1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.503 2007/04/16 10:31:58 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.
221 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
222 fixed by 4.65/MH/01 (is this a record?) are fixed:
224 (i) An empty string was always treated as zero by the numeric comparison
225 operators. This behaviour has been restored.
227 (ii) It is documented that the numeric comparison operators always treat
228 their arguments as decimal numbers. This was broken in that numbers
229 starting with 0 were being interpreted as octal.
231 While fixing these problems I realized that there was another issue that
232 hadn't been noticed. Values of message_size_limit (both the global option
233 and the transport option) were treated as octal if they started with 0.
234 The documentation was vague. These values are now always treated as
235 decimal, and I will make that clear in the documentation.
241 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
242 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
245 MH/01 Don't check that the operands of numeric comparison operators are
246 integers when their expansion is in "skipping" mode (fixes bug
247 introduced by 4.64-PH/07).
249 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
250 child addresses, Exim now panics and dies. Previously, because the count
251 is held in a short int, deliveries were likely to be lost. As such a
252 large number of recipients for a single message is ridiculous
253 (performance will be very, very poor), I have chosen to impose a limit
254 rather than extend the field.
260 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
261 leftover -K file (the existence of which was triggered by #402).
262 While we were at it, introduced process PID as part of the -K
263 filename. This should rule out race conditions when creating
266 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
267 processing considerably. Previous code took too long for large mails,
268 triggering a timeout which in turn triggers #401.
270 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
271 in the DK code in transports.c. sendfile() is not really portable,
272 hence the _LINUX specificness.
274 TF/01 In the add_headers option to the mail command in an Exim filter,
275 there was a bug that Exim would claim a syntax error in any
276 header after the first one which had an odd number of characters
279 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
280 callout verification, Exim cached a "reject" for the entire domain. This
281 is correct for most verifications, but it is not correct for a recipient
282 verification with use_sender or use_postmaster set, because in that case
283 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
284 case of MAIL FROM:<> rejection from other early rejections (e.g.
285 rejection of HELO). When verifying a recipient using a non-null MAIL
286 address, the cache is ignored if it shows MAIL FROM:<> rejection.
287 Whatever the result of the callout, the value of the domain cache is
288 left unchanged (for any other kind of callout, getting as far as trying
289 RCPT means that the domain itself is ok).
291 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
294 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
295 manifest itself as EPIPE rather than ECONNECT. When tidying away a
296 session, the daemon ignores ECONNECT errors and logs others; it now
297 ignores EPIPE as well.
299 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
300 (quoted-printable decoding).
302 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
303 later the small subsequent patch to fix an introduced bug.
305 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
307 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
309 PH/08 An error is now given if message_size_limit is specified negative.
311 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
312 to be given (somewhat) arbitrary names.
314 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
317 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
318 miscellaneous code fixes
320 PH/10 Added the log_reject_target ACL modifier to specify where to log
323 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
324 hostname. This is wrong, because it relates to the incoming message (and
325 probably the interface on which it is arriving) and not to the outgoing
326 callout (which could be using a different interface). This has been
327 changed to use the value of the helo_data option from the smtp transport
328 instead - this is what is used when a message is actually being sent. If
329 there is no remote transport (possible with a router that sets up host
330 addresses), $smtp_active_hostname is used.
332 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
333 tweaks were necessary in order to get it to work (see also 21 below):
334 (a) The code assumed that strncpy() returns a negative number on buffer
335 overflow, which isn't the case. Replaced with Exim's string_format()
337 (b) There were several signed/unsigned issues. I just did the minimum
338 hacking in of casts. There is scope for a larger refactoring.
339 (c) The code used strcasecmp() which is not a standard C function.
340 Replaced with Exim's strcmpic() function.
341 (d) The code set only $1; it now sets $auth1 as well.
342 (e) A simple test gave the error "authentication client didn't specify
343 service in request". It would seem that Dovecot has changed its
344 interface. Fortunately there's a specification; I followed it and
345 changed what the client sends and it appears to be working now.
347 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
350 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
351 address A is aliased to B and C, where B exists and C does not. Without
352 -v the output is "A verified" because verification stops after a
353 successful redirection if more than one address is generated. However,
354 with -v the child addresses are also verified. Exim was outputting "A
355 failed to verify" and then showing the successful verification for C,
356 with its parentage. It now outputs "B failed to verify", showing B's
357 parentage before showing the successful verification of C.
359 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
360 look up a TXT record in a specific list after matching in a combined
363 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
364 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
365 they consult the DNS. I had assumed they would set it the way they
366 wanted; and indeed my experiments on Linux seem to show that in some
367 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
368 To be on the safe side, however, I have now made the interface to
369 host_find_byname() similar to host_find_bydns(), with an argument
370 containing the DNS resolver options. The host_find_byname() function now
371 sets these options at its start, just as host_find_bydns() does. The smtp
372 transport options dns_qualify_single and dns_search_parents are passed to
373 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
374 of host_find_byname() use the default settings of RES_DEFNAMES
375 (qualify_single) but not RES_DNSRCH (search_parents).
377 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
378 spool_read_header() do less string testing, by means of a preliminary
379 switch on the second character of optional "-foo" lines. (This is
380 overdue, caused by the large number of possibilities that now exist.
381 Originally there were few.) While I was there, I also converted the
382 str(n)cmp tests so they don't re-test the leading "-" and the first
383 character, in the hope this might squeeze out yet more improvement.
385 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
386 flag allowing group syntax was set by the header_syntax check but not
387 turned off, possible causing trouble later; (2) The flag was not being
388 set at all for the header_verify test, causing "group"-style headers to
389 be rejected. I have now set it in this case, and also caused header_
390 verify to ignore an empty address taken from a group. While doing this, I
391 came across some other cases where the code for allowing group syntax
392 while scanning a header line wasn't quite right (mostly, not resetting
393 the flag correctly in the right place). These bugs could have caused
394 trouble for malformed header lines. I hope it is now all correct.
396 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
397 with the "reply" argument non-NULL. The code, however (which originally
398 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
399 but it didn't always do it. This confused somebody who was copying the
400 code for some other use. I have removed all the tests.
402 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
403 feature that was used to support insecure browsers during the U.S. crypto
404 embargo. It requires special client support, and Exim is probably the
405 only MTA that supported it -- and would never use it because real RSA is
406 always available. This code has been removed, because it had the bad
407 effect of slowing Exim down by computing (never used) parameters for the
408 RSA_EXPORT functionality.
410 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
411 authenticator to fail if there's a tab character in the incoming data
412 (there should never be unless someone is messing about, as it's supposed
413 to be base64-encoded). Also added, on Timo's advice, the "secured" option
414 if the connection is using TLS or if the remote IP is the same as the
415 local IP, and the "valid-client-cert option" if a client certificate has
418 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
419 authenticators. This can be used for authorization after authentication
420 succeeds. (In the case of plaintext, it servers for both authentication
423 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
424 if any retry times were supplied.
426 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
427 connection, where there is no client IP address to check. In this
428 situation, the verify now always succeeds.
430 PH/25 Applied John Jetmore's -Mset patch.
432 PH/26 Added -bem to be like -Mset, but loading a message from a file.
434 PH/27 In a string expansion for a processed (not raw) header when multiple
435 headers of the same name were present, leading whitespace was being
436 removed from all of them, but trailing whitespace was being removed only
437 from the last one. Now trailing whitespace is removed from each header
438 before concatenation. Completely empty headers in a concatenation (as
441 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
442 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
444 PH/29 [Removed. This was a change that I later backed out, and forgot to
445 correct the ChangeLog entry (that I had efficiently created) before
446 committing the later change.]
448 PH/30 Exim was sometimes attempting to deliver messages that had suffered
449 address errors (4xx response to RCPT) over the same connection as other
450 messages routed to the same hosts. Such deliveries are always "forced",
451 so retry times are not inspected. This resulted in far too many retries
452 for the affected addresses. The effect occurred only when there were more
453 hosts than the hosts_max_try setting in the smtp transport when it had
454 the 4xx errors. Those hosts that it had tried were not added to the list
455 of hosts for which the message was waiting, so if all were tried, there
456 was no problem. Two fixes have been applied:
458 (i) If there are any address or message errors in an SMTP delivery, none
459 of the hosts (tried or untried) are now added to the list of hosts
460 for which the message is waiting, so the message should not be a
461 candidate for sending over the same connection that was used for a
462 successful delivery of some other message. This seems entirely
463 reasonable: after all the message is NOT "waiting for some host".
464 This is so "obvious" that I'm not sure why it wasn't done
465 previously. Hope I haven't missed anything, but it can't do any
466 harm, as the worst effect is to miss an optimization.
468 (ii) If, despite (i), such a delivery is accidentally attempted, the
469 routing retry time is respected, so at least it doesn't keep
470 hammering the server.
472 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
473 in ${readsocket because some servers need this prod.
475 PH/32 Added some extra debug output when updating a wait-xxx database.
477 PH/33 The hint "could be header name not terminated by colon", which has been
478 given for certain expansion errors for a long time, was not being given
479 for the ${if def:h_colon_omitted{... case.
481 PH/34 The spec says: "With one important exception, whenever a domain list is
482 being scanned, $domain contains the subject domain." There was at least
483 one case where this was not true.
485 PH/35 The error "getsockname() failed: connection reset by peer" was being
486 written to the panic log as well as the main log, but it isn't really
487 panic-worthy as it just means the connection died rather early on. I have
488 removed the panic log writing for the ECONNRESET error when getsockname()
491 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
492 runs only) independently of the message's sender address. This meant
493 that, if the 4xx error was in fact related to the sender, a different
494 message to the same recipient with a different sender could confuse
495 things. In particualar, this can happen when sending to a greylisting
496 server, but other circumstances could also provoke similar problems.
497 I have changed the default so that the retry time for these errors is now
498 based a combination of the sender and recipient addresses. This change
499 can be overridden by setting address_retry_include_sender=false in the
502 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
503 remote server are returned as part of bounce messages. This was not
504 happening for LMTP over a pipe (the lmtp transport), but now it is the
505 same for both kinds of LMTP.
507 PH/38 Despite being documented as not happening, Exim was rewriting addresses
508 in header lines that were in fact CNAMEs. This is no longer the case.
510 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
511 and queue runs started by the daemon processed all messages. This has
512 been fixed so that -R and -S can now usefully be given with -q<time>.
514 PH/40 Import PCRE release 6.7 (fixes some bugs).
516 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
518 PH/42 Give an error if -q is specified more than once.
520 PH/43 Renamed the variables $interface_address and $interface_port as
521 $received_ip_address and $received_port, to make it clear that these
522 values apply to message reception, and not to the outgoing interface when
523 a message is delivered. (The old names remain recognized, of course.)
525 PH/44 There was no timeout on the connect() call when using a Unix domain
526 socket in the ${readsocket expansion. There now is.
528 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
529 be meaningful with "accept".
531 SC/01 Eximstats V1.43
532 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
534 SC/02 Eximstats V1.44
535 Use a glob alias rather than an array ref in the generated
536 parser. This improves both readability and performance.
538 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
539 Collect SpamAssassin and rejection statistics.
540 Don't display local sender or destination tables unless
541 there is data to show.
542 Added average volumes into the top table text output.
544 SC/04 Eximstats V1.46
545 Collect data on the number of addresses (recipients)
546 as well as the number of messages.
548 SC/05 Eximstats V1.47
549 Added 'Message too big' to the list of mail rejection
550 reasons (thanks to Marco Gaiarin).
552 SC/06 Eximstats V1.48
553 Mainlog lines which have GMT offsets and are too short to
554 have a flag are now skipped.
556 SC/07 Eximstats V1.49 (Alain Williams)
557 Added the -emptyok flag.
559 SC/08 Eximstats V1.50
560 Fixes for obtaining the IP address from reject messages.
562 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
563 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
564 whitesspace changes from 4.64-PH/27
566 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
569 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
570 are found, allow negative numbers in numeric criteria)
572 JJ/06 exipick.20061117.2, added new $message_body_missing variable
574 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
575 to match changes made in 4.64-PH/43
577 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
579 PH/47 Put in an explicit test for a DNS lookup of an address record where the
580 "domain" is actually an IP address, and force a failure. This locks out
581 those revolvers/nameservers that support "A-for-A" lookups, in
582 contravention of the specifications.
584 PH/48 When a host name was looked up from an IP address, and the subsequent
585 forward lookup of the name timed out, the host name was left in
586 $sender_host_name, contrary to the specification.
588 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
589 restricted to single-key lookups, Exim was not diagnosing an error if
590 * or *@ was used with a query-style lookup.
592 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
594 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
595 long ago, but noone interested enough thought of it. Let's just say that
596 the "1.1" means that there are some new functions that weren't there at
597 some point in the past.
599 PH/51 Error processing for expansion failure of helo_data from an smtp
600 transport during callout processing was broken.
602 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
603 tested/used via the -bh/-bhc/-bs options.
605 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
606 bug, fixed in subsequent PCRE releases).
608 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
609 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
611 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
617 SC/01 Use a glob alias rather than an array ref in eximstats generated
618 parser. This improves both readability and performance.
620 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
621 Don't display local sender or destination tables in eximstats unless
622 there is data to show.
623 Added average volumes into the eximstats top table text output.
625 SC/03 Collect data on the number of addresses (recipients) as well
626 as the number of messages in eximstats.
628 TF/01 Correct an error in the documentation for the redirect router. Exim
629 does (usually) call initgroups() when daemonizing.
631 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
632 with consistent privilege compared to when running as a daemon.
634 TF/03 Note in the spec that $authenticated_id is not set for local
635 submissions from trusted users.
637 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
638 Thanks to Dean Brooks <dean@iglou.com> for the patch.
640 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
641 by adding some example configuration directives to the default
642 configuration file. A little bit of work is required to uncomment the
643 directives and define how usernames and passwords are checked, but
644 there is now a framework to start from.
646 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
647 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
648 without this. I don't know how relevant this is to other LDAP libraries.
650 PH/02 Add the verb name to the "unknown ACL verb" error.
652 PH/03 Magnus Holmgren's patch for filter_prepend_home.
654 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
656 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
657 directory not expanded when it should be if an expanded home directory
658 was set for the address (which is overridden by the transport).
660 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
663 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
664 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
665 because it is too late at that time, and has no effect.
667 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
668 security issue with \' (bugzilla #107). I could not use the
669 PQescapeStringConn() function, because it needs a PGconn value as one of
672 PH/08 When testing addresses using -bt, indicate those final addresses that
673 are duplicates that would not cause an additional delivery. At least one
674 person was confused, thinking that -bt output corresponded to deliveries.
675 (Suppressing duplicates isn't a good idea as you lose the information
676 about possibly different redirections that led to the duplicates.)
678 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
679 systems where poll() doesn't work, in particular OS X.
681 PH/10 Added more information to debugging output for retry time not reached.
683 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
684 operations in malware.c.
686 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
689 PH/13 If write_rejectlog was set false when logging was sent to syslog with
690 syslog_duplication set false, log lines that would normally be written
691 both the the main log and to the reject log were not written to syslog at
694 PH/14 In the default configuration, change the use of "message" in ACL warn
695 statements to "add_header".
697 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
698 not followed by a command (e.g. "seen endif").
700 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
701 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
704 PH/17 Added extra conditions to the default value of delay_warning_condition
708 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
709 { match{$h_precedence:}{(?i)bulk|list|junk} } \
710 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
713 The Auto-Submitted: and various List- headers are standardised, whereas I
714 don't think Precedence: ever was.
716 PH/18 Refactored debugging code in route_finduser() to show more information,
717 in particular, the error code if getpwnam() issues one.
719 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
720 This is apparently needed in addition to the PH/07 change above to avoid
721 any possible encoding problems.
723 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
724 but not after initializing Perl.
726 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
727 output them only if debugging. By default they are written stderr,
728 apparently, which is not desirable.
730 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
733 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
736 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
738 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
739 authenticated or an ident call has been made. Suppress the default
740 values for $authenticated_id and $authenticated_sender (but permit -oMai
741 and -oMas) when testing with -bh.
743 PH/24 Re-jigged the order of the tests in the default configuration so that the
744 tests for valid domains and recipients precede the DNS black list and CSA
745 tests, on the grounds that those ones are more expensive.
747 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
748 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
749 This bug exists in every version of Exim that I still have, right back to
752 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
753 However, an attempt to turn on case-sensitivity in a regex key by
754 including (?-i) didn't work because the subject string was already
755 lowercased, and the effects were non-intuitive. It turns out that a
756 one-line patch can be used to allow (?-i) to work as expected.
762 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
763 other effects) broke the use of negated acl sub-conditions.
765 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
768 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
769 "Deny" causes Exim to reject the incoming connection with a 554 error.
770 Unfortunately, if there is a major crisis, such as a disk failure,
771 tcp-wrappers gives "deny", whereas what one would like would be some
772 kind of temporary error. A kludge has been added to help with this.
773 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
774 554 error is used if errno is still zero or contains ENOENT (which occurs
775 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
778 PH/03 Add -lutil to the default FreeBSD LIBS setting.
780 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
781 errors. Otherwise a message that provokes a temporary error (when other
782 messages do not) can cause a whole host to time out.
784 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
785 addresses were routed directly to files or pipes from a redirect router.
786 File deliveries just didn't batch; pipe deliveries might have suffered
789 PH/06 A failure to get a lock for a hints database would erroneously always say
790 "Failed to get write lock", even when it was really a read lock.
792 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
793 of 0600. This has been changed to use the value of the lockfile_mode
794 option (which defaults to 0600).
796 PH/08 Applied small patch from the Sieve maintainer.
798 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
799 folder from quota calculations, a direct delivery into this folder messed
800 up the contents of the maildirsize file. This was because the regex was
801 used only to exclude .Trash (or whatever) when the size of the mailbox
802 was calculated. There was no check that a delivery was happening into an
803 excluded directory. This bug has been fixed by ignoring all quota
804 processing for deliveries into excluded directories.
