased, but I don't think I want to be around
4266 when emails are that large.
4273 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4274 iconv() is not standard in FreeBSD.
4276 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4277 IPv6 enabled. The observed symptom was a segmentation fault on return from
4278 the function os_common_find_running_interfaces() in src/os.c.
4280 3. In the check_special_case() function in daemon.c I had used "errno" as an
4281 argument name, which causes warnings on some systems. This was basically a
4282 typo, since it was named "eno" in the comments!
4284 4. The code that waits for the clock to tick (at a resolution of some fraction
4285 of a second) so as to ensure message-id uniqueness was always waiting for
4286 at least one whole tick, when it could have waited for less. [This is
4287 almost certainly not relevant at current processor speeds, where it is
4288 unlikely to ever wait at all. But we try to future-proof.]
4290 5. The function that sleeps for a time interval that includes fractions of a
4291 second contained a race. It did not block SIGALRM between setting the
4292 timer, and suspending (a couple of lines later). If the interval was short
4293 and the sigsuspend() was delayed until after it had expired, the suspension
4294 never ended. On busy systems this could lead to processes getting stuck for
4297 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4298 process, before it forks any delivery processes. The open lookup caching
4299 mechanism meant that the open file or database connection was passed into
4300 the delivery process. The problem was that delivery processes always tidy
4301 up cached lookup data. This could cause a problem for the next delivery
4302 process started by the queue runner, because the external queue runner
4303 process does not know about the closure. So the next delivery process
4304 still has data in the lookup cache. In the case of a file lookup, there was
4305 no problem because closing a file descriptor in a subprocess doesn't affect
4306 the parent. However, if the lookup was caching a connection to a database,
4307 the connection was closed, and the second delivery process was likely to
4308 see errors such as "PGSQL: query failed: server closed the connection
4309 unexpectedly". The problem has been fixed by closing all cached lookups
4310 in a queue runner before running a delivery process.
4312 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4313 seem to have the "const" qualifier which it has on other OS. I've
4316 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4317 *of the same type* (client or server) with the same public name that an
4318 error should be diagnosed.
4320 9. When Exim looked up a host name for an IP address, but failed to find the
4321 original IP address when looking up the host name (a safety check), it
4322 output the message "<ip address> does not match any IP for NULL", which was
4323 confusing, to say the least. The bug was that the host name should have
4324 appeared instead of "NULL".
4326 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4327 user that is built into the binary, and the -C or -D options is used, root
4328 privilege is dropped before the configuration file is read. In addition,
4329 logging is switched to stderr instead of the normal log files. If the
4330 configuration then re-defines the Exim user, the unprivileged environment
4331 is probably not what is expected, so Exim logs a panic warning message (but
4334 However, if deliver_drop_privilege is set, the unprivileged state may well
4335 be exactly what is intended, so the warning has been cut out in that case,
4336 and Exim is allowed to try to write to its normal log files.
4342 1. smtp_return_error_details was not giving details for temporary sender
4343 or receiver verification errors.
4345 2. Diagnose a configuration error if two authenticators have the same public
4348 3. Exim used not to create the message log file for a message until the first
4349 delivery attempt. This could be confusing when incoming messages were held
4350 for policy or load reasons. The message log file is now created at the time
4351 the message is received, and an initial "Received" line is written to it.
4353 4. The automatically generated man page for command line options had a minor
4354 bug that caused no ill effects; however, a more serious problem was that
4355 the procedure for building the man page automatically didn't always
4356 operate. Consequently, release 4.20 contains an out-of-date version. This
4357 shouldn't happen again.
4359 5. When building Exim with embedded Perl support, the script that builds the
4360 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4361 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4363 6. The freeze_tell option was not being used for messages that were frozen on
4364 arrival, either by an ACL or by local_scan().
4366 7. Added the smtp_incomplete_transaction log selector.
4368 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4369 was accepting AUTH without a new EHLO.
4371 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4372 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4375 10. Logging of TCP/IP connections (when configured) now happens in the main
4376 daemon process instead of the child process, so that the TCP/IP connection
4377 count is more accurate (but it can never be perfect).
4379 11. The use of "drop" in a nested ACL was not being handled correctly in the
4380 outer ACL. Now, if condition failure induced by the nested "drop" causes
4381 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4382 or "require"), the connection is dropped.
4384 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4385 that yield "discard" can now be used with an "accept" or a "discard" verb,
4386 but an error is generated for any others (because I can't see a useful way
4387 to define what should happen).