806 PH/10 Added the maildirfolder_create_regex option to appendfile.
812 PH/01 The code for finding all the local interface addresses on a FreeBSD
813 system running IPv6 was broken. This may well have applied to all BSD
814 systems, as well as to others that have similar system calls. The broken
815 code found IPv4 interfaces correctly, but gave incorrect values for the
816 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
817 that it would not match correctly against @[] and not recognize the IPv6
820 PH/02 The ipliteral router was not recognizing addresses of the form user@
821 [ipv6:....] because it didn't know about the "ipv6:" prefix.
823 PH/03 Added disable_ipv6.
825 PH/04 Changed $reply_address to use the raw form of the headers instead of the
826 decoded form, because it is most often used to construct To: headers
827 lines in autoreplies, and the decoded form may well be syntactically
828 invalid. However, $reply_address has leading white space removed, and all
829 newlines turned into spaces so that the autoreply transport does not
832 PH/05 If group was specified without a user on a router, and no group or user
833 was specified on a transport, the group from the router was ignored.
835 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
836 for visible compile-time settings that can be used to change these
837 numbers, for those that want even more. Backwards compatibility with old
838 spool files has been maintained. However, going back to a previous Exim
839 release will lost any variables that are in spool files.
841 PH/07 Two small changes when running in the test harness: increase delay when
842 passing a TCP/IP connection to a new process, in case the original
843 process has to generate a bounce, and remove special handling of
844 127.0.0.2 (sic), which is no longer necessary.
846 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
847 be the same on different OS.
849 PH/09 Moved a debug statement in filter processing to avoid a race problem when
852 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
853 whether --show-vars was specified or not
855 JJ/02 exipick: Added support for new ACL variable spool format introduced
858 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
859 syntactically invalid From: or Reply-to: line, and a filter used this to
860 generate an autoreply, and therefore failed to obtain an address for the
861 autoreply, Exim could try to deliver to a non-existent relative file
862 name, causing unrelated and misleading errors. What now happens is that
863 it logs this as a hard delivery error, but does not attempt to create a
866 PH/11 The exinext utility has a -C option for testing purposes, but although
867 the given file was scanned by exinext itself; it wasn't being passed on
868 when Exim was called.
870 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
871 an end-of-file indication when reading a command response.
873 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
874 compiled. In many other places in Exim, IPv6 addresses are always
875 recognized, so I have changed this. It also means that IPv4 domain
876 literals of the form [IPV4:n.n.n.n] are now always recognized.
878 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
879 used if the router is not running as root, for example, when verifying at
880 ACL time, or when using -bh. The debugging output from this situation was
881 non-existent - all you got was a failure to exec. I have made two
884 (a) Failures to set uid/gid, the current directory, or a process leader
885 in a subprocess such as that created by queryprogram now generate
886 suitable debugging ouput when -d is set.
888 (b) The queryprogram router detects when it is not running as root,
889 outputs suitable debugging information if -d is set, and then runs
890 the subprocess without attempting to change uid/gid.
892 PH/15 Minor change to Makefile for building test_host (undocumented testing
895 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
896 additional section of a DNS packet that returns MX or SRV records.
897 Instead, it always explicitly searches for A/AAAA records. This avoids
898 major problems that occur when a DNS server includes only records of one
899 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
900 fixed another bug: if SRV records were looked up and the corresponding
901 address records were *not* found in the additional section, the port
902 values from the SRV records were lost.
904 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
905 using the correct key (the original address) when searching the retry
906 rules in order to find which one to use for generating the retry hint.
908 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
909 adding the default one. Similarly, if it contains a Reply-To: header, the
910 errors_reply_to option, if set, is not used.
912 PH/19 When calculating a retry time, Exim used to measure the "time since
913 failure" by looking at the "first failed" field in the retry record. Now
914 it does not use this if it is later than than the arrival time of the
915 message. Instead it uses the arrival time. This makes for better
916 behaviour in cases where some deliveries succeed, thus re-setting the
917 "first failed" field. An example is a quota failure for a huge message
918 when small messages continue to be delivered. Without this change, the
919 "time since failure" will always be short, possible causing more frequent
920 delivery attempts for the huge message than are intended.
921 [Note: This change was subsequently modified - see PH/04 for 4.62.]
923 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
924 $1, $2, $3) because the numerical variables can be reset during some
925 expansion items (e.g. "match"), thereby losing the authentication data.
927 PH/21 Make -bV show the size of off_t variables so that the test suite can
928 decide whether to run tests for quotas > 2G.
930 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
931 mailbox_size, and mailbox_filecount in the appendfile transport. If a
932 filecount value is greater than 2G or if a quota value is greater than 2G
933 on a system where the size of off_t is not greater than 4, a panic error
936 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
937 never match. The debug and -bh output now contains an explicit error
938 message indicating a malformed IPv4 address or mask.
940 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
941 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
944 PH/25 Do not write to syslog when running in the test harness. The only
945 occasion when this arises is a failure to open the main or panic logs
946 (for which there is an explicit test).
948 PH/26 Added the /no_tell option to "control=freeze".
950 PH/27 If a host name lookup failed very early in a connection, for example, if
951 the IP address matched host_lookup and the reverse lookup yielded a name
952 that did not have a forward lookup, an error message of the form "no IP
953 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
954 could be logged. Now it outputs the IP address instead of "NULL".
956 PH/28 An enabling patch from MH: add new function child_open_exim2() which
957 allows the sender and the authenticated sender to be set when
958 submitting a message from within Exim. Since child_open_exim() is
959 documented for local_scan(), the new function should be too.
961 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
962 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
963 results in an empty string is now treated as unset.
965 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
967 PH/31 Added sender_verify_fail logging option.
969 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
970 needed by RFC 822 but not by RFC 2822 was commented out. I have now
971 tidied the source and removed it altogether.
973 PH/33 When a queue run was abandoned because the load average was too high, a
974 log line was always written; now it is written only if the queue_run log
975 selector is set. In addition, the log line for abandonment now contains
976 information about the queue run such as the pid. This is always present
977 in "start" and "stop" lines but was omitted from the "abandon" line.
979 PH/34 Omit spaces between a header name and the colon in the error message that
980 is given when verify = headers_syntax fails (if there are lots of them,
981 the message gets confusing).
983 PH/35 Change the default for dns_check_names_pattern to allow slashes within
984 names, as there are now some PTR records that contain slashes. This check
985 is only to protect against broken name servers that fall over on strange
986 characters, so the fact that it applies to all lookups doesn't matter.
988 PH/36 Now that the new test suite is complete, we can remove some of the
989 special code in Exim that was needed for the old test suite. For example,
990 sorting DNS records because real resolvers return them in an arbitrary
991 order. The new test suite's fake resolver always returns records in the
994 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
995 bounces) except when queue_only is set, to avoid logging races between
996 the different processes.
998 PH/38 Panic-die if .include specifies a non-absolute path.
1000 PH/39 A tweak to the "H" retry rule from its user.
1002 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1003 a label. They prevented compilation on older perls.
1005 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1006 a warning to be raised on newish perls.
1008 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1009 on queue. Changes to match documented behaviour of showing count of
1010 messages matching specified criteria.
1012 PH/40 Changed the default ident timeout from 30s to 5s.
1014 PH/41 Added support for the use of login_cap features, on those BSD systems
1015 that have them, for controlling the resources used by pipe deliveries.
1017 PH/42 The content-scanning code uses fopen() to create files in which to put
1018 message data. Previously it was not paying any attention to the mode of
1019 the files. Exim runs with umask(0) because the rest of the code creates
1020 files with open(), and sets the required mode explicitly. Thus, these
1021 files were ending up world-writeable. This was not a big issue, because,
1022 being within the spool directory, they were not world-accessible. I have
1023 created a function called modefopen, which takes an additional mode
1024 argument. It sets umask(777), creates the file, chmods it to the required
1025 mode, then resets the umask. All the relevant calls to fopen() in the
1026 content scanning code have been changed to use this function.
1028 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1029 to 24 hours. This avoids potential overflow problems when processing G
1030 and H retry rules. I suspect nobody ever tinkers with this value.
1032 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1034 PH/45 When the plaintext authenticator is running as a client, the server's
1035 challenges are checked to ensure they are valid base64 strings. By
1036 default, the authentication attempt is cancelled if an invalid string is
1037 received. Setting client_ignore_invalid_base64 true ignores these errors.
1038 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1039 they are received. Thus, the responses can be made to depend on the
1040 challenges. If an invalid string is ignored, an empty string is placed in
1043 PH/46 Messages that are created by the autoreply transport now contains a
1044 References: header, in accordance with RFCs 2822 and 3834.
1046 PH/47 Added authenticated_sender_force to the smtp transport.
1048 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1050 PH/49 Installed latest patch from the Sieve maintainer.
1052 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1053 was also set, the appendfile transport was unnecessarily scanning a
1054 directory of message files (e.g. for maildir delivery) to find the count
1055 of files (along with the size), even though it did not need this
1056 information. It now does the scan only if it needs to find either the
1057 size of the count of files.
1059 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1061 PH/52 Two bugs concerned with error handling when the smtp transport is
1064 (i) Exim was not creating retry information for temporary errors given
1065 for individual recipients after the DATA command when the smtp transport
1066 was used in LMTP mode. This meant that they could be retried too
1067 frequently, and not timed out correctly.
1069 (ii) Exim was setting the flag that allows error details to be returned
1070 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1071 recipients that were returned after the DATA command.
1073 PH/53 This is related to PH/52, but is more general: for any failing address,
1074 when detailed error information was permitted to be returned to the
1075 sender, but the error was temporary, then after the final timeout, only
1076 "retry timeout exceeded" was returned. Now it returns the full error as
1077 well as "retry timeout exceeded".
1079 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1080 do this, and (what is worse) MTAs that accept it.
1082 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1083 will now be deprecated.
1085 PH/56 New os.c-cygwin from the Cygwin maintainer.
1087 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1088 formats (previously only available in exim formats via -bpr, -bpru,
1089 and -bpra. Now also available in native and exiqgrep formats)
1091 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1092 with very large, slow to parse queues
1094 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1096 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1098 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1099 responses to authentication challenges, though it was showing the
1100 challenges; (ii) I've removed the CR characters from the debug output for
1103 PH/58 Allow for the insertion of a newline as well as a space when a string
1104 is turned into more than one encoded-word during RFC 2047 encoding. The
1105 Sieve code now uses this.
1107 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1108 data_4xx, lost_connection, tls_required.
1110 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1111 message was being sent as an SMTP response.
1113 PH/61 Add -l and -k options to exicyclog.
1115 PH/62 When verifying, if an address was redirected to one new address, so that
1116 verification continued, and the new address failed or deferred after
1117 having set something in $address_data, the value of $address_data was not
1118 passed back to the ACL. This was different to the case when no
1119 redirection occurred. The value is now passed back in both cases.
1121 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1122 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1123 use login_cap.h, so on its own it isn't the distinguishing feature. The
1124 new name refers directly to the setclassresources() function.
1126 PH/65 Added configuration files for NetBSD3.
1128 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1130 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1131 is preferred over IPv4.
1133 PH/68 The bounce_return_message and bounce_return_body options were not being
1134 honoured for bounces generated during the reception of non-SMTP messages.
1135 In particular, this applied to messages rejected by the ACL. This bug has
1136 been fixed. However, if bounce_return_message is true and bounce_return_
1137 body is false, the headers that are returned for a non-SMTP message
1138 include only those that have been read before the error was detected.
1139 (In the case of an ACL rejection, they have all been read.)
1141 PH/69 The HTML version of the specification is now built in a directory called
1142 spec_html instead of spec.html, because the latter looks like a path with
1143 a MIME-type, and this confuses some software.
1145 PH/70 Catch two compiler warnings in sieve.c.
1147 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1148 function verify_get_ident() calls ip_connect() to connect a socket, but
1149 if the "connect()" function timed out, ip_connect() used to close the
1150 socket. However, verify_get_ident() also closes the socket later, and in
1151 between Exim writes to the log, which may get opened at this point. When
1152 the socket was closed in ip_connect(), the log could get the same file
1153 descriptor number as the socket. This naturally causes chaos. The fix is
1154 not to close the socket in ip_connect(); the socket should be closed by
1155 the function that creates it. There was only one place in the code where
1156 this was missing, in the iplookup router, which I don't think anybody now
1157 uses, but I've fixed it anyway.
1159 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1160 well as to direct DNS lookups. Otherwise the handling of names in host
1161 lists is inconsistent and therefore confusing.
1167 PH/01 Two changes to the default runtime configuration:
1169 (1) Move the checks for relay_from_hosts and authenticated clients from
1170 after to before the (commented out) DNS black list checks.
1172 (2) Add control=submission to the relay_from_hosts and authenticated
1173 clients checks, on the grounds that messages accepted by these
1174 statements are most likely to be submissions.
1176 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1178 (1) Generate an error if the third argument for the ${prvs expansion is
1181 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1184 (3) Reset the variables that are obtained from the first argument of
1185 ${prvscheck and used in the second argument before leaving the code,
1186 because their memory is reclaimed, so using them afterwards may do
1189 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1190 one (it's much easier than Tom thought :-).
1192 (5) Because of (4), we can now allow for the use of $prvscheck_result
1193 inside the third argument.
1195 PH/03 For some reason, the default setting of PATH when running a command from
1196 a pipe transport was just "/usr/bin". I have changed it to
1199 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1200 anything to be listed in the output from -bV.
1202 PH/05 When a filter generated an autoreply, the entire To: header line was
1203 quoted in the delivery log line, like this:
1205 => >A.N.Other <ano@some.domain> <original@ddress> ...
1207 This has been changed so that it extracts the operative address. There
1208 may be more than one such address. If so, they are comma-separated, like
1211 => >ano@some.domain,ona@other.domain <original@ddress> ...
1213 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1214 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1215 not being looked up in the rDNS to get a host name, Exim was showing the
1216 IP address twice in Received: lines, even though the IP addresses were
1217 identical. For example:
1219 Received: from [1.2.3.4] (helo=[1.2.3.4])
1221 However, if the real host name was known, it was omitting the HELO data
1222 if it matched the actual IP address. This has been tidied up so that it
1223 doesn't show the same IP address twice.
1225 PH/07 When both +timestamp and +memory debugging was on, the value given by
1226 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1227 called by the debug printing, thereby overwriting the timestamp buffer.
1228 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1230 PH/08 When the original message was included in an autoreply transport, it
1231 always said "this is a copy of the message, including all the headers",
1232 even if body_only or headers_only was set. It now gives an appropriate
1235 PH/09 Applied a patch from the Sieve maintainer which:
1237 o fixes some comments
1238 o adds the (disabled) notify extension core
1239 o adds some debug output for the result of if/elsif tests
1240 o points to the current vacation draft in the documentation
1241 and documents the missing references header update
1245 o fixes a bug in processing the envelope test (when testing
1246 multiple envelope elements, the last element determinted the
1249 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1250 Electronic Mail") by including:
1252 Auto-submitted: auto-generated
1254 in the messages that it generates (bounce messages and others, such as
1255 warnings). In the case of bounce messages for non-SMTP mesages, there was
1256 also a typo: it was using "Auto_submitted" (underscore instead of
1257 hyphen). Since every message generated by Exim is necessarily in response
1258 to another message, thes have all been changed to:
1260 Auto-Submitted: auto-replied
1262 in accordance with these statements in the RFC:
1264 The auto-replied keyword:
1266 - SHOULD be used on messages sent in direct response to another
1267 message by an automatic process,
1269 - MUST NOT be used on manually-generated messages,
1271 - MAY be used on Delivery Status Notifications (DSNs) and Message
1272 Disposition Notifications (MDNs),
1274 - MUST NOT be used on messages generated by automatic or periodic
1275 processes, except for messages which are automatic responses to
1278 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1279 to the default Received: header definition.
1281 PH/12 Added log selector acl_warn_skipped (default on).
1283 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1284 variables because (a) they are in the wrong storage pool and (b) even if
1285 they were copied, it wouldn't work properly because of the caching.
1287 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1288 checking when decoding. Apparently there are clients that generate
1289 overlong encoded strings. Why am I not surprised?
1291 PH/15 If the first argument of "${if match_address" was not empty, but did not
1292 contain an "@" character, Exim crashed. Now it writes a panic log message
1293 and treats the condition as false.
1295 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1297 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1298 extension "envelope-auth". The code is finished and in agreement with
1299 other implementations, but there is no documentation so far and in fact,
1300 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1301 not changing the active code.
1303 Print executed "if" and "elsif" statements when debugging is used. This
1304 helps a great deal to understand what a filter does.
1306 Document more things not specified clearly in RFC3028. I had all this
1307 sorted out, when out of a sudden new issues came to my mind. Oops."
1309 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1312 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1313 canonical form (as documented). However, after a host name lookup from
1314 the IP address, check_host() was doing a simple string comparison with
1315 addresses acquired from the DNS when checking that the found name did
1316 have the original IP as one of its addresses. Since any found IPv6
1317 addresses are likely to be in abbreviated form, the comparison could
1318 fail. Luckily, there already exists a function for doing the comparison
1319 by converting both addresses to binary, so now that is used instead of
1320 the text comparison.
1322 PH/20 There was another similar case to PH/19, when a complete host name was
1323 given in a host list; looking up its IP address could give an abbreviated
1324 form, whereas the current host's name might or might not be abbreviated.
1325 The same fix has been applied.
1331 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1332 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1335 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1336 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1338 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1340 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1341 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1342 Also, for "Auto-Submitted", treat anything other than "no" as
1343 non-personal, in accordance with RFC 3834. (Previously it treated
1344 anything starting "auto-" as non-personal.)
1346 TF/01 The control=submission/name=... option had a problem with syntax
1347 errors if the name included a slash character. The /name= option
1348 now slurps the rest of the string, so it can include any characters
1349 but it must come last in the list of options (after /sender_retain
1352 PH/05 Some modifications to the interface to the fake nameserver for the new
1360 TK/01 Added the "success_on_redirect" address verification option. See
1361 NewStuff for rationale and an example.
1363 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1365 PH/02 Patch to exigrep to allow it to work on syslog lines.
1367 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1368 fread() to skip over the body file's header line, because in Cygwin the
1369 header line is locked and is inaccessible.
1371 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1372 co-exist for some time) to make it clear that it is the Exim ID that is
1373 referenced, not the Message-ID: header line.
1375 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1376 string_format() function, because snprintf() does not exist on all
1379 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1380 ${stat: expansion item.
1382 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1383 protocol synchronization error", to keep the pedants happy.
1385 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1386 well as for IRIX systems, when gcc is being used. See the host.c source
1389 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1391 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1394 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1395 transport and to the smtp transport in LMTP mode.
1397 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1399 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1400 run a filter in a subprocess. This could lead to confusion in subsequent
1401 lookups in the parent process. There should also be a search_tidyup() at
1402 the end of the subprocess.
1404 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1405 only if the host matched helo_try_verify_hosts, which caused the
1406 verification to occur when the EHLO/HELO command was issued. The ACL just
1407 tested the remembered result. Now, if a previous verification attempt has
1408 not happened, "verify = helo" does it there and then.
1410 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1412 TK/03 Fix log output including CR from clamd.
1414 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1415 exist provoked a memory error which could cause a segfault.
1417 PH/15 Installed PCRE 6.2
1419 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1421 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1422 of the problem. Specifically, suggested +O2 rather than +O1 for the
1425 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1427 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1428 record, the port was not correctly logged when the outgoing_port log
1429 selector was set (it logged the transort's default port).
1431 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1432 fallback_hosts, and "hosts" in the smtp transport.
1434 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1435 host errors such as "Connection refused".
1437 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1438 authentication with radiusclient 0.4.9:
1440 - Error returned from rc_read_config was caught wrongly
1441 - Username/password not passed on to radius server due to wrong length.
1443 The presumption is that some radiusclient API changes for 4.51/PH/17
1444 were not taken care of correctly. The code is still untested by me (my
1445 Linux distribution still has 0.3.2 of radiusclient), but it was
1446 contributed by a Radius user.
1448 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1449 expanding the "port" option of the smtp transport.
1451 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1452 while reading a MIME header. Thanks to Tom Hughes for a patch.
1454 PH/24 Include config.h inside local_scan.h so that configuration settings are
1457 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1458 that in an ACL for RCPT (for example), you can examine exactly what was
1461 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1462 commands, but it was not correctly comparing the address with the actual
1463 client host address. Thus, it would show the EHLO address in Received:
1464 header lines when this was not necessary.
1466 PH/27 Added the % operator to ${eval:}.
1468 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1469 it should be ignoring failures (because with -C, for example, it has lost
1470 privilege). It wasn't ignoring creation failures other than "already
1473 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1476 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1477 because an input line was too long, either on its own, or by virtue of
1478 too many continuations, the temporary file was not being removed, and the
1479 return code was incorrect.
1481 PH/31 Missing "BOOL" in function definition in filtertest.c.
1483 PH/32 Applied Sieve patches from the maintainer.
1485 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1487 PH/33 Added "verify = not_blind".
1489 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1490 Local/Makefile (with some defaults set). These are used in built scripts
1491 such as exicyclog, but they have never been used in the exim_install
1492 script (though there are many overriding facilities there). I have
1493 arranged that the exim_install script now takes note of these two
1496 PH/35 Installed configuration files for Dragonfly.
1498 PH/36 When a locally submitted message by a trusted user did not contain a
1499 From: header, and the sender address was obtained from -f or from an SMTP
1500 MAIL command, and the trusted user did not use -F to supply a sender
1501 name, $originator_name was incorrectly used when constructing a From:
1502 header. Furthermore, $originator_name was used for submission mode
1503 messages from external hosts without From: headers in a similar way,
1504 which is clearly wrong.
1506 PH/37 Added control=suppress_local_fixups.
1508 PH/38 When log_selector = +received_sender was set, and the addition of the
1509 sender made the log line's construction buffer exactly full, or one byte
1510 less than full, an overflow happened when the terminating "\n" was
1513 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1514 when the result of a list match is failure because a DNS lookup failed.
1516 PH/40 RM_COMMAND is now used in the building process.
1518 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1519 the "build-* directories that it finds.
1521 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1522 address in a domain literal was a prefix of an interface address.
1524 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1525 when verifying a sender address, unless rewrite_headers is false.
1527 PH/44 Wrote a long comment about why errors_to addresses are verified as
1528 recipients, not senders.
1530 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1531 the ratelimit ACL was added.
1533 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1535 PH/46 Added extra information about PostgreSQL errors to the error string.
1537 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1538 avoiding the need to install special zones in a real server. This is
1539 backwards compatible; if it can't find the fake resolver, it drops back.
1540 Thus, both old and new test suites can be run.
1542 TF/02 Added util/ratelimit.pl
1544 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1545 clock is set back in time.
1547 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1548 Candler <B.Candler@pobox.com>.
1550 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1551 OK for addresses that are the result of redirections.
1553 PH/48 A number of further additions for the benefit of the new test suite,
1554 including a fake gethostbyname() that interfaces to the fake DNS resolver
1557 TF/06 The fix for widen_domains has also been applied to qualify_single and
1558 search_parents which are the other dnslookup options that can cause
1561 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1564 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1566 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1567 service" if it encountered an MX record with an empty target hostname.
1568 The message is now "an MX or SRV record indicated no SMTP service".
1570 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1571 if the DNS of the sending site is misconfigured. This is quite a
1572 common situation. This change restores the behaviour of treating a
1573 helo verification defer as a failure.
1575 PH/51 If self=fail was set on a router, the bounce message did not include the
1576 actual error message.
1582 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1584 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1585 command itself ended in error, the underlying message about the transport
1586 filter timeout was being overwritten with the pipe command error. Now the
1587 underlying error message should be appended to the second error message.
1589 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1591 PH/02 Reduce the amount of output that "make" produces by default. Full output
1592 can still be requested.
1594 PH/03 The warning log line about a condition test deferring for a "warn" verb
1595 was being output only once per connection, rather than after each
1596 occurrence (because it was using the same function as for successful
1597 "warn" verbs). This seems wrong, so I have changed it.
1599 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1600 it should not have, which might have caused a crash in the right
1601 circumstances, but probably never did.
1603 PH/04 Installed a modified version of Tony Finch's patch to make submission
1604 mode fix the return path as well as the Sender: header line, and to
1605 add a /name= option so that you can make the user's friendly name appear
1608 TF/03 Added the control = fakedefer ACL modifier.
1610 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1611 Mark Lowes for thorough testing.
1613 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1615 TK/03 Merged latest SRS patch from Miles Wilton.
1617 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1618 with the definition in sysexits.h (which is #included earlier).
1619 Fortunately, Exim does not actually use EX_OK. The code used to try to
1620 preserve the sysexits.h value, by assumimg that macro definitions were
1621 scanned for macro replacements. I have been disabused of this notion,
1622 so now the code just undefines EX_OK before #including unistd.h.
1624 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1625 in the smtp transport. When a block could not be written in a single
1626 write() function, the timeout was being re-applied to each part-write.
1627 This seems wrong - if the receiver was accepting one byte at a time it
1628 would take for ever. The timeout is now adjusted when this happens. It
1629 doesn't have to be particularly precise.
1631 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1632 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1634 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1635 without a domain if the check to <postmaster@domain> fails.
1637 SC/01 Eximstats: added -xls and the ability to specify output files
1638 (patch written by Frank Heydlauf).
1640 SC/02 Eximstats: use FileHandles for outputing results.
1642 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1644 SC/04 Eximstats: fixed display of large numbers with -nvr option
1646 SC/05 Eximstats: fixed merging of reports with empty tables.
1648 SC/06 Eximstats: added the -include_original_destination flag
1650 SC/07 Eximstats: removed tabs and trailing whitespace.
1652 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1654 TK/06 MBOX spool code: Add real "From " MBOX separator line
1655 so the .eml file is really in mbox format (even though
1656 most programs do not really care). Patch from Alex Miller.
1658 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1659 The latter is generated from $received_to and is only set if the
1660 message has one envelope recipient. SA can use these headers,
1661 obviously out-of-the-box. Patch from Alex Miller.
1663 PH/08 The ${def test on a variable was returning false if the variable's
1664 value was "0", contrary to what the specification has always said!
1665 The result should be true unless the variable is empty.
1667 PH/09 The syntax error of a character other than { following "${if
1668 def:variable_name" (after optional whitespace) was not being diagnosed.
1669 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1670 accidental colon was present, for example, could give incorrect results.
1672 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1673 result is used (not including appendfile, where other changes are about
1676 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1677 This involved changing a lot of size variables from int to off_t. It
1678 should work with maildirs and everything.
1680 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1681 spamd dying while we are connected to it.
1683 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1686 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1687 function for BDB 4.3.
1689 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1691 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1692 the sources for PCRE have been re-arranged and more files are now
1695 PH/15 The code I had for printing potentially long long variables in PH/11
1696 above was not the best (it lost precision). The length of off_t variables
1697 is now inspected at build time, and an appropriate printing format (%ld
1698 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1699 to be "long long int" or "long int". This is needed for the internal
1700 formatting function string_vformat().
1702 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1703 the configuration file to be ":syslog", then the script "guesses" where
1704 the logs files are, rather than using the compiled in default. In our
1705 case the guess is not the same as the compiled default, so the script
1706 suddenly stopped working when I started to use syslog. The patch checks
1707 to see if log_file_path is "". If so, it attempts to read it from exim
1708 with no configuration file to get the compiled in version, before it
1709 falls back to the previous guessing code."
1711 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1712 implementing BATV in an Exim configuration. See NewStuff for the gory
1715 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1716 Makefile that are specific to HP-UX.
1718 PH/18 If the "use_postmaster" option was set for a recipient callout together
1719 with the "random" option, the postmaster address was used as the MAIL
1720 FROM address for the random test, but not for the subsequent recipient
1721 test. It is now used for both.
1723 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1724 patch removes a few documentation additions to RFC 3028, because the
1725 latest draft now contains them. It adds the new en;ascii-case comparator
1726 and a new error check for 8bit text in MIME parts. Comparator and
1727 require names are now matched exactly. I enabled the subaddress
1728 extension, but it is not well tested yet (read: it works for me)."
1730 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1731 rework some of the code of TK/09 above to avoid the hardwired use of
1732 "%lld" and "long long". Replaced the call to snprintf() with a call to
1735 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1736 records point to non-existent hosts", "retry timeout exceeded", and
1737 "retry time not reached for any host after a long failure period".
1739 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1740 experimental DomainKeys support:
1742 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1743 (2) On an error such as an illegally used "control", the wrong name for
1744 the control was given.
1746 These problems did NOT occur unless DomainKeys support was compiled.
1748 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1750 PH/24 Added ${if match_ip condition.
1752 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1753 databases so that it will be absolutely obvious if a crash occurs in the
1754 DB library. This is a regular occurrence (often caused by mis-matched
1757 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1758 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1759 picked up on a user's system that detects such things. There doesn't seem
1760 to be a gcc warning option for this - only an attribute that has to be
1761 put on the function's prototype. It seems that in Fedora Core 4 they have
1762 set this on a number of new functions. No doubt there will be more in due
1765 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1766 specify a transport. However, if an address that was verified by such a
1767 router was the subject of a callout, Exim crashed because it tried to
1768 read the rcpt_include_affixes from the non-existent transport. Now it
1769 just assumes that the setting of that option is false. This bug was
1770 introduced by 4.51/PH/31.
1772 PH/28 Changed -d+all to exclude +memory, because that information is very
1773 rarely of interest, but it makes the output a lot bigger. People tend to
1774 do -d+all out of habit.
1776 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1777 code in os-type was giving problems when libc.so lives in lib64, like on
1780 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1781 aren't the modern standard, and it seems that some systems' include files
1782 don't always have them. Exim was already checking for some of the newer
1783 ones like T_AAAA, and defining it itself. I've added checks for all the
1784 record types that Exim uses.
1786 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1787 not automatically generating a new one, as it is supposed to. This
1788 prevented TLS from working. If the file did exist, but contained invalid
1789 data, a new version was generated, as expected. It was only the case of a
1790 non-existent file that was broken.
1792 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1793 with a change in libdomainkeys > 0.64.
1795 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1796 from DNS. If the selector record carries the flag, it now has
1797 precedence over the domain-wide flag.
1799 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1801 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1802 the use of a port name that isn't defined in /etc/services) occurred, the
1803 message was deferred as in a normal delivery, and thus remained on the
1804 spool, instead of being failed because of the mua_wrapper setting. This
1805 is now fixed, and I tidied up some of the mua_wrapper messages at the
1808 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1809 the messages in a hash of arrays rather than using individual hashes.
1810 This is a bit cleaner and results in dramatic memory savings, albeit
1811 at a slight CPU cost.
1813 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1814 as requested by Marc Sherman.
1816 SC/10 Eximstats: added histograms for user specified patterns as requested
1819 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1821 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1822 fopen() in the content-scanning modules that did not already have it.
1828 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1829 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1831 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1833 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1835 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1836 to test Sieve filters that use "vacation".
1838 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1839 that changes the way the GnuTLS parameters are stored in the cache file.
1840 The new format can be generated externally. For backward compatibility,
1841 if the data in the cache doesn't make sense, Exim assumes it has read an
1842 old-format file, and it generates new data and writes a new file. This
1843 means that you can't go back to an older release without removing the
1846 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1847 work if there are any delivery delays because "one_time" forces the
1848 parent to be marked "delivered", so its unseen clone is never tried
1849 again. For this reason, Exim now forbids the simultaneous setting of
1852 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1853 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1854 entry if you want to know the details. The fix, however, neglected to
1855 consider the case when local delivery batching is involved. The test for
1856 "previously delivered" was not happening when checking to see if an
1857 address could be batched with a previous (undelivered) one; under
1858 certain circumstances this could lead to multiple deliveries to the same
1861 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1862 in its include files, and this causes problems building Exim.
1864 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1865 header_syntax) but Exim was just ignoring anything given after a slash.
1866 In particular, this caused confusion with an attempt to use "verify =
1867 reverse_host_lookup/defer_ok". An error is now given when options are
1868 supplied for verify items that do not have them. (Maybe reverse_host_
1869 lookup should have a defer_ok option, but that's a different point.)
1871 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1872 defined by RFC 821) to 2048, because there were problems with some AUTH
1873 commands, and RFC 1869 says the size should be increased for extended
1874 SMTP commands that take arguments.
1876 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1879 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1880 "unknown" error; now it says that the functionality isn't in the binary.
1882 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1883 an address' error message when a string expansion fails (syntax or
1884 whatever). Otherwise the password may appear in the log. Following change
1885 PH/42 below, there is no longer a chance of it appearing in a bounce
1888 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1890 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1891 of its addresses was ever tried. (Bugzilla bug #2.)
1893 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1894 the result incorrectly in the debug output. (It correctly added a newline
1895 to what was transported.)
1897 TF/01 Added $received_time.
1899 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1900 commented out examples of how to interface to a virus scanner and to
1901 SpamAssassin. Also added commented examples of av_scanner and
1902 spamd_address settings.
1904 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1905 and controls are allowed in which ACLs. There were a couple of minor
1906 errors. Some of the entries in the conditions table (which is a table of
1907 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1908 negation of where the condition IS allowed.
1910 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1912 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1913 header file does not have a version number, so I've had to invent a new
1914 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1915 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1916 radiusclient), but it was contributed by a Radius user.
1918 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1919 files or directories, for OpenSSL.
1921 PH/19 When an Exim process that is running as root has to create an Exim log
1922 file, it does so in a subprocess that runs as exim:exim so as to get the
1923 ownership right at creation (otherwise, other Exim processes might see
1924 the file with the wrong ownership). There was no test for failure of this
1925 fork() call, which would lead to the process getting stuck as it waited
1926 for a non-existent subprocess. Forks do occasionally fail when resources
1927 run out. I reviewed all the other calls to fork(); they all seem to check
1930 PH/20 When checking for unexpected SMTP input at connect time (before writing
1931 the banner), Exim was not dealing correctly with a non-positive return
1932 from the read() function. If the client had disconnected by this time,
1933 the result was a log entry for a synchronization error with an empty
1934 string after "input=" when read() returned zero. If read() returned -1
1935 (an event I could not check), uninitialized data bytes were printed.
1936 There were reports of junk text (parts of files, etc) appearing after
1939 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1941 PH/22 Added support for macro redefinition, and (re)definition in between
1942 driver and ACL definitions.
1944 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1945 forgetting to use the resulting value; it was using the unexpanded value.
1947 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1948 hadn't been configured. The fix is from Juergen Kreileder, who
1949 understands it better than I do:
1951 "Here's what I see happening with three configured cyrus_sasl
1952 authenticators configured (plain, login, cram-md5):
1954 On startup auth_cyrus_sasl_init() gets called for each of these.
1955 This means three calls to sasl_listmech() without a specified mech_list.
1956 => SASL tests which mechs of all available mechs actually work
1957 => three warnings about OTP not working
1958 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1960 With the patch, sasl_listmech() also gets called three times. But now
1961 SASL's mech_list option is set to the server_mech specified in the the
1962 authenticator. Or in other words, the answer from sasl_listmech()
1963 gets limited to just the mech you're testing for (which is different
1965 => the return list contains just 'plain' or 'login', 'cram-md5' or
1966 nothing depending on the value of ob->server_mech.
1968 I've just tested the patch: Authentication still works fine,
1969 unavailable mechs specified in the exim configuration are still
1970 caught, and the auth.log warnings about OTP are gone."