4389 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4390 are now processed in the same way as lines in the Exim configuration file.
4391 In particular, continuation lines are supported.
4393 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4395 15. Added -ti meaning -t -i.
4397 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4398 domains, and warn by logging if others are found.
4400 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4401 returned from GIFCONF when Exim is trying to find the list of interfaces on
4402 a host. The code in os.c has been modified to copy each ifreq to an aligned
4403 structure in all cases.
4405 Also, in some cases, the returned ifreq's were being copied to a 'struct
4406 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4407 means the last couple of bytes of an IPv6 address could be chopped if the
4408 ifreq contained only a normal sockaddr (14 bytes storage).
4410 18. Named domain lists were not supported in the hosts_treat_as_local option.
4411 An entry such as +xxxx was not recognized, and was treated as a literal
4414 19. Ensure that header lines added by a DATA ACL are included in the reject log
4415 if the ACL subsequently rejects the message.
4417 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4418 MD5 (which is deprecated).
4420 21. When testing a filter file using -bf, Exim was writing a message when it
4421 took the sender from a "From " line in the message, but it was not doing so
4422 when it took $return_path from a Return-Path: header line. It now does.
4424 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4425 with a valid header line field name (a series of printing characters
4426 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4428 23. Changed "disc" in the source to "disk" to conform to the documentation and
4429 the book and for uniformity.
4431 24. Ignore Sendmail's -Ooption=value command line item.
4433 25. When execve() failed while trying to run a command in a pipe transport,
4434 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4435 could be confused with a return value of 69 from the command itself. This
4436 has been changed to 127, the value the shell returns if it is asked to run
4437 a non-existent command. The wording for the related log line suggests a
4438 non-existent command as the problem.
4440 26. If received_header_text expands to an empty string, do not add a Received:
4441 header line to the message. (Well, it adds a token one on the spool, but
4442 marks it "old" so that it doesn't get used or transmitted.)
4444 27. Installed eximstats 1.28 (addition of -nt option).
4446 28. There was no check for failure on the call to getsockname() in the daemon
4447 code. This can fail if there is a shortage of resources on the system, with
4448 ENOMEM, for example. A temporary error is now given on failure.
4450 29. Contrary to the C standard, it seems that in some environments, the
4451 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4452 program. Exim now does this explicitly; it affects the formatting of
4453 timestamps using strftime().
4455 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4456 complaints. I've now initialized all the variables, to avoid this.
4458 32. Header lines added by a system filter were not being "seen" during
4459 transport-time rewrites.
4461 33. The info_callback() function passed to OpenSSL is set up with type void
4462 (*)(SSL *, int, int), as described somewhere. However, when calling the
4463 function (actually a macro) that sets it up, the type void(*)() is
4464 expected. I've put in a cast to prevent warnings from picky compilers.
4466 34. If a DNS black list lookup found a CNAME record, but there were no A
4467 records associated with the domain it pointed at, Exim crashed.
4469 35. If a DNS black list lookup returned more than one A record, Exim ignored
4470 all but the first. It now scans all returned addresses if a particular IP
4471 value is being sought. In this situation, the contents of the
4472 $dnslist_value variable are a list of all the addresses, separated by a
4475 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4476 to accept a host name and all its aliases if the forward lookup for any of
4477 them yielded the IP address of the incoming connection. Now it accepts only
4478 those names whose forward lookup yields the correct IP address. Any other
4479 names are discarded. This closes a loophole whereby a rogue DNS
4480 administrator could create reverse DNS records to break through a
4481 wildcarded host restriction in an ACL.
4483 37. If a user filter or a system filter that ran in a subprocess used any of
4484 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4485 the wrong values were passed to the pipe command ($thisaddress had the
4486 value of $0, $0 had the value of $1, etc). This bug was introduced by
4487 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4489 38. Improved the line breaking for long SMTP error messages from ACLs.
4490 Previously, if there was no break point between 40 and 75 characters, Exim
4491 left the rest of the message alone. Two changes have been made: (a) I've
4492 reduced the minimum length to 35 characters; (b) if it can't find a break
4493 point between 35 and 75 characters, it looks ahead and uses the first one
4494 that it finds. This may give the occasional overlong line, but at least the
4495 remaining text gets split now.
4497 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4498 file descriptors might be low, and that setting 1000 would always raise it.
4499 It turns out that in some environments, the limit is already over 1000 and
4500 that lowering it causes trouble. So now Exim takes care not to decrease it.