1972 PH/25 When debugging is enabled, the contents of the command line are added
1973 to the debugging output, even when log_selector=+arguments is not
1976 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1977 answer is "GNU", and only if the return is "GNU/something" is the answer
1980 PH/27 $acl_verify_message is now set immediately after the failure of a
1981 verification in an ACL, and so is available in subsequent modifiers. In
1982 particular, the message can be preserved by coding like this:
1984 warn !verify = sender
1985 set acl_m0 = $acl_verify_message
1987 Previously, $acl_verify_message was set only while expanding "message"
1988 and "log_message" when a very denied access.
1990 PH/28 Modified OS/os.c-Linux with
1992 -#ifndef OS_LOAD_AVERAGE
1993 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1995 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1996 nomenclature these days.)
1998 PH/29 Installed patch from the Sieve maintainer that adds the options
1999 sieve_useraddress and sieve_subaddress to the redirect router.
2001 PH/30 In these circumstances:
2002 . Two addresses routed to the same list of hosts;
2003 . First host does not offer TLS;
2004 . First host accepts first address;
2005 . First host gives temporary error to second address;
2006 . Second host offers TLS and a TLS session is established;
2007 . Second host accepts second address.
2008 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2009 and peerdn, if requested) that were in fact used only for the second
2012 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2013 attention to any local part prefix or suffix that was matched by the
2014 router that accepted the address. It now behaves in the same way as it
2015 does for delivery: the affixes are removed from the local part unless
2016 rcpt_include_affixes is set on the transport.
2018 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2019 timeout during the DATA phase of an incoming message.
2021 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2022 applied a patch sanctioned by the Sieve maintainer.
2024 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2025 the uid or gid is negative. A case of a negative gid caused this to be
2026 noticed. The fix allows for either to be negative.
2028 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2029 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2032 PH/36 The change PH/12 above was broken. Fixed it.
2034 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2035 the grounds that routing the same address twice would always produce the
2036 same answer. This might have been true once, but it is certainly no
2037 longer true now. Routing a child address may depend on the previous
2038 routing that produced that child. Some complicated redirection strategies
2039 went wrong when messages had multiple recipients, and made Exim's
2040 behaviour dependent on the order in which the addresses were given.
2042 I have moved the duplicate checking until after the routing is complete.
2043 Exim scans the addresses that are assigned to local and remote
2044 transports, and removes any duplicates. This means that more work will be
2045 done, as duplicates will always all be routed, but duplicates are
2046 presumably rare, so I don't expect this is of any significance.
2048 For deliveries to pipes, files, and autoreplies, the duplicate checking
2049 still happens during the routing process, since they are not going to be
2052 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2053 It corrects a timeout issue with spamd. This is Ian's comment: "The
2054 background is that sometimes spamd either never reads data from a
2055 connection it has accepted, or it never writes response data. The exiscan
2056 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2057 blindly assumes that writes won't block so it may never time out."
2059 PH/39 Allow G after quota size as well as K and M.
2061 PH/40 The value set for $authenticated_id in an authenticator may not contain
2062 binary zeroes or newlines because the value is written to log lines and
2063 to spool files. There was no check on this. Now the value is run through
2064 the string_printing() function so that such characters are converted to
2065 printable escape sequences.
2067 PH/41 $message_linecount is a new variable that contains the total number of
2068 lines in the message. Compare $body_linecount, which is the count for the
2071 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2072 bounce and delay warning messages, except in certain special cases, which
2075 (a) An SMTP error message from a remote host;
2076 (b) A message specified in a :fail: redirection;
2077 (c) A message specified in a "fail" command in a system filter;
2078 (d) A message specified in a FAIL return from the queryprogram router;
2079 (e) A message specified by the cannot_route_message router option.
2081 In these cases only, Exim does include the error details in bounce and
2082 warning messages. There are also a few cases where bland messages such
2083 as "unrouteable address" or "local delivery error" are given.
2085 PH/43 $value is now also set for the "else" part of a ${run expansion.
2087 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2088 being worked on, but at least Exim now implements the latest version to
2091 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2092 process to complete was treated as a delivery failure, a timeout while
2093 writing the message to the pipe was logged, but erroneously treated as a
2094 successful delivery. Such timeouts include transport filter timeouts. For
2095 consistency with the overall process timeout, these timeouts are now
2096 treated as errors, giving rise to delivery failures by default. However,
2097 there is now a new Boolean option for the pipe transport called
2098 timeout_defer, which, if set TRUE, converts the failures into defers for
2099 both kinds of timeout. A transport filter timeout is now identified in
2102 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2103 systems where "make" and "gmake" are different, calling "gmake" at top
2104 level broke things. I've arranged for the value of $(MAKE) to be passed
2105 from the Makefile to this script so that it can call the same version of
2109 A note about Exim versions 4.44 and 4.50
2110 ----------------------------------------
2112 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2113 changes of various kinds. As a consequence, a big documentation update was
2114 needed. This delayed the release for rather longer than seemed good, especially
2115 in the light of a couple of (minor) security issues. Therefore, the changes
2116 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2117 release. So 4.44 and 4.50 are in effect two different branches that both start
2120 I have left the 4.50 change log unchanged; it contains all the changes since
2121 4.43. The change log for 4.44 is below; many of its items are identical to
2122 those for 4.50. This seems to be the most sensible way to preserve the
2123 historical information.
2129 1. Minor wording change to the doc/README.SIEVE file.
2131 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2132 computation of the current number of files was incorrect.
2134 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2135 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2138 4. Give more explanation in the error message when the command for a transport
2139 filter fails to execute.
2141 5. There are several places where Exim runs a non-Exim command in a
2142 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2143 was being done only for the command run by the queryprogram router. It is
2144 now done for all such subprocesses. The other cases are: ${run, transport
2145 filters, and the commands run by the lmtp and pipe transports.
2147 6. Added CONFIGURE_GROUP build-time option.
2149 7. Some older OS have a limit of 256 on the maximum number of file
2150 descriptors. Exim was using setrlimit() to set 1000 as a large value
2151 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2152 systems. I've change it so that if it can't get 1000, it tries for 256.
2154 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2155 was an oversight, and furthermore, ever since the addition of extra
2156 controls (e.g. 4.43/32), the checks on when to allow different forms of
2157 "control" were broken. There should now be diagnostics for all cases when a
2158 control that does not make sense is encountered.
2160 9. Added the /retain_sender option to "control=submission".
2162 10. $recipients is now available in the predata ACL (oversight).
2164 11. Tidy the search cache before the fork to do a delivery from a message
2165 received from the command line. Otherwise the child will trigger a lookup
2166 failure and thereby defer the delivery if it tries to use (for example) a
2167 cached ldap connection that the parent has called unbind on.
2169 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2170 of $address_data from the recipient verification was clobbered by the
2171 sender verification.
2173 13. The value of address_data from a sender verification is now available in
2174 $sender_address_data in subsequent conditions in the ACL statement.
2176 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2178 15. Added a new option "connect=<time>" to callout options, to set a different
2181 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2182 was its contents. (It was OK if the option was not defined at all.)
2184 17. A "Completed" log line is now written for messages that are removed from
2185 the spool by the -Mrm option.
2187 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2188 information about exactly what failed.
2190 19. Added -dd to debug only the daemon process.
2192 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2193 handles timeouts, both on the server side and network timeouts. Renamed the
2194 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2196 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2197 It is now set to "smtps".
2199 22. $host_address is now set to the target address during the checking of
2200 ignore_target_hosts.
2202 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2203 being passed; this would have caused $sender_host_name to have been used if
2204 matching the list had actually called for a host name (not very likely,
2205 since this list is usually IP addresses). A host name is now passed as
2208 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2209 code that specifies a non-restarting handler (typically sigaction() in
2210 modern systems) in an attempt to fix a rare and obscure crash bug.
2212 25. Narrowed the window for a race in the daemon that could cause it to ignore
2213 SIGCHLD signals. This is not a major problem, because they are used only to
2214 wake it up if nothing else does.
2216 26. A malformed maildirsize file could cause Exim to calculate negative values
2217 for the mailbox size or file count. Odd effects could occur as a result.
2218 The maildirsize information is now recalculated if the size or filecount
2221 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2222 support for a long time. Removed HAVE_SYS_VFS_H.
2224 28. Installed the latest version of exipick from John Jetmore.
2226 29. In an address list, if the pattern was not a regular expression, an empty
2227 subject address (from a bounce message) matched only if the pattern was an
2228 empty string. Non-empty patterns were not even tested. This was the wrong
2229 because it is perfectly reasonable to use an empty address as part of a
2230 database query. An empty address is now tested by patterns that are
2231 lookups. However, all the other forms of pattern expect the subject to
2232 contain a local part and a domain, and therefore, for them, an empty
2233 address still always fails if the pattern is not itself empty.
2235 30. Exim went into a mad DNS loop when attempting to do a callout where the
2236 host was specified on an smtp transport, and looking it up yielded more
2237 than one IP address.
2239 31. Re-factored the code for checking spool and log partition space into a
2240 function that finds that data and another that does the check. The former
2241 is then used to implement four new variables: $spool_space, $log_space,
2242 $spool_inodes, and $log_inodes.
2244 32. The RFC2047 encoding function was originally intended for short strings
2245 such as real names; it was not keeping to the 75-character limit for
2246 encoded words that the RFC imposes. It now respects the limit, and
2247 generates multiple encoded words if necessary. To be on the safe side, I
2248 have increased the buffer size for the ${rfc2047: expansion operator from
2251 33. It is now permitted to omit both strings after an "if" condition; if the
2252 condition is true, the result is "true". As before, when the second string
2253 is omitted, a false condition yields an empty string. This makes it less
2254 cumbersome to write custom ACL and router conditions.
2256 34. Failure to deliver a bounce message always caused it to be frozen, even if
2257 there was an errors_to setting on the router. The errors_to setting is now
2260 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2261 canonical form (fully expanded) before being placed in
2262 $sender_host_address.
2264 36. The table in the code that translates DNS record types into text (T_A to
2265 "A" for instance) was missing entries for NS and CNAME. It is just possible
2266 that this could have caused confusion if both these types were looked up
2267 for the same domain, because the text type is used as part of Exim's
2268 per-process caching. But the chance of anyone hitting this buglet seems
2271 37. The dnsdb lookup has been extended in a number of ways.
2273 (1) There is a new type, "zns", which walks up the domain tree until it
2274 finds some nameserver records. It should be used with care.
2276 (2) There is a new type, "mxh", which is like "mx" except that it returns
2277 just the host names, not the priorities.
2279 (3) It is now possible to give a list of domains (or IP addresses) to be
2280 looked up. The behaviour when one of the lookups defers can be
2281 controlled by a keyword.
2283 (4) It is now possible to specify the separator character for use when
2284 multiple records are returned.
2286 38. The dnslists ACL condition has been extended: it is now possible to supply
2287 a list of IP addresses and/or domains to be looked up in a particular DNS
2290 39. Added log_selector=+queue_time_overall.
2292 40. When running the queue in the test harness, wait just a tad after forking a
2293 delivery process, to get repeatability of debugging output.
2295 41. Include certificate and key file names in error message when GnuTLS fails
2296 to set them up, because the GnuTLS error message doesn't include the name
2297 of the failing file when there is a problem reading it.
2299 42. Allow both -bf and -bF in the same test run.
2301 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2303 44. The "Exiscan patch" is now merged into the mainline Exim source.
2305 45. Sometimes the final signoff response after QUIT could fail to get
2306 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2307 < 0 before doing a fflush(). This bug looks as though it goes back to the
2308 introduction of TLS in release 3.20, but "sometimes" must have been rare
2309 because the tests only now provoked it.
2311 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2312 (this can affect the format of dates).
2314 47. exim_tidydb, when checking for the continued existence of a message for
2315 which it has found a message-specific retry record, was not finding
2316 messages that were in split spool directories. Consequently, it was
2317 deleting retry records that should have stayed in existence.
2319 48. Steve fixed some bugs in eximstats.
2321 49. The SPA authentication driver was not abandoning authentication and moving
2322 on to the next authenticator when an expansion was forced to fail,
2323 contradicting the general specification for all authenticators. Instead it
2324 was generating a temporary error. It now behaves as specified.
2326 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2327 (the order specifies the preference for clients). The order is now AES256,
2328 AES128, 3DES, ARCFOUR128.
2330 51. Small patch to Sieve code - explicitly set From: when generating an
2333 52. Exim crashed if a remote delivery caused a very long error message to be
2334 recorded - for instance if somebody sent an entire SpamAssassin report back
2335 as a large number of 550 error lines. This bug was coincidentally fixed by
2336 increasing the size of one of Exim's internal buffers (big_buffer) that
2337 happened as part of the Exiscan merge. However, to be on the safe side, I
2338 have made the code more robust (and fixed the comments that describe what
2341 53. Now that there can be additional text after "Completed" in log lines (if
2342 the queue_time_overall log selector is set), a one-byte patch to exigrep
2343 was needed to allow it to recognize "Completed" as not the last thing in
2346 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2347 patch that reportedly fixes this has been added. I am not expert enough to
2348 create a test for it. This is what the patch creator wrote:
2350 "I found a little strange behaviour of ldap code when working with
2351 Windows 2003 AD Domain, where users was placed in more than one
2352 Organization Units. When I tried to give exim partial DN, the exit code
2353 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2354 But simultaneously result of request was absolutely normal ldap result,
2355 so I produce this patch..."
2357 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2358 REFERENCE, so I have modified the code to exclude the patch when that macro
2361 55. Some experimental protocols are using DNS PTR records for new purposes. The
2362 keys for these records are domain names, not reversed IP addresses. The
2363 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2364 leaves it alone. Component reversal etc. now happens only for IP addresses.
2367 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2369 57. Double the size of the debug message buffer (to 2048) so that more of very
2370 long debug lines gets shown.
2372 58. The exicyclog utility now does better if the number of log files to keep
2373 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2375 59. Two changes related to the smtp_active_hostname option:
2377 (1) $smtp_active_hostname is now available as a variable.
2378 (2) The default for smtp_banner uses $smtp_active_hostname instead
2379 of $primary_hostname.
2381 60. The host_aton() function is supposed to be passed a string that is known
2382 to be a valid IP address. However, in the case of IPv6 addresses, it was
2383 not checking this. This is a hostage to fortune. Exim now panics and dies
2384 if the condition is not met. A case was found where this could be provoked
2385 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2386 components; fortuitously, this particular loophole had already been fixed
2387 by change 4.50/55 above.
2389 If there are any other similar loopholes, the new check in host_aton()
2390 itself should stop them being exploited. The report I received stated that
2391 data on the command line could provoke the exploit when Exim was running as
2392 exim, but did not say which command line option was involved. All I could
2393 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2394 running as the user.
2397 61. There was a buffer overflow vulnerability in the SPA authentication code
2398 (which came originally from the Samba project). I have added a test to the
2399 spa_base64_to_bits() function which I hope fixes it.
2402 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2403 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2405 63. The daemon start-up calls getloadavg() while still root for those OS that
2406 need the first call to be done as root, but it missed one case: when
2407 deliver_queue_load_max is set with deliver_drop_privilege. This is
2408 necessary for the benefit of the queue runner, because there is no re-exec
2409 when deliver_drop_privilege is set.
2411 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2412 This has been fixed.
2414 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2415 was in use, was not putting the data itself into the right store pool;
2416 consequently, it could be overwritten for a subsequent message in the same
2417 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2420 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2422 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2423 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2424 the function were treating the return as a boolean value, which happened to
2425 work because 0=false and not-0=true, but is not correct code.
2427 68. The host_aton() function was not handling scoped IPv6 addresses (those
2428 with, for example, "%eth0" on the end) correctly.
2430 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2431 negated items (that is, ~something) in unsigned ints. Some compilers
2432 apparently mutter when there is no cast.
2434 70. If an address verification called from an ACL failed, and did not produce a
2435 user-specific message (i.e. there was only a "system" message), nothing was
2436 put in $acl_verify_message. In this situation, it now puts the system
2439 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2440 session; change 4.31/43 added the unwanted input to the log line - except
2441 that it did not do this in the start of session case. It now does.
2443 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2444 This is wrong and can cause the other end to generate a synchronization
2445 error if it is another Exim or anything else that does the synchronization
2446 check. A QUIT command is no longer sent after a timeout.
2448 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2449 during host lookups.
2451 74. The defer_ok option of callout verification was not working if it was used
2452 when verifying addresses in header lines, that is, for this case:
2454 verify = header_sender/callout=defer_ok
2456 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2457 those file descriptors could be used for SMTP connections. If anything
2458 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2459 could be sent to the SMTP client, causing chaos. The daemon now opens
2460 stdin, stdout, and stderr to /dev/null when it puts itself into the
2463 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2464 log by default. The user can override this with suitable Perl magic.
2466 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2467 the log message when discard triggers, was not working for the DATA ACL or
2468 for the non-SMTP ACL.
2470 78. Error message wording change in sieve.c.
2472 79. If smtp_accept_max_per_host was set, the number of connections could be
2473 restricted to fewer than expected, because the daemon was trying to set up
2474 a new connection before checking whether the processes handling previous
2475 connections had finished. The check for completed processes is now done
2476 earlier. On busy systems, this bug wouldn't be noticed because something
2477 else would have woken the daemon, and it would have reaped the completed
2480 80. If a message was submitted locally by a user whose login name contained one
2481 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2482 It caused a spool format error. I have fixed the spool reading code. A
2483 related problem was that the "from" clause in the Received: line became
2484 illegal because of the space(s). It is now covered by ${quote_local_part.
2486 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2489 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2491 83. Patch from Sieve maintainer to fix unterminated string problem in
2492 "vacation" handling.
2494 84. Some minor changes to the Linux configuration files to help with other
2495 OS variants using glibc.
2497 85. One more patch for Sieve to update vacation handling to latest spec.
2500 ----------------------------------------------------
2501 See the note above about the 4.44 and 4.50 releases.
2502 ----------------------------------------------------
2508 1. Change 4.43/35 introduced a bug that caused file counts to be
2509 incorrectly computed when quota_filecount was set in an appendfile
2512 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2513 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2516 3. Give more explanation in the error message when the command for a transport
2517 filter fails to execute.
2519 4. There are several places where Exim runs a non-Exim command in a
2520 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2521 was being done only for the command run by the queryprogram router. It is
2522 now done for all such subprocesses. The other cases are: ${run, transport
2523 filters, and the commands run by the lmtp and pipe transports.
2525 5. Some older OS have a limit of 256 on the maximum number of file
2526 descriptors. Exim was using setrlimit() to set 1000 as a large value
2527 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2528 systems. I've change it so that if it can't get 1000, it tries for 256.
2530 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2531 was an oversight, and furthermore, ever since the addition of extra
2532 controls (e.g. 4.43/32), the checks on when to allow different forms of
2533 "control" were broken. There should now be diagnostics for all cases when a
2534 control that does not make sense is encountered.
2536 7. $recipients is now available in the predata ACL (oversight).
2538 8. Tidy the search cache before the fork to do a delivery from a message
2539 received from the command line. Otherwise the child will trigger a lookup
2540 failure and thereby defer the delivery if it tries to use (for example) a
2541 cached ldap connection that the parent has called unbind on.
2543 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2544 of $address_data from the recipient verification was clobbered by the
2545 sender verification.
2547 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2548 was its contents. (It was OK if the option was not defined at all.)
2550 11. A "Completed" log line is now written for messages that are removed from
2551 the spool by the -Mrm option.
2553 12. $host_address is now set to the target address during the checking of
2554 ignore_target_hosts.
2556 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2557 being passed; this would have caused $sender_host_name to have been used if
2558 matching the list had actually called for a host name (not very likely,
2559 since this list is usually IP addresses). A host name is now passed as
2562 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2563 code that specifies a non-restarting handler (typically sigaction() in
2564 modern systems) in an attempt to fix a rare and obscure crash bug.
2566 15. Narrowed the window for a race in the daemon that could cause it to ignore
2567 SIGCHLD signals. This is not a major problem, because they are used only to
2568 wake it up if nothing else does.
2570 16. A malformed maildirsize file could cause Exim to calculate negative values
2571 for the mailbox size or file count. Odd effects could occur as a result.
2572 The maildirsize information is now recalculated if the size or filecount
2575 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2576 support for a long time. Removed HAVE_SYS_VFS_H.
2578 18. Updated exipick to current release from John Jetmore.
2580 19. Allow an empty sender to be matched against a lookup in an address list.
2581 Previously the only cases considered were a regular expression, or an
2584 20. Exim went into a mad DNS lookup loop when doing a callout where the
2585 host was specified on the transport, if the DNS lookup yielded more than
2588 21. The RFC2047 encoding function was originally intended for short strings
2589 such as real names; it was not keeping to the 75-character limit for
2590 encoded words that the RFC imposes. It now respects the limit, and
2591 generates multiple encoded words if necessary. To be on the safe side, I
2592 have increased the buffer size for the ${rfc2047: expansion operator from
2595 22. Failure to deliver a bounce message always caused it to be frozen, even if
2596 there was an errors_to setting on the router. The errors_to setting is now
2599 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2600 canonical form (fully expanded) before being placed in
2601 $sender_host_address.
2603 24. Updated eximstats to version 1.33
2605 25. Include certificate and key file names in error message when GnuTLS fails
2606 to set them up, because the GnuTLS error message doesn't include the name
2607 of the failing file when there is a problem reading it.
2609 26. Expand error message when OpenSSL has problems setting up cert/key files.
2612 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2613 (this can affect the format of dates).
2615 28. exim_tidydb, when checking for the continued existence of a message for
2616 which it has found a message-specific retry record, was not finding
2617 messages that were in split spool directories. Consequently, it was
2618 deleting retry records that should have stayed in existence.
2620 29. eximstats updated to version 1.35
2621 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2622 1.35 - bugfix such that pie charts by volume are generated correctly
2624 30. The SPA authentication driver was not abandoning authentication and moving
2625 on to the next authenticator when an expansion was forced to fail,
2626 contradicting the general specification for all authenticators. Instead it
2627 was generating a temporary error. It now behaves as specified.
2629 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2630 (the order specifies the preference for clients). The order is now AES256,
2631 AES128, 3DES, ARCFOUR128.
2633 31. Small patch to Sieve code - explicitly set From: when generating an
2636 32. Exim crashed if a remote delivery caused a very long error message to be
2637 recorded - for instance if somebody sent an entire SpamAssassin report back
2638 as a large number of 550 error lines. This bug was coincidentally fixed by
2639 increasing the size of one of Exim's internal buffers (big_buffer) that
2640 happened as part of the Exiscan merge. However, to be on the safe side, I
2641 have made the code more robust (and fixed the comments that describe what
2644 33. Some experimental protocols are using DNS PTR records for new purposes. The
2645 keys for these records are domain names, not reversed IP addresses. The
2646 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2647 leaves it alone. Component reversal etc. now happens only for IP addresses.
2650 34. The host_aton() function is supposed to be passed a string that is known
2651 to be a valid IP address. However, in the case of IPv6 addresses, it was
2652 not checking this. This is a hostage to fortune. Exim now panics and dies
2653 if the condition is not met. A case was found where this could be provoked
2654 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2655 components; fortuitously, this particular loophole had already been fixed
2656 by change 4.50/55 or 4.44/33 above.
2658 If there are any other similar loopholes, the new check in host_aton()
2659 itself should stop them being exploited. The report I received stated that
2660 data on the command line could provoke the exploit when Exim was running as
2661 exim, but did not say which command line option was involved. All I could
2662 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2663 running as the user.
2666 35. There was a buffer overflow vulnerability in the SPA authentication code
2667 (which came originally from the Samba project). I have added a test to the
2668 spa_base64_to_bits() function which I hope fixes it.
2671 36. The daemon start-up calls getloadavg() while still root for those OS that
2672 need the first call to be done as root, but it missed one case: when
2673 deliver_queue_load_max is set with deliver_drop_privilege. This is
2674 necessary for the benefit of the queue runner, because there is no re-exec
2675 when deliver_drop_privilege is set.
2677 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2678 was in use, was not putting the data itself into the right store pool;
2679 consequently, it could be overwritten for a subsequent message in the same
2680 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2683 38. Sometimes the final signoff response after QUIT could fail to get
2684 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2685 < 0 before doing a fflush(). This bug looks as though it goes back to the
2686 introduction of TLS in release 3.20, but "sometimes" must have been rare
2687 because the tests only now provoked it.
2693 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2694 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2695 Now it returns an integer. A number of calls were still expecting a T/F
2696 return. Fortuitously, in all cases, the tests worked in OK situations,
2697 which is the norm. However, things would have gone wrong on any write
2698 failures on the smtp file descriptor. This function is used when sending
2699 messages over SMTP and also when doing verify callouts.
2701 2. When Exim is called to do synchronous delivery of a locally submitted
2702 message (the -odf or -odi options), it no longer closes stderr before doing
2705 3. Implemented the mua_wrapper option.
2707 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2709 5. Implemented the functions header_remove(), header_testname(),
2710 header_add_at_position(), and receive_remove_recipient(), and exported them
2713 6. If an ACL "warn" statement specified the addition of headers, Exim already
2714 inserted X-ACL-Warn: at the start if there was no header name. However, it
2715 was not making this test for the second and subsequent header lines if
2716 there were newlines in the string. This meant that an invalid header could
2717 be inserted if Exim was badly configured.
2719 7. Allow an ACL "warn" statement to add header lines at the start or after all
2720 the Received: headers, as well as at the end.
2722 8. Added the rcpt_4xx retry error code.
2724 9. Added postmaster_mailfrom=xxx to callout verification option.
2726 10. Added mailfrom=xxxx to the callout verification option, for verify=
2729 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2730 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2732 12. Inserted some casts to stop certain compilers warning when using pointer
2733 differences as field lengths or precisions in printf-type calls (mostly
2734 affecting debugging statements).
2736 13. Added optional readline() support for -be (dynamically loaded).
2738 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2739 same clock tick as a message's arrival, so that its received time was the
2740 same as the "first fail" time on the retry record, and that message
2741 remained on the queue past the ultimate address timeout, every queue runner
2742 would try a delivery (because it was past the ultimate address timeout) but
2743 after another failure, the ultimate address timeout, which should have then
2744 bounced the address, did not kick in. This was a "< instead of <=" error;
2745 in most cases the first failure would have been in the next clock tick
2746 after the received time, and all would be well.
2748 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2749 being recognized when the domain list was tested by the match_domain
2750 condition in an expansion string.
2752 16. Added the ${str2b64: operator.
2754 17. Exim was always calling setrlimit() to set a large limit for the number of
2755 processes, without checking whether the existing limit was already
2756 adequate. (It did check for the limit on file descriptors.) Furthermore,
2757 errors from getrlimit() and setrlimit() were being ignored. Now they are
2758 logged to the main and panic logs, but Exim does carry on, to try to do its
2759 job under whatever limits there are.
2761 18. Imported PCRE 5.0.
2763 19. Trivial typo in log message " temporarily refused connection" (the leading
2766 20. If the log selector return_path_on_delivery was set and an address was
2767 redirected to /dev/null, the delivery process crashed because it assumed
2768 that a return path would always be set for a "successful" delivery. In this
2769 case, the whole delivery is bypassed as an optimization, and therefore no
2772 21. Internal re-arrangement: the function for sending a challenge and reading
2773 a response while authentication was assuming a zero-terminated challenge
2774 string. It's now changed to take a pointer and a length, to allow for
2775 binary data in such strings.
2777 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2779 23. Exim was not respecting finduser_retries when seeking the login of the
2780 uid under which it was called; it was always trying 10 times. (The default
2781 setting of finduser_retries is zero.) Also, it was sleeping after the final
2782 failure, which is pointless.
2784 24. Implemented tls_on_connect_ports.
2786 25. Implemented acl_smtp_predata.
2788 26. If the domain in control=submission is set empty, Exim assumes that the
2789 authenticated id is a complete email address when it generates From: or
2790 Sender: header lines.
2792 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2793 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2794 chown and chgrp in /bin and hostname in /usr/bin.
2796 28. Exim was keeping the "process log" file open after each use, just as it
2797 does for the main log. This opens the possibility of it remaining open for
2798 long periods when the USR1 signal hits a daemon. Occasional processlog
2799 errors were reported, that could have been caused by this. Anyway, it seems
2800 much more sensible not to leave this file open at all, so that is what now
2803 29. The long-running daemon process does not normally write to the log once it
2804 has entered its main loop, and it closes the log before doing so. This is
2805 so that log files can straightforwardly be renamed and moved. However,
2806 there are a couple of unusual error situations where the daemon does write
2807 log entries, and I had neglected to close the log afterwards.
2809 30. The text of an SMTP error response that was received during a remote
2810 delivery was being truncated at 512 bytes. This is too short for some of
2811 the long messages that one sometimes sees. I've increased the limit to
2814 31. It is now possible to make retry rules that apply only when a message has a
2815 specific sender, in particular, an empty sender.
2817 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2818 it possible to be selective about when SMTP synchronization is enforced.
2820 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2822 32. Implemented hosts_connection_nolog.
2824 33. Added an ACL for QUIT.
2826 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2829 35. Added mailbox_size and mailbox_filecount to appendfile.
2831 36. Added control = no_multiline_responses to ACLs.
2833 37. There was a bug in the logic of the code that waits for the clock to tick
2834 in the case where the clock went backwards by a substantial amount such
2835 that the microsecond fraction of "now" was more than the microsecond
2836 fraction of "then" (but the whole seconds number was less).
2838 38. Added support for the libradius Radius client library this is found on
2839 FreeBSD (previously only the radiusclient library was supported).
2845 1. When certain lookups returned multiple values in the form name=value, the
2846 quoting of the values was not always being done properly. Specifically:
2847 (a) If the value started with a double quote, but contained no whitespace,
2849 (b) If the value contained whitespace other than a space character (i.e.
2850 tabs or newlines or carriage returns) it was not quoted.
2851 This fix has been applied to the mysql and pgsql lookups by writing a
2852 separate quoting function and calling it from the lookup code. The fix
2853 should probably also be applied to nisplus, ibase and oracle lookups, but
2854 since I cannot test any of those, I have not disturbed their existing code.
2856 2. A hit in the callout cache for a specific address caused a log line with no
2857 reason for rejecting RCPT. Now it says "Previous (cached) callout
2858 verification failure".
2860 3. There was an off-by-one bug in the queryprogram router. An over-long
2861 return line was truncated at 256 instead of 255 characters, thereby
2862 overflowing its buffer with the terminating zero. As well as fixing this, I
2863 have increased the buffer size to 1024 (and made a note to document this).
2865 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2866 when Exim is waiting for an SMTP response from a remote server, Exim
2867 restarts its select() call on the socket, thereby resetting its timeout.
2868 This is not a problem when such interrupts are rare. Somebody set up a cron
2869 job to run exiwhat every 2 minutes, which is less than the normal select()
2870 timeout (5 or 10 minutes). This meant that the select() timeout never
2871 kicked in because it was always reset. I have fixed this by comparing the
2872 time when an interrupt arrives with the time at the start of the first call
2873 to select(). If more time than the timeout has elapsed, the interrupt is
2874 treated as a timeout.
2876 5. Some internal re-factoring in preparation for the addition of Sieve
2877 extensions (by MH). In particular, the "personal" test is moved to a
2878 separate function, and given an option for scanning Cc: and Bcc: (which is
2879 not set for Exim filters).
2881 6. When Exim created an email address using the login of the caller as the
2882 local part (e.g. when creating a From: or Sender: header line), it was not
2883 quoting the local part when it contained special characters such as @.
2885 7. Installed new OpenBSD configuration files.
2887 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2888 try to make them clearer.
2890 9. Callout options, other than the timeout value, were being ignored when
2891 verifying sender addresses in header lines. For example, when using
2893 verify = header_sender/callout=no_cache
2895 the cache was (incorrectly) being used.
2897 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2898 only the Cygwin environment.
2900 11. When return_path_on_delivery was set as a log selector, if different remote
2901 addresses in the same message used different return paths and parallel
2902 remote delivery occurred, the wrong values would sometimes be logged.
2903 (Whenever a remote delivery process finished, the return path value from
2904 the most recently started remote delivery process was logged.)
2906 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2907 lines when AUTH and/or TLS are in use. This is the "received protocol"
2908 field. Exim used to use "asmtp" for authenticated SMTP, without any
2909 indication (in the protocol name) for TLS use. Now it follows the RFC and
2910 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2911 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2912 names appear in log lines as well as in Received: header lines.
2914 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2915 extensions, and comparison tests, and to fix some bugs.
2917 14. Changes to the "personal" filter test:
2919 (1) The test was buggy in that it was just doing the equivalent of
2920 "contains" tests on header lines. For example, if a user's address was
2921 anne@some.where, the "personal" test would incorrectly be true for
2923 To: susanne@some.where
2925 This test is now done by extracting each address from the header in turn,
2926 and checking the entire address. Other tests that are part of "personal"
2927 are now done using regular expressions (for example, to check local parts
2928 of addresses in From: header lines).
2930 (2) The list of non-personal local parts in From: addresses has been
2931 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2932 taken from the Sieve specification recommendations.
2934 (3) If the message contains any header line starting with "List-" it is
2935 treated as non-personal.
2937 (4) The test for "circular" in the Subject: header line has been removed
2938 because it now seems ill-conceived.
2940 15. Minor typos in src/EDITME comments corrected.
2942 16. Installed latest exipick from John Jetmore.
2944 17. If headers_add on a router specified a text string that was too long for
2945 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2946 of string_sprintf() is now avoided.
2948 18. $message_body_size was not set (it was always zero) when running the DATA
2949 ACL and the local_scan() function.
2951 19. For the "mail" command in an Exim filter, no default was being set for
2952 the once_repeat time, causing a random time value to be used if "once" was
2953 specified. (If the value happened to be <= 0, no repeat happened.) The
2954 default is now 0s, meaning "never repeat". The "vacation" command was OK
2955 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2956 (I found it when inspecting the code).
2958 20. There is now an overall timeout for performing a callout verification. It
2959 defaults to 4 times the callout timeout, which applies to individual SMTP
2960 commands during the callout. The overall timeout applies when there is more
2961 than one host that can be tried. The timeout is checked before trying the
2962 next host. This prevents very long delays if there are a large number of
2963 hosts and all are timing out (e.g. when the network connections are timing
2964 out). The value of the overall timeout can be changed by specifying an
2965 additional sub-option for "callout", called "maxwait". For example:
2967 verify = sender/callout=5s,maxwait=20s
2969 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2970 to the end before writing, but this should make it even safer).
2972 22. Exim was forgetting that it had advertised PIPELINING for the second and
2973 subsequent messages on an SMTP connection. It was also not resetting its
2974 memory on STARTTLS and an internal HELO.
2976 23. When Exim logs an SMTP synchronization error within a session, it now
2977 records whether PIPELINING has been advertised or not.
2979 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2980 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2981 rather than long int.
2983 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2985 26. Added the never_mail option to autoreply.
2991 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2992 crash if the getsockname() call failed; this can happen if a connection is
2993 closed very soon after it is established. The problem was simply in the
2994 order in which certain operations were done, causing Exim to try to write
2995 to the SMTP stream before it had set up the file descriptor. The bug has
2996 been fixed by making things happen in the correct order.
3002 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3003 before the connection was closed, thus losing the rejection response.
3005 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3006 some early Solaris releases, but causes trouble in current releases where
3007 socklen_t is defined.
3009 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3012 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3015 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3016 that releasing the top of it at the end releases what was used for sub-
3017 expansions (unless the block got too big). However, discard this block if
3018 the first thing is a variable or header, so that we can use its block when
3019 it is dynamic (useful for very large $message_headers, for example).
3021 6. Lookups now cache *every* query, not just the most recent. A new, separate
3022 store pool is used for this. It can be recovered when all lookup caches are
3023 flushed. Lookups now release memory at the end of their result strings.
3024 This has involved some general refactoring of the lookup sources.
3026 7. Some code has been added to the store_xxx() functions to reduce the amount
3027 of flapping under certain conditions.
3029 8. log_incoming_interface used to affect only the <= reception log lines. Now
3030 it causes the local interface and port to be added to several more SMTP log
3031 lines, for example "SMTP connection from", and rejection lines.
3033 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3035 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3037 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3038 could be overwritten at the end of the current message (or the start of a
3039 new message if it was set in a HELO ACL). The value is now preserved for
3040 the duration of the SMTP connection.
3042 12. If a transport had a headers_rewrite setting, and a matching header line
3043 contained an unqualified address, that address was qualified, even if it
3044 did not match any rewriting rules. The underlying bug was that the values
3045 of the flags that permit the existence of unqualified sender and recipient
3046 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3047 non-local messages, and by -bnq for local messages) were not being
3048 preserved with the message after it was received.
3050 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3051 "next input=" as part of the text comprising the host identity instead of
3052 the correct text. The code was using the same buffer for two different
3053 strings. However, depending on which order the printing function evaluated
3054 its arguments, the bug did not always show up. Under Linux, for example, my
3055 test suite worked just fine.
3057 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3058 doesn't work with some older versions of Perl. It has been changed to "my",
3059 which in any case is probably the better facility to use.
3061 15. A really picky compiler found some instances of statements for creating
3062 error messages that either had too many or two few arguments for the format
3065 16. The size of the buffer for calls to the DNS resolver has been increased
3066 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3067 for addresses that have a lot of PTR records. This alleviates a problem; it
3068 does not fully solve it.
3070 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3071 buffer now truncates the list and logs the incident, which is the same
3072 action as happens when Exim is looking up a host name and its aliases.
3073 Previously in this situation something unpredictable would happen;
3074 sometimes it was "internal error: store_reset failed".
3076 18. If a server dropped the connection unexpectedly when an Exim client was
3077 using GnuTLS and trying to read a response, the client delivery process
3078 crashed while trying to generate an error log message.
3080 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3081 string, for example:
3083 warn message = H1: something\nH2: something
3085 the text was added as a single header line from Exim's point of view
3086 though it ended up OK in the delivered message. However, searching for the
3087 second and subsequent header lines using $h_h2: did not work. This has been
3088 fixed. Similarly, if a system filter added multiple headers in this way,
3089 the routers could not see them.
3091 20. Expanded the error message when iplsearch is called with an invalid key to
3092 suggest using net-iplsearch in a host list.
3094 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3095 ACLs are no longer actually imposed (and a message to that effect is
3098 22. If a "gecos" field in a passwd entry contained escaped characters, in
3099 particular, if it contained a \" sequence, Exim got it wrong when building
3100 a From: or a Sender: header from that name. A second bug also caused
3101 incorrect handling when an unquoted " was present following a character
3102 that needed quoting.
3104 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3105 was not being matched caselessly.
3107 24. Arranged for all hyphens in the exim.8 source to be escaped with
3110 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3111 properly. Recipient callout cache records were still being keyed to include
3112 the sender, even when use_sender was set false. This led to far more
3113 callouts that were necessary. The sender is no longer included in the key
3114 when use_sender is false.
3116 26. Added "control = submission" modifier to ACLs.
3118 27. Added the ${base62d: operator to decode base 62 numbers.
3120 28. dnsdb lookups can now access SRV records.
3122 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3123 the configuration file.
3125 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3126 (-v) mode. This makes the output for a verbose queue run more intelligible.
3128 31. Added a use_postmaster feature to recipient callouts.
3130 32. Added the $body_zerocount variable, containing the number of binary zero
3131 bytes in the message body.
3133 33. The time of last modification of the "new" subdirectory is now used as the
3134 "mailbox time last read" when there is a quota error for a maildir
3137 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3139 35. Added +ignore_unknown as a special item in host lists.
3141 36. Code for decoding IPv6 addresses in host lists is now included, even if
3142 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3143 address was recognized as an IP address, but was then not correctly decoded
3144 into binary, causing unexpected and incorrect effects when compared with
3151 1. Very minor rewording of debugging text in manualroute to say "list of
3152 hosts" instead of "hostlist".
3154 2. If verify=header_syntax was set, and a header line with an unqualified
3155 address (no domain) and a large number of spaces between the end of the
3156 name and the colon was received, the reception process suffered a buffer
3157 overflow, and (when I tested it) crashed. This was caused by some obsolete
3158 code that should have been removed. The fix is to remove it!
3160 3. When running in the test harness, delay a bit after writing a bounce
3161 message to get a bit more predictability in the log output.
3163 4. Added a call to search_tidyup() just before forking a reception process. In
3164 theory, someone could use a lookup in the expansion of smtp_accept_max_
3165 per_host which, without the tidyup, could leave open a database connection.
3167 5. Added the variables $recipient_data and $sender_data which get set from a
3168 lookup success in an ACL "recipients" or "senders" condition, or a router
3169 "senders" option, similar to $domain_data and $local_part_data.
3171 6. Moved the writing of debug_print from before to after the "senders" test
3174 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3175 problems for message scanning, either using a data ACL, or using
3176 local_scan() because the Received: header was not generated till after they
3177 were called (in order to set the time as the time of reception completion).
3178 I have revised the way this works. The header is now generated after the
3179 body is received, but before the ACL or local_scan() are called. After they
3180 are run, the timestamp in the header is updated.
3186 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3187 before starting a queue runner without re-exec. This happened only when
3188 deliver_drop_privilege was set or when the Exim user was set to root. The
3189 effect of the bug was that timeouts during subsequent deliveries caused
3190 crashes instead of being properly handled. The handler is now left at its
3191 default (and expected) setting.
3193 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3194 message, again when deliver_drop_privilege is set or Exim is run as root.
3195 The bug described in (1) was not present in this case, but the tidying up
3196 of the other signals was missing. I have made the two cases consistent.
3198 3. The ignore_target_hosts setting on a manualroute router was being ignored
3199 for hosts that were looked up using the /MX notation.
3201 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3204 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3205 operated on the sender address. After changing the $sender_address to <>
3206 for the sender address verify, Exim was re-instated it as the original
3207 (before rewriting) address, but remembering that it had rewritten it, so it
3208 wasn't rewriting it again. This bug also had the effect of breaking the
3209 sender address verification caching when the sender address was rewritten.
3211 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3212 This has been changed so that if the ip literal address matches
3213 ignore_target_hosts, the router declines.
3215 7. Added expansion conditions match_domain, match_address, and match_local_
3216 part (NOT match_host).
3218 8. The placeholder for the Received: header didn't have a length field set.
3220 9. Added code to Exim itself and to exim_lock to test for a specific race
3221 condition that could lead to file corruption when using MBX delivery. The
3222 issue is with the lockfile that is created in /tmp. If this file is removed
3223 after a process has opened it but before that process has acquired a lock,
3224 there is the potential for a second process to recreate the file and also
3225 acquire a lock. This could lead to two Exim processes writing to the file
3226 at the same time. The added code performs the same test as UW imapd; it
3227 checks after acquiring the lock that its file descriptor still refers to
3228 the same named file.
3230 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3231 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3234 11. Added the smtp_active_hostname option. If used, this will typically be made
3235 to depend on the incoming interface address. Because $interface_address is
3236 not set up until the daemon has forked a reception process, error responses
3237 that can happen earlier (such as "too many connections") no longer contain
3240 12. If an expansion in a condition on a "warn" statement fails because a lookup
3241 defers, the "warn" statement is abandoned, and the next ACL statement is
3242 processed. Previously this caused the whole ACL to be aborted.
3244 13. Added the iplsearch lookup type.
3246 14. Added ident_timeout as a log selector.
3248 15. Added tls_certificate_verified as a log selector.
3250 16. Added a global option tls_require_ciphers (compare the smtp transport
3251 option of the same name). This controls incoming TLS connections.
3253 17. I finally figured out how to make tls_require_ciphers do a similar thing
3254 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3255 before starting the TLS session.
3257 18. Tabs are now shown as \t in -bP output.
3259 19. If the log selector return_path_on_delivery was set, Exim crashed when
3260 bouncing a message because it had too many Received: header lines.
3262 20. If two routers both had headers_remove settings, and the first one included
3263 a superfluous trailing colon, the final name in the first list and the
3264 first name in the second list were incorrectly joined into one item (with a
3265 colon in the middle).
3271 1. Added -C and -D options to the exinext utility, mainly to make it easier
3272 to include in the automated testing, but these could be helpful when
3273 multiple configurations are in use.
3275 2. The exinext utility was not formatting the output nicely when there was
3276 an alternate port involved in the retry record key, nor when there was a
3277 message id as well (for retries that were specific to a specific message
3278 and a specific host). It was also confused by IPv6 addresses, because of
3279 the additional colons they contain. I have fixed the IPv4 problem, and
3280 patched it up to do a reasonable job for IPv6.
3282 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3283 delivery, the log line now contains "pipelined" if PIPELINING was used.
3285 4. An SMTP transport process used to panic and die if the bind() call to set
3286 an explicit outgoing interface failed. This has been changed; it is now
3287 treated in the same way as a connect() failure.
3289 5. A reference to $sender_host_name in the part of a conditional expansion
3290 that was being skipped was still causing a DNS lookup. This no longer
3293 6. The def: expansion condition was not recognizing references to header lines
3294 that used bh_ and bheader_.
3296 7. Added the _cache feature to named lists.
3298 8. The code for checking quota_filecount in the appendfile transport was
3299 allowing one more file than it should have been.
3301 9. For compatibility with Sendmail, the command line option
3309 and sets the incoming protocol and host name (for trusted callers). The
3310 host name and its colon can be omitted when only the protocol is to be set.
3311 Note the Exim already has two private options, -pd and -ps, that refer to
3312 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3313 "s", but I don't think that's a major issue.
3315 10. A number of refactoring changes to the code, none of which should affect
3318 (a) The number of logging options was getting close to filling up the
3319 32-bit word that was used as a bit map. I have split them into two classes:
3320 those that are passed in the argument to log_write(), and those that are
3321 only ever tested independently outside of that function. These are now in
3322 separate 32-bit words, so there is plenty of room for expansion again.
3323 There is no change in the user interface or the logging behaviour.
3325 (b) When building, for example, log lines, the code previously used a
3326 macro that called string_cat() twice, in order to add two strings. This is
3327 not really sufficiently general. Furthermore, there was one instance where
3328 it was actually wrong because one of the argument was used twice, and in
3329 one call a function was used. (As it happened, calling the function twice
3330 did not affect the overall behaviour.) The macro has been replaced by a
3331 function that can join an arbitrary number of extra strings onto a growing
3334 (c) The code for expansion conditions now uses a table and a binary chop
3335 instead of a serial search (which was left over from when there were very
3336 few conditions). Also, it now recognizes conditions like "pam" even when
3337 the relevant support is not compiled in: a suitably worded error message is
3338 given if an attempt is made to use such a condition.
3340 11. Added ${time_interval:xxxxx}.
3342 12. A bug was causing one of the ddress fields not to be passed back correctly
3343 from remote delivery subprocesses. The field in question was not being
3344 subsequently used, so this caused to problems in practice.
3346 13. Added new log selectors queue_time and deliver_time.
3348 14. Might have fixed a bug in maildirsizefile handling that threw up
3349 "unexpected character" debug warnings, and recalculated the data
3350 unnecessarily. In any case, I expanded the warning message to give more
3353 15. Added the message "Restricted characters in address" to the statements in
3354 the default ACL that block characters like @ and % in local parts.
3356 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3357 Three changes have been made:
3359 (a) There was a serious bug; a negative response to MAIL caused the whole
3360 recipient domain to be cached as invalid, thereby blocking all messages
3361 to all local parts at the same domain, from all senders. This bug has
3362 been fixed. The domain is no longer cached after a negative response to
3363 MAIL if the sender used is not empty.
3365 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3368 (c) A new callout option, "use_sender" has been added for people who want
3369 the modified behaviour.
3375 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3378 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3379 indeed breaks things for older releases.
3381 3. Added additional logging to the case where there is a problem reading data
3382 from a filter that is running in a subprocess using a pipe, in order to
3383 try to track down a specific problem.
3385 4. Testing facility fudge: when running in the test harness and attempting
3386 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3387 getting "No route to host". Convert this to a timeout.
3389 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3392 6. Some OS don't have socklen_t but use size_t instead. This affects the
3393 fifth argument of getsockopt() amongst other things. This is now
3394 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3395 can be set for individual OS. I have set it for SunOS5, OSF1, and
3396 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3397 some earlier ones do not.
3399 7. Change 4.30/15 was not doing the test caselessly.
3401 8. The standard form for an IPv6 address literal was being rejected by address
3402 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3403 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3404 this, as well as the form without the "IPv6" on the front (but only when
3405 address literals are enabled, of course).
3407 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3409 10. Exim crashed if a message with an empty sender address specified by -f
3410 encountered a router with an errors_to setting. This could be provoked only
3411 by a command such as
3415 where an empty string was supplied; "<>" did not hit this bug.
3417 11. Installed PCRE release 4.5.
3419 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3420 remained set. It is now erased.
3422 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3423 times from message ids (which are base 36 rather than the normal 62).
3425 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3426 were being counted as actual protocol errors, and logged if the log
3427 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3428 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3429 and DATA following a set of rejected RCPTs do not count as protocol errors.
3430 In other words, Exim assumes they were pipelined, though this may not
3431 actually be the case. Of course, in all cases the client gets an
3432 appropriate error code.
3434 15. If a lookup fails in an ACL condition, a message about the failure may
3435 be available; it is used if testing the ACL cannot continue, because most
3436 such messages specify what the cause of the deferral is. However, some
3437 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3438 that caused an old message to be retained and used if a later statement
3439 caused a defer, replacing the real cause of the deferral.
3441 16. If an IP address had so many PTR records that the DNS lookup buffer
3442 was not large enough to hold them, Exim could crash while trying to process
3443 the truncated data. It now detects and logs this case.
3445 17. Further to 4.21/58, another change has been made: if (and only if) the
3446 first line of a message (the first header line) ends with CRLF, a bare LF
3447 in a subsequent header line has a space inserted after it, so as not to
3448 terminate the header.
3450 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3451 unnecessarily, used atoi() instead of strtol(), and didn't check the
3452 termination when getting file sizes from file names by regex.
3454 19. Completely re-implemented the support for maildirsize files, in the light
3455 of a number of problems with the previous contributed implementation
3456 (4.30/29). In particular:
3458 . If the quota is zero, the maildirsize file is maintained, but no quota is
3461 . If the maildir directory does not exist, it is created before any attempt
3462 to write a maildirsize file.
3464 . The quota value in the file is just a cache; if the quota is changed in
3465 the transport, the new value overrides.
3467 . A regular expression is available for excluding directories from the
3470 20. The autoreply transport checks the characters in options that define the
3471 message's headers; it allows continued headers, but it was checking with
3472 isspace() after an embedded newline instead of explicitly looking for a
3475 21. If all the "regular" hosts to which an address was routed had passed their
3476 expiry times, and had not reached their retry times, the address was
3477 bounced, even if fallback hosts were defined. Now Exim should go on to try
3480 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3481 equivalent code in the SMTP transport. Some hosts send humungous responses
3482 to HELO/EHLO, more than 1024 it seems.
3484 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3485 gives compiler warnings in some environments. I've now done it "properly",
3488 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3489 (because of problems with the built-in one) was declared to return uschar *
3490 instead of char *, causing compiler failure.
3492 25. Fixed a file descriptor leak when processing alias/forward files.
3494 26. Fixed a minor format string issue in dbfn.c.
3496 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3498 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3499 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3502 29. When a sender address is verified, it is cached, to save repeating the test
3503 when there is more than one recipient in a message. However, when the
3504 verification involves a callout, it is possible for different callout
3505 options to be set for different recipients. It is too complicated to keep
3506 track of this in the cache, so now Exim always runs a verification when a
3507 callout is required, relying on the callout cache for the optimization.
3508 The overhead is duplication of the address routing, but this should not be
3511 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3512 to be verified with callout=postmaster, and the main callout worked but the
3513 postmaster check failed, the verification correctly failed. However, if a
3514 subsequent RCPT command asked for sender verification *without* the
3515 postmaster check, incorrect caching caused this verification also to fail,
3518 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3519 it was not caching the DNS options (qualify_single, search_parents) that
3520 were used when the lookup failed. A subsequent lookup with different
3521 options therefore always gave the same answer, though there were cases
3522 where it should not have. (Example: a "domains = !$mx_any" option on a
3523 dnslookup router: the "domains" option is always processed without any
3524 widening, but the router might have qualify_single set.) Now Exim uses the
3525 cached value only when the same options are set.
3527 32. Added John Jetmore's "exipick" utility to the distribution.
3529 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3530 than a timeout (e.g. a certificate is required, and is not provided), an
3531 Exim server now closes the connection immediately. Previously it waited for
3532 the client to close - but if the client is SSL, it seems that they each
3533 wait for each other, leading to a delay before one of them times out.
3535 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3536 maintained 0.8.x compatibility because I don't think many are using it, and
3537 it is clearly obsolete.
3539 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3542 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3543 client certificate was expired. A simple patch fixes this, though I don't
3544 understand the full logic of why the verify callback is called multiple
3547 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3548 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3549 which causes problems with some clients (such as the Certicom SSL Plus
3550 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3551 disables the coutermeasure allowing Eudora to connect."