4502 40. When delivering a message, the value of $return_path is set to $sender_
4503 address at the start of routing (routers may change the value). By an
4504 oversight, this default was not being set up when an address was tested by
4505 -bt or -bv, which affected the outcome if any router or filter referred to
4508 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4509 log only when "message" or "log_message" are set. However, if one of the
4510 conditions was an address verification, or a call to a nested ACL, the
4511 messages generated by the underlying test were being passed through. This
4512 no longer happens. The underlying message is available in $acl_verify_
4513 message for both "message" and "log_message" expansions, so it can be
4514 passed through if needed.
4516 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4517 new expansion $bh_ to give the encoded byte string without charset
4518 translation. Translation happens only if iconv() is available; HAVE_ICONV
4519 indicates this at build time. HEADERS_CHARSET gives the charset to
4520 translate to; headers_charset can change it in the configuration, and
4521 "headers charset" can change it in an individual filter file.
4523 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4524 that creates RFC 2047 encoded "words" labels them as that charset instead
4525 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4526 expansion operator; (ii) when Exim creates a From: line for a local
4527 message; (iii) when a header line is rewritten to include a "phrase" part.
4529 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4530 buggy, causing it to skip the first lines of messages whose message ID
4531 ended in 'D'. This would not have bitten before Exim release 4.14, because
4532 message IDs were unlikely to end in 'D' before then. The effect was to have
4533 incorrect size information for certain domains.
4535 45. #include "config.h" was missing at the start of the crypt16.c module. This
4536 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4539 46. If there was a timeout during a "random" callout check, Exim treated it as
4540 a failure of the random address, and carried on sending RSET and the real
4541 address. If the delay was just some slowness somewhere, the response to the
4542 original RCPT would be taken as a response to RSET and so on, causing
4543 mayhem of various kinds.
4545 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4546 when I implemented it. It didn't allow for the fact that some option values
4547 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4548 the right test for positive values.
4550 48. Domain names in DNS records are case-independent. Exim always looks them up
4551 in lower case. Some resolvers return domain names in exactly the case they
4552 appear in the zone file, that is, they may contain uppercase letters. Not
4553 all resolvers do this - some return always lower case. Exim was treating a
4554 change of case by a resolver as a change of domain, similar to a widening
4555 of a domain abbreviation. This triggered its re-routing code and so it was
4556 trying to route what was effectively the same domain again. This normally
4557 caused routing to fail (because the router wouldn't handle the domain
4558 twice). Now Exim checks for this case specially, and just changes the
4559 casing of the domain that it ultimately uses when it transmits the message
4562 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4565 50. If a filter generated a file delivery with a non-absolute name (possible if
4566 no home directory exists for the router), the forbid_file option was not
4569 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4570 the existing equality matching.
4572 52. Exim was using ints instead of ino_t variables in some places where it was
4573 dealing with inode numbers.
4575 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4576 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4577 called TMPDIR, and if it finds it is different, it changes its value.
4579 54. The smtp_printf() function is now made available to local_scan() so
4580 additional output lines can be written before returning. There is also an
4581 smtp_fflush() function to enable the detection of a dropped connection.
4582 The variables smtp_input and smtp_batched_input are exported to
4585 55. Changed the default runtime configuration: the message "Unknown user"
4586 has been removed from the ACL, and instead placed on the localuser router,
4587 using the cannot_route_message feature. This means that any verification
4588 failures that generate their own messages won't get overridden. Similarly,
4589 the "Unrouteable address" message that was in the ACL for unverifiable
4590 relay addresses has also been removed.
4592 56. Added hosts_avoid_esmtp to the smtp transport.
4594 57. The exicyclog script was not checking for the esoteric option
4595 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4596 will work only if exicyclog is run under the appropriate euid.
4598 58. Following a discussion on the list, the rules by which Exim recognises line
4599 endings on incoming messages have been changed. The -dropcr and drop_cr
4600 options are now no-ops, retained only for backwards compatibility. The
4601 following line terminators are recognized: LF CRLF CR. However, special
4602 processing applies to CR:
4604 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4605 nor a local message in the state where . is a terminator.
4607 (ii) If a bare CR is encountered in a header line, an extra space is added
4608 after the line terminator so as not to end the header. The reasoning
4609 behind this is that bare CRs in header lines are most likely either
4610 to be mistakes, or people trying to play silly games.
4612 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4613 was being incorrectly given as 18 bytes larger than it should have been.