3553 38. Exim was not checking that a write() to a log file succeeded. This could
3554 lead to Bad Things if a log got too big, in particular if it hit a file
3555 size limit. Exim now panics and dies if it cannot write to a log file, just
3556 as it does if it cannot open a log file.
3558 39. Modified OS/Makefile-Linux so that it now contains
3560 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3562 The two -D definitions ensure that Exim is compiled with large file
3563 support, which makes it possible to handle log files that are bigger than
3566 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3567 instance) a domain was checked against a named list that involved a lookup,
3568 causing $domain_data to be set, then another domain was checked against the
3569 same list, then the first domain was re-checked, the value of $domain_data
3570 after the final check could be wrong. In particular, if the second check
3571 failed, it could be set empty. This bug probably also applied to
3574 41. The strip_trailing_dot option was not being applied to the address given
3575 with the -f command-line option.
3577 42. The code for reading a message's header from the spool was incrementing
3578 $received_count, but never initializing it. This meant that the value was
3579 incorrect (doubled) while delivering a message in the same process in which
3580 it was received. In the most common configuration of Exim, this never
3581 happens - a fresh exec is done - but it can happen when
3582 deliver_drop_privilege is set.
3584 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3585 it now includes up to 150 characters of the unexpected data in the log
3588 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3589 and building data strings. The size of both of these buffers was 10 000
3590 bytes - far larger than anybody would *ever* want, thought I. Needless to
3591 say, somebody hit the limit. I have increased the maximum line length to
3592 20 000 and the maximum data length of concatenated lines to 100 000. I have
3593 also fixed two bugs, because there was no checking on these buffers. Tsk,
3594 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3595 buffer is too small.
3597 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3598 lsearch lookups. Now it does.
3600 46. When parsing a route_list item in a manualroute router, a fixed-length
3601 buffer was used for the list of hosts. I made this 1024 bytes long,
3602 thinking that nobody would ever have a list of hosts that long. Wrong.
3603 Somebody had a whole pile of complicated expansion conditions, and the
3604 string was silently truncated, leading to an expansion error. It turns out
3605 that it is easier to change to an unlimited length (owing to other changes
3606 that have happened since this code was originally written) than to build
3607 structure for giving a limitation error. The length of the item that
3608 expands into the list of hosts is now unlimited.
3610 47. The lsearch lookup could not handle data where the length of text line was
3611 more than 4095 characters. Such lines were truncated, leading to shortened
3612 data being returned. It should now handle lines of any length.
3614 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3615 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3618 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3619 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3620 no longer happens in comment lines. A list of replacements is now placed
3621 at the head of all of the source files, except those whose only change is
3622 to replace PERL_COMMAND in the very first #! line.
3624 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3625 messages on the queue, with a bottom-up merge sort, using code contributed
3626 by Michael Haardt. This should make operations like -bp somewhat faster on
3627 large queues. It won't affect queue runners, except when queue_run_in_order
3630 51. Installed eximstats 1.31 in the distribution.
3632 52. Added support for SRV lookups to the dnslookup router.
3634 53. If an ACL referred to $message_body or $message_body_end, the value was not
3635 reset for any messages that followed in the same SMTP session.
3637 54. The store-handling optimization for building very long strings was not
3638 differentiating between the different store pools. I don't think this
3639 actually made any difference in practice, but I've tidied it.
3641 55. While running the routers to verify a sender address, $sender_address
3642 was still set to the sender address. This is wrong, because when routing to
3643 send a bounce to the sender, it would be empty. Therefore, I have changed
3644 it so that, while verifying a sender address, $sender_address is set to <>.
3645 (There is no change to what happens when verifying a recipient address.)
3647 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3648 target A or AAAA records (if not already returned) without resetting the
3649 qualify_single or search_parents options of the DNS resolver. These are
3650 inappropriate in this case because the targets of MX and SRV records must
3651 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3652 target that, when qualified, matched something in the local domain. These
3653 two options are now turned off when doing these lookups.
3655 57. It seems that at least some releases of Reiserfs (which does not have the
3656 concept of a fixed number of inodes) returns zero and not -1 for the
3657 number of available inodes. This interacted badly with check_spool_inodes,
3658 which assumed that -1 was the "no such thing" setting. What I have done is
3659 to check that the total number of inodes is greater than zero before doing
3660 the test of how many are available.
3662 58. When a "warn" ACL statement has a log_message modifier, the message is
3663 remembered, and not repeated. This is to avoid a lot of repetition when a
3664 message has many recipients that cause the same warning to be written.
3665 Howewer, Exim was preserving the list of already written lines for an
3666 entire SMTP session, which doesn't seem right. The memory is now reset if a
3667 new message is started.
3669 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3670 debugging output unless log_rewrite was also set.
3672 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3673 of (int)(handle) when we know that handle contains (void *)(-1).
3675 61. The Exim daemon panic-logs an error return when it closes the incoming
3676 connection. However "connection reset by peer" seems to be common, and
3677 isn't really an error worthy of noting specially, so that particular error
3680 62. When Exim is trying to find all the local interfaces, it used to panic and
3681 die if the ioctl to get the interface flags failed. However, it seems that
3682 on at least one OS (Solaris 9) it is possible to have an interface that is
3683 included in the list of interfaces, but for which you get a failure error
3684 for this call. This happens when the interface is not "plumbed" into a
3685 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3686 failure of the "get flags" call assumes that the interface is down.
3688 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3689 makes life easier for people who are doing arithmetic on fields extracted
3690 from dates, where you often get leading zeros that should not be
3691 interpreted as octal.
3693 64. Added qualify_domain to the redirect router, to override the global
3696 65. If a pathologically long header line contained very many addresses (the
3697 report of this problem mentioned 10 000) and each of them was rewritten,
3698 Exim could use up a very large amount of memory. (It kept on making new
3699 copies of the header line as it rewrote, and never released the old ones.)
3700 At the expense of a bit more processing, the header rewriting function has
3701 been changed so that it no longer eats memory in this way.
3703 66. The generation of the Received: header has been moved from the time that a
3704 message starts to be received, to the time that it finishes. The timestamp
3705 in the Received: header should now be very close to that of the <= log
3706 line. There are two side-effects of this change:
3708 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3709 logged header lines no longer include the local Received: line, because
3710 it has not yet been created. The same applies to a copy of the message
3711 that is returned to a non-SMTP sender when a message is rejected.
3713 (b) When a filter file is tested using -bf, no additional Received: header
3714 is added to the test message. After some thought, I decided that this
3717 This change does not affect the value of $received_for. It is still set
3718 after address rewriting, but before local_scan() is called.
3720 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3722 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3723 gave an unhelpful panic error message, and a defer error. I have managed to
3724 change this behaviour so that it now rejects any supplied certificate,
3725 which seems right, as the list of acceptable certificates is empty.
3727 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3728 gave an unhelpful defer error. I have not managed to make this reject any
3729 supplied certificates, but the error message it gives is "no certificate
3730 supplied", which is not helpful.
3732 70. exigrep's output now also includes lines that are not associated with any
3733 message, but which match the given pattern. Implemented by a patch from
3734 Martin Sluka, which also tidied up the Perl a bit.
3736 71. Recipient callout verification, like sender verification, was using <> in
3737 the MAIL FROM command. This isn't really the right thing, since the actual
3738 sender may affect whether the remote host accepts the recipient or not. I
3739 have changed it to use the actual sender in the callout; this means that
3740 the cache record is now keyed on a recipient/sender pair, not just the
3741 recipient address. There doesn't seem to be a real danger of callout loops,
3742 since a callout by the remote host to check the sender would use <>.
3743 [SEE ABOVE: changed after hitting problems.]
3745 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3746 temporary errors. However, in the case of such a code being given after
3747 the end of a data transmission (i.e. after ".") Exim was failing to write
3748 a retry record for the message. (Yes, there was some broken host that was
3749 actually sending 8xx at this point.)
3751 73. An unknown lookup type in a host list could cause Exim to panic-die when
3752 the list was checked. (An example that provoked this was putting <; in the
3753 middle of a list instead of at the start.) If this happened during a DATA
3754 ACL check, a -D file could be left lying around. This kind of configuration
3755 error no longer causes Exim to die; instead it causes a defer errror. The
3756 incident is still logged to the main and panic logs.
3758 74. Buglet left over from Exim 3 conversion. The message "too many messages
3759 in one connection" was written to the rejectlog but not the mainlog, except
3760 when address rewriting (yes!) was being logged.
3762 75. Added write_rejectlog option.
3764 76. When a system filter was run not as root (that is, when system_filter_user
3765 was set), the values of the $n variables were not being returned to the
3766 main process; thus, they were not subsequently available in the $sn
3769 77. Added +return_path_on_delivery log selector.
3771 78. A connection timeout was being treated differently from recipients deferred
3772 when testing hosts_max_try with a message that was older than the host's
3773 retry timeout. (The host should not be counted, thus allowing all hosts to
3774 be tried at least once before bouncing.) This may have been the cause of an
3775 occasionally reported bug whereby a message would remain on the queue
3776 longer than the retry timeout, but would be bounced if a delivery was
3777 forced. I say "may" because I never totally pinned down the problem;
3778 setting up timeout/retry tests is difficult. See also the next item.
3780 79. The ultimate address timeout was not being applied to errors that involved
3781 a combination of host plus message (for example, a timeout on a MAIL
3782 command). When an address resolved to a number of possible hosts, and they
3783 were not all tried for each delivery (e.g. because of hosts_max_try), a
3784 message could remain on the queue longer than the retry timeout.
3786 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3789 81. Fixed an obscure SMTP outgoing bug which required at least the following
3790 conditions: (a) there was another message waiting for the same server;
3791 (b) the server returned 5xx to all RCPT commands in the first message so
3792 that the message was not completed; (c) the server dropped the connection
3793 or gave a negative response to the RSET that Exim sends to abort the
3794 transaction. The observed case was a dropped connection after DATA that had
3795 been sent in pipelining mode. That is, the server had advertised PIPELINING
3796 but was not implementing it correctly. The effect of the bug was incorrect
3797 behaviour, such as trying another host, and this could lead to a crash.
3803 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3804 and daemon.c were passed as pointers to ints; they should have been
3805 pointers to socklen_t variables (which are typically unsigned ints).
3807 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3810 3. Fixed a really odd bug that affected only the testing scheme; patching a
3811 certain fixed string in the binary changed the value of another string that
3812 happened to be identical to the end of the original first string.
3814 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3815 name", it returns that address as the IP address. On some operating
3816 systems (e.g. Solaris), it also passes back the IP address string as the
3817 "host name". However, on others (e.g. Linux), it passes back an empty
3818 string. Exim wasn't checking for this, and was changing the host name to an
3819 empty string, assuming it had been canonicized.
3821 5. Although rare, it is permitted to have more than one PTR record for a given
3822 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3823 all the names associated with an address, because they do in Solaris.
3824 However, it seems that they do not in Linux for data that comes from the
3825 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3826 I found this out when I moved to a new Linux workstation and tried to run
3827 the Exim test suite.
3829 To get round this problem I have changed the code so that it now does its
3830 own call to the DNS to look up PTR records when searching for a host name.
3831 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3832 addresses that are only in /etc/hosts are still found.
3834 This behaviour is, however, controlled by an option called host_lookup_
3835 order, which defaults to "bydns:byaddr". If people want to use the other
3836 order, or indeed, just use one or the other means of lookup, they can
3837 specify it in this variable.
3839 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3840 some operating systems, this comes back from gethostbyaddr() as an empty
3841 string, and this is what Exim used to test for. However, it seems that in
3842 other systems, "." is yielded. Exim now tests for this case too.
3844 7. The values of check_spool_space and check_log_space are now held internally
3845 as a number of kilobytes instead of an absolute number of bytes. If a
3846 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3847 kilobyte. This means that much larger values can be stored.
3849 8. Exim monitor: an attempt to get the action menu when not actually pointing
3850 at a message produces an empty menu entitled "No message selected". This
3851 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3852 no entries in it ("Shell widget menu has zero width and/or height"). So I
3853 have added a single, blank menu entry in this case.
3855 9. Added ${quote_local_part.
3857 10. MIME decoding is now applied to the contents of Subject: header lines when
3860 11. Now that a reference to $sender_host_address automatically causes a reverse
3861 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3862 host lookup before query-style lookups in lists that might use this
3863 variable. This has therefore been abolished, and the "net-" prefix is no
3864 longer necessary for query-style lookups.
3866 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3867 have been a typo for LFLAGS, so it has been changed.
3869 13. The install script calls Exim with "-C /dev/null" in order to find the
3870 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3871 to be output. Howeve, since Exim outputs its version number before the
3872 error, it didn't break the script. It just looked ugly. I fixed this by
3873 always allowing "-C /dev/null" if the caller is root.
3875 14. Ignore overlarge ACL variable number when reading spool file - insurance
3876 against a later release with more variables having written the file.
3878 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3879 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3880 this, as well as the form without the "IPv6" on the front.
3882 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3883 OS/Makefile-Darwin file.
3885 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3886 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3888 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3889 variable. However, some OS use ints for inodes. I've added cast to long int
3890 to get rid of the compiler warning.
3892 19. I had forgotten to lock out "/../" in configuration file names when
3893 ALT_CONFIG_PREFIX was set.
3895 20. Routers used for verification do not need to specify transports. However,
3896 if such a router generated a host list, and callout was configured, Exim
3897 crashed, because it could not find a port number from the (non-existent)
3898 transport. It now assumes port 25 in this circumstance.
3900 21. Added the -t option to exigrep.
3902 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3903 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3904 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3905 them is set, LOOKUP_LSEARCH is forced.
3907 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3909 24. Added sender and host information to the "rejected by local_scan()" log
3910 line; previously there was no indication of these.
3912 25. Added .include_if_exists.
3914 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3915 Linux. It turns out that not all file systems support this. Apparently some
3916 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3917 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3918 not supported on the file descriptor, is now ignored when Exim is trying to
3919 sync a directory. This applies only to Linux.
3921 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3923 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3924 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3927 29. Added support for maildirsize files from supplied patch (modified a bit).
3929 30. The use of :fail: followed by an empty string could lead Exim to respond to
3930 sender verification failures with (e.g.):
3932 550 Verification failed for <xxx>
3933 550 Sender verify failed
3935 where the first response line was missing the '-' that indicates it is not
3936 the final line of the response.
3938 31. The loop for finding the name of the user that called Exim had a hardwired
3939 limit of 10; it now uses the value of finduser_retries, which is used for
3940 all other user lookups.
3942 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3945 33. Exim was neglecting to zero errno before one call of strtol() when
3946 expanding a string and expecting an integer value. On some systems this
3947 resulted in spurious "integer overflow" errors. Also, it was casting the
3948 result into an int without checking.
3950 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3951 did not work. The code looks as if it has *never* worked, though it appears
3952 to have been documented since at least releast 1.62. I have made it work.
3954 35. The "timeout_DNS" error in retry rules, also documented since at least
3955 1.62, also never worked. As it isn't clear exactly what this means, and
3956 clearly it isn't a major issue, I have abolished the feature by treating it
3957 as "timeout", and writing a warning to the main and panic logs.
3959 36. The display of retry rules for -brt wasn't always showing the error code
3962 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3963 timeout_connect_A, timeout_connect_MX.
3965 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3966 to the empty sender.
3968 39. The daemon was not analysing the content of -oX till after it had closed
3969 stderr and disconnected from the controlling terminal. This meant that any
3970 syntax errors were only noted on the panic log, and the return code from
3971 the command was 0. By re-arranging the code a little, I've made the
3972 decoding happen first, so such errors now appear on stderr, and the return
3973 code is 1. However, the actual setting up of the sockets still happens in
3974 the disconnected process, so errors there are still only recorded on the
3977 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3978 connections (as happens on some IP stacks) was logged at start up time as
3979 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3980 it from "IPv6 and IPv4", which means that two separate sockets are being
3983 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3984 says whether AF_INET or AF_INET6 was passed as an argument.
3986 42. Exiwhat output was messed up when time zones were included in log
3989 43. Exiwhat now gives more information about the daemon's listening ports,
3990 and whether -tls-on-connect was used.
3992 44. The "port" option of the smtp transport is now expanded.
3994 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3995 silently ignored. Now an error message is written to the main and panic
3998 46. There's a new ACL modifier called "logwrite" which writes to a log file
3999 as soon as it is encountered.
4001 47. Added $local_user_uid and $local_user_gid at routing time.
4003 48. Exim crashed when trying to verify a sender address that was being
4006 49. Exim was recognizing only a space character after ".include". It now also
4007 recognizes a tab character.
4009 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4010 extracting the relevant information from the specification. The man page no
4011 longer contains scrambled data for the -d option, and I've added a section
4012 at the front about calling Exim under different names.
4014 51. Added "extra_headers" argument to the "mail" command in filter files.
4016 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4019 53. Installed eximstats 1.29.
4021 54. Added transport_filter_timeout as a generic transport option.
4023 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4024 Cc: header lines. This was required by RFC 822, but it not required by RFC
4027 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4028 incoming messages that did not have them. Now it does so only if the
4029 message originates locally, that is, if there is no associated remote host
4030 address. When Resent- header lines are present, this applies to the Resent-
4031 lines rather than the non-Resent- lines.
4033 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4034 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4036 58. Messages for configuration errors now include the name of the main
4037 configuration file - useful now that there may be more than one file in a
4038 list (.included file names were always shown).
4040 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4041 for those rare installations that do not start the daemon as root or run it
4042 setuid root. I've cut out the call to initgroups() if the daemon is not
4045 60. The Exim user and group can now be bound into the binary as text strings
4046 that are looked up at the start of Exim's processing.
4048 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4050 62. Added $mailstore_basename variable.
4052 63. Installed patch to sieve.c from Michael Haardt.
4054 64. When Exim failed to open the panic log after failing to open the main log,
4055 the original message it was trying to log was written to stderr and debug
4056 output, but if they were not available (the usual case in production), it
4057 was lost. Now it is written to syslog before the two lines that record the
4058 failures to open the logs.
4060 65. Users' Exim filters run in subprocesses under the user's uid. It is
4061 possible for a "deliver" command or an alias in a "personal" command to
4062 provoke an address rewrite. If logging of address rewriting is configured,
4063 this fails because the process is not running as root or exim. There may be
4064 a better way of dealing with this, but for the moment (because 4.30 needs
4065 to be released), I have disabled address rewrite logging when running a
4066 filter in a non-root, non-exim process.
4072 1. The buildconfig auxiliary program wasn't quoting the value set for
4073 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4074 not defined. This bug was masked in 4.22 by the effect that was fixed in
4077 2. Some messages that were rejected after a message id was allocated were
4078 shown as "incomplete" by exigrep. It no longer does this for messages that
4079 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4081 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4082 have allow_domain_literals set, the ID did not get logged in the <= line.
4083 Domain literals are now always recognized in Message-ID: header lines.
4085 4. The first argument for a ${extract expansion item is the key name or field
4086 number. Leading and trailing spaces in this item were not being ignored,
4087 causing some misleading effects.
4089 5. When deliver_drop_privilege was set, single queue runner processes started
4090 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4091 same command in the process it spins off) were not dropping privilege.
4093 6. When the daemon running as "exim" started a queue runner, it always
4094 re-executed Exim in the spun-off process. This is a waste of effort when
4095 deliver_drop_privilege is set. The new process now just calls the
4096 queue-runner function directly.
4102 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4105 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4106 ignored. Though the use of -oP was forcing the writing of a pid file, it
4107 was always written to the default place.
4109 3. If the message "no IP address found for host xxxx" is generated during
4110 incoming verification, it is now followed by identification of the incoming
4111 connection (so you can more easily find what provoked it).
4113 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4115 5. Added some features to "harden" Exim a bit more against certain attacks:
4117 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4118 be put in Local/Makefile. This is like the never_users runtime option,
4119 but it cannot be overridden. The default setting is "root".
4121 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4122 prefix string with which any file named in a -C command line option
4125 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4126 is retained for -C and -D only if the caller of Exim is root. Without
4127 it, the exim user may also use -C and -D and retain privilege.
4129 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4130 command line option is disabled.
4132 6. Macro names set by the -D option must start with an upper case letter, just
4133 like macro names defined in the configuration file.
4135 7. Added "dereference=" facility to LDAP.
4137 8. Two instances of the typo "uknown" in the source files are fixed.
4139 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4140 the Configure-Makefile script screwed up while processing it.
4142 10. Incorporated PCRE 4.4.
4144 11. The SMTP synchronization check was not operating right at the start of an
4145 SMTP session. For example, it could not catch a HELO sent before the client
4146 waited for the greeting. There is now a check for outstanding input at the
4147 point when the greeting is written. Because of the duplex, asynchronous
4148 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4149 way, but not yet received, when the check is performed.
4151 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4152 on TCP/IP sockets, because this apparently causes some broken clients to
4155 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4156 unchanged) from the Cygwin maintainer.
4158 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4159 was supported instead of testing for mbx. Effectively a typo.
4161 15. The spa authenticator server code was not checking that the input it
4162 received was valid base64.
4164 16. The debug output line for the "set" modifier in ACLs was not showing the
4165 name of the variable that was being set.
4167 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4169 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4170 lookup on its own. Something else was needed to trigger the lookup. For
4171 example, a match in host_lookup or the need for a host name in a host list.
4172 Now, if $sender_host_name is referenced and the host name has not yet been
4173 looked up, a lookup is performed. If the lookup fails, the variable remains
4174 empty, and $host_lookup_failed is set to "1".
4176 19. Added "eqi" as a case-independent comparison operator.
4178 20. The saslauthd authentication condition could segfault if neither service
4179 nor realm was specified.
4181 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4182 error message that was logged was misleading, and incoming SMTP
4183 connections were dropped. The message is now more accurate, and temporary
4184 errors are given to SMTP connections.
4186 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4187 (except RSET) with a 421 error, until QUIT is received. However, it was
4188 failing to send a response to QUIT.
4190 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4191 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4192 ACL is now run after the helo_try_verify_hosts code.
4194 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4195 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4196 case-independent, so other case variants are also recognized). Apparently
4197 some systems use these upper case variants.
4199 25. If more than two messages were waiting for the same host, and a transport
4200 filter was specified for the transport, Exim sent two messages over the
4201 same TCP/IP connection, and then failed with "socket operation on non-
4202 socket" when it tried to send the third.
4204 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4206 27. The extern definition of crypt16() in expand.c was not being excluded when
4207 the OS had its own crypt16() function.
4209 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4210 as a preferred synonym for return_size_limit, both as an option and as an
4213 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4215 30. Changed the default configuration ACL to relax the local part checking rule
4216 for addresses that are not in any local domains. For these addresses,
4217 slashes and pipe symbols are allowed within local parts, but the sequence
4218 /../ is explicitly forbidden.
4220 31. SPA server authentication was not clearing the challenge buffer before
4223 32. log_message in a "warn" ACL statement was writing to the reject log as
4224 well as to the main log, which contradicts the documentation and doesn't
4225 seem right (because no rejection is happening). So I have stopped it.
4227 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4228 However, I am unable to do any testing of this.
4230 34. Fixed an infelicity in the appendfile transport. When checking directories
4231 for a mailbox, to see if any needed to be created, it was accidentally
4232 using path names with one or more superfluous leading slashes; tracing
4233 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4235 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4236 contents are added to the log line that is written for every discarded
4237 recipient. (Previously a log_message setting was ignored.)
4239 36. The ${quote: operator now quotes the string if it is empty.
4241 37. The install script runs exim in order to find its version number. If for
4242 some reason other than non-existence or emptiness, which it checks, it
4243 could not run './exim', it was installing it with an empty version number,
4244 i.e. as "exim-". This error state is now caught, and the installation is
4247 38. An argument was missing from the function that creates an error message
4248 when Exim fails to connect to the socket for saslauthd authentication.
4249 This could cause Exim to crash, or give a corrupted message.
4251 39. Added isip, isip4, and isip6 to ${if conditions.
4253 40. The ACL variables $acl_xx are now saved with the message, and can be
4254 accessed later in routers, transports, and filters.
4256 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4257 strings in the file are not string-expanded.
4259 42. If a MAIL command specified a SIZE value that was too large to fit into an
4260 int variable, the check against message_size_limit failed. Such values are
4261 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4262 day this will have to be increased, but I don't think I want to be around
4263 when emails are that large.
4270 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4271 iconv() is not standard in FreeBSD.
4273 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4274 IPv6 enabled. The observed symptom was a segmentation fault on return from
4275 the function os_common_find_running_interfaces() in src/os.c.
4277 3. In the check_special_case() function in daemon.c I had used "errno" as an
4278 argument name, which causes warnings on some systems. This was basically a
4279 typo, since it was named "eno" in the comments!
4281 4. The code that waits for the clock to tick (at a resolution of some fraction
4282 of a second) so as to ensure message-id uniqueness was always waiting for
4283 at least one whole tick, when it could have waited for less. [This is
4284 almost certainly not relevant at current processor speeds, where it is
4285 unlikely to ever wait at all. But we try to future-proof.]
4287 5. The function that sleeps for a time interval that includes fractions of a
4288 second contained a race. It did not block SIGALRM between setting the
4289 timer, and suspending (a couple of lines later). If the interval was short
4290 and the sigsuspend() was delayed until after it had expired, the suspension
4291 never ended. On busy systems this could lead to processes getting stuck for
4294 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4295 process, before it forks any delivery processes. The open lookup caching
4296 mechanism meant that the open file or database connection was passed into
4297 the delivery process. The problem was that delivery processes always tidy
4298 up cached lookup data. This could cause a problem for the next delivery
4299 process started by the queue runner, because the external queue runner
4300 process does not know about the closure. So the next delivery process
4301 still has data in the lookup cache. In the case of a file lookup, there was
4302 no problem because closing a file descriptor in a subprocess doesn't affect
4303 the parent. However, if the lookup was caching a connection to a database,
4304 the connection was closed, and the second delivery process was likely to
4305 see errors such as "PGSQL: query failed: server closed the connection
4306 unexpectedly". The problem has been fixed by closing all cached lookups
4307 in a queue runner before running a delivery process.
4309 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4310 seem to have the "const" qualifier which it has on other OS. I've
4313 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4314 *of the same type* (client or server) with the same public name that an
4315 error should be diagnosed.
4317 9. When Exim looked up a host name for an IP address, but failed to find the
4318 original IP address when looking up the host name (a safety check), it
4319 output the message "<ip address> does not match any IP for NULL", which was
4320 confusing, to say the least. The bug was that the host name should have
4321 appeared instead of "NULL".
4323 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4324 user that is built into the binary, and the -C or -D options is used, root
4325 privilege is dropped before the configuration file is read. In addition,
4326 logging is switched to stderr instead of the normal log files. If the
4327 configuration then re-defines the Exim user, the unprivileged environment
4328 is probably not what is expected, so Exim logs a panic warning message (but
4331 However, if deliver_drop_privilege is set, the unprivileged state may well
4332 be exactly what is intended, so the warning has been cut out in that case,
4333 and Exim is allowed to try to write to its normal log files.
4339 1. smtp_return_error_details was not giving details for temporary sender
4340 or receiver verification errors.
4342 2. Diagnose a configuration error if two authenticators have the same public
4345 3. Exim used not to create the message log file for a message until the first
4346 delivery attempt. This could be confusing when incoming messages were held
4347 for policy or load reasons. The message log file is now created at the time
4348 the message is received, and an initial "Received" line is written to it.
4350 4. The automatically generated man page for command line options had a minor
4351 bug that caused no ill effects; however, a more serious problem was that
4352 the procedure for building the man page automatically didn't always
4353 operate. Consequently, release 4.20 contains an out-of-date version. This
4354 shouldn't happen again.
4356 5. When building Exim with embedded Perl support, the script that builds the
4357 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4358 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4360 6. The freeze_tell option was not being used for messages that were frozen on
4361 arrival, either by an ACL or by local_scan().
4363 7. Added the smtp_incomplete_transaction log selector.
4365 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4366 was accepting AUTH without a new EHLO.
4368 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4369 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4372 10. Logging of TCP/IP connections (when configured) now happens in the main
4373 daemon process instead of the child process, so that the TCP/IP connection
4374 count is more accurate (but it can never be perfect).
4376 11. The use of "drop" in a nested ACL was not being handled correctly in the
4377 outer ACL. Now, if condition failure induced by the nested "drop" causes
4378 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4379 or "require"), the connection is dropped.
4381 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4382 that yield "discard" can now be used with an "accept" or a "discard" verb,
4383 but an error is generated for any others (because I can't see a useful way
4384 to define what should happen).
4386 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4387 are now processed in the same way as lines in the Exim configuration file.
4388 In particular, continuation lines are supported.
4390 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4392 15. Added -ti meaning -t -i.
4394 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4395 domains, and warn by logging if others are found.
4397 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4398 returned from GIFCONF when Exim is trying to find the list of interfaces on
4399 a host. The code in os.c has been modified to copy each ifreq to an aligned
4400 structure in all cases.
4402 Also, in some cases, the returned ifreq's were being copied to a 'struct
4403 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4404 means the last couple of bytes of an IPv6 address could be chopped if the
4405 ifreq contained only a normal sockaddr (14 bytes storage).
4407 18. Named domain lists were not supported in the hosts_treat_as_local option.
4408 An entry such as +xxxx was not recognized, and was treated as a literal
4411 19. Ensure that header lines added by a DATA ACL are included in the reject log
4412 if the ACL subsequently rejects the message.
4414 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4415 MD5 (which is deprecated).
4417 21. When testing a filter file using -bf, Exim was writing a message when it
4418 took the sender from a "From " line in the message, but it was not doing so
4419 when it took $return_path from a Return-Path: header line. It now does.
4421 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4422 with a valid header line field name (a series of printing characters
4423 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4425 23. Changed "disc" in the source to "disk" to conform to the documentation and
4426 the book and for uniformity.
4428 24. Ignore Sendmail's -Ooption=value command line item.
4430 25. When execve() failed while trying to run a command in a pipe transport,
4431 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4432 could be confused with a return value of 69 from the command itself. This
4433 has been changed to 127, the value the shell returns if it is asked to run
4434 a non-existent command. The wording for the related log line suggests a
4435 non-existent command as the problem.
4437 26. If received_header_text expands to an empty string, do not add a Received:
4438 header line to the message. (Well, it adds a token one on the spool, but
4439 marks it "old" so that it doesn't get used or transmitted.)
4441 27. Installed eximstats 1.28 (addition of -nt option).
4443 28. There was no check for failure on the call to getsockname() in the daemon
4444 code. This can fail if there is a shortage of resources on the system, with
4445 ENOMEM, for example. A temporary error is now given on failure.
4447 29. Contrary to the C standard, it seems that in some environments, the
4448 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4449 program. Exim now does this explicitly; it affects the formatting of
4450 timestamps using strftime().
4452 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4453 complaints. I've now initialized all the variables, to avoid this.
4455 32. Header lines added by a system filter were not being "seen" during
4456 transport-time rewrites.
4458 33. The info_callback() function passed to OpenSSL is set up with type void
4459 (*)(SSL *, int, int), as described somewhere. However, when calling the
4460 function (actually a macro) that sets it up, the type void(*)() is
4461 expected. I've put in a cast to prevent warnings from picky compilers.
4463 34. If a DNS black list lookup found a CNAME record, but there were no A
4464 records associated with the domain it pointed at, Exim crashed.
4466 35. If a DNS black list lookup returned more than one A record, Exim ignored
4467 all but the first. It now scans all returned addresses if a particular IP
4468 value is being sought. In this situation, the contents of the
4469 $dnslist_value variable are a list of all the addresses, separated by a
4472 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4473 to accept a host name and all its aliases if the forward lookup for any of
4474 them yielded the IP address of the incoming connection. Now it accepts only
4475 those names whose forward lookup yields the correct IP address. Any other
4476 names are discarded. This closes a loophole whereby a rogue DNS
4477 administrator could create reverse DNS records to break through a
4478 wildcarded host restriction in an ACL.
4480 37. If a user filter or a system filter that ran in a subprocess used any of
4481 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4482 the wrong values were passed to the pipe command ($thisaddress had the
4483 value of $0, $0 had the value of $1, etc). This bug was introduced by
4484 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4486 38. Improved the line breaking for long SMTP error messages from ACLs.
4487 Previously, if there was no break point between 40 and 75 characters, Exim
4488 left the rest of the message alone. Two changes have been made: (a) I've
4489 reduced the minimum length to 35 characters; (b) if it can't find a break
4490 point between 35 and 75 characters, it looks ahead and uses the first one
4491 that it finds. This may give the occasional overlong line, but at least the
4492 remaining text gets split now.
4494 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4495 file descriptors might be low, and that setting 1000 would always raise it.
4496 It turns out that in some environments, the limit is already over 1000 and
4497 that lowering it causes trouble. So now Exim takes care not to decrease it.
4499 40. When delivering a message, the value of $return_path is set to $sender_
4500 address at the start of routing (routers may change the value). By an
4501 oversight, this default was not being set up when an address was tested by
4502 -bt or -bv, which affected the outcome if any router or filter referred to
4505 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4506 log only when "message" or "log_message" are set. However, if one of the
4507 conditions was an address verification, or a call to a nested ACL, the
4508 messages generated by the underlying test were being passed through. This
4509 no longer happens. The underlying message is available in $acl_verify_
4510 message for both "message" and "log_message" expansions, so it can be
4511 passed through if needed.
4513 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4514 new expansion $bh_ to give the encoded byte string without charset
4515 translation. Translation happens only if iconv() is available; HAVE_ICONV
4516 indicates this at build time. HEADERS_CHARSET gives the charset to
4517 translate to; headers_charset can change it in the configuration, and
4518 "headers charset" can change it in an individual filter file.
4520 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4521 that creates RFC 2047 encoded "words" labels them as that charset instead
4522 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4523 expansion operator; (ii) when Exim creates a From: line for a local
4524 message; (iii) when a header line is rewritten to include a "phrase" part.
4526 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4527 buggy, causing it to skip the first lines of messages whose message ID
4528 ended in 'D'. This would not have bitten before Exim release 4.14, because
4529 message IDs were unlikely to end in 'D' before then. The effect was to have
4530 incorrect size information for certain domains.
4532 45. #include "config.h" was missing at the start of the crypt16.c module. This
4533 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4536 46. If there was a timeout during a "random" callout check, Exim treated it as
4537 a failure of the random address, and carried on sending RSET and the real
4538 address. If the delay was just some slowness somewhere, the response to the
4539 original RCPT would be taken as a response to RSET and so on, causing
4540 mayhem of various kinds.
4542 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4543 when I implemented it. It didn't allow for the fact that some option values
4544 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4545 the right test for positive values.
4547 48. Domain names in DNS records are case-independent. Exim always looks them up
4548 in lower case. Some resolvers return domain names in exactly the case they
4549 appear in the zone file, that is, they may contain uppercase letters. Not
4550 all resolvers do this - some return always lower case. Exim was treating a
4551 change of case by a resolver as a change of domain, similar to a widening
4552 of a domain abbreviation. This triggered its re-routing code and so it was
4553 trying to route what was effectively the same domain again. This normally
4554 caused routing to fail (because the router wouldn't handle the domain
4555 twice). Now Exim checks for this case specially, and just changes the
4556 casing of the domain that it ultimately uses when it transmits the message
4559 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4562 50. If a filter generated a file delivery with a non-absolute name (possible if
4563 no home directory exists for the router), the forbid_file option was not
4566 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4567 the existing equality matching.
4569 52. Exim was using ints instead of ino_t variables in some places where it was
4570 dealing with inode numbers