4614 This is a VOB (very old bug).
4616 60. This may never have affected anything current, but just in case it has:
4617 When the local host is found other than at the start of a list of hosts,
4618 the local host, those with the same MX, and any that follow, are discarded.
4619 When the list in question was part of a longer list of hosts, the following
4620 hosts (not currently being processed) were also being discarded. This no
4621 longer happens. I'm not sure if this situation could ever has previously
4624 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4627 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4628 header. This is something that is recommended in a new Internet Draft, and
4629 is something that is documented as being done by Sendmail. There are two
4630 possible values. For messages generated by the autoreply transport, Exim
4633 Auto-Submitted: auto-replied
4635 whereas for all other generated messages (e.g. bounces) it adds
4637 Auto-Submitted: auto-generated
4639 63. The "personal" condition in filters now includes a test for the
4640 Auto-Submitted: header. If it contains the string "auto-" the message it
4641 not considered personal.
4643 64. Added rcpt_include_affixes as a generic transport option.
4645 65. Added queue_only_override (default true).
4647 66. Added the syslog_duplication option.
4649 67. If what should have been the first header line of a message consisted of
4650 a space followed by a colon, Exim was mis-interpreting it as a header line.
4651 It isn't of course - it is syntactically invalid and should therefore be
4652 treated as the start of the message body. The misbehaviour could have
4653 caused a number of strange effects, including loss of data in subsequent
4654 header lines, and spool format errors.
4656 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4657 client host had authenticated. This control can now be exercised by an ACL
4658 for more flexibility.
4660 69. By default, callouts do not happen when testing with -bh. There is now a
4661 variant, -bhc, which does actually run the callout code, including
4662 consulting and updating the callout cache.
4664 70. Added support for saslauthd authentication, courtesy of Alexander
4667 71. If statvfs() failed on the spool or log directories while checking their
4668 size for availability, Exim confusingly gave the error "space shortage".
4669 Furthermore, in debugging mode it crashed with a floating point exception.
4670 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4671 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4672 serious problem, Exim now writes to the main and panic logs when this
4673 happens, with details of the failure. It then refuses to accept the
4674 incoming message, giving the message "spool directory problem" or "log
4675 directory problem" with a 421 code for SMTP messages.
4677 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4678 0, 1, and 2 exist, because some OS complain for execs without them (see
4679 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4680 descriptors. However, the code omitted to check that the open succeeded,
4681 causing mysterious errors if for some reason the permissions on /dev/null
4682 got screwed. Now Exim writes a message to the main and panic logs, and
4683 bombs out if it can't open /dev/null.
4685 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4686 interact so that it is all more flexible. It is supposed to remain
4687 backwards compatible. Also added extra_local_interfaces.
4689 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4690 to bomb out with an assertion failure - to the client this appears as a
4691 connection drop. This problem occurs in the part of the code that was taken
4692 from the Samba project. Fortunately, the assertion is in a very simple
4693 function, so I have fixed this by reproducing the function inline in the
4694 one place where it is called, and arranging for authentication to fail
4695 instead of killing the process with assert().
4697 75. The SPA client code was not working when the server requested OEM rather
4698 than Unicode encoding.
4700 76. Added code to make require_files with a specific uid setting more usable in
4701 the case where statting the file as root fails - usually a non-root-mounted
4702 NFS file system. When this happens and the failure is EACCES, Exim now
4703 forks a subprocess and does the per-uid checking as the relevant uid.
4705 77. Added process_log_path.
4707 78. If log_file_path was not explicitly set, a setting of check_log_space or
4708 check_log_inodes was ignored.
4710 79. If a space check for the spool or log partitions fails, the incident is now
4711 logged. Of course, in the latter case the data may get lost...
4713 80. Added the %p formatting code to string_format() so that it can be used to
4714 print addresses in debug_print(). Adjusted all the address printing in the
4715 debugging in store.c to use %p rather than %d.
4717 81. There was a concern that a line of code in smtp_in.c could overflow a
4718 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4719 initially expressed, the concern was not well-founded, because trailing
4720 spaces are removed early. However, if the trailing spaces were followed by
4721 a NULL, they did not get removed, so the overflow was possible. Two fixes
4724 (a) I re-wrote the offending code in a cleaner fashion.
4725 (b) If an incoming SMTP command contains a NULL character, it is rejected
4728 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4729 runs initgroups(), so that if the Exim user is in any additional groups,
4730 they will be used during message reception.
4736 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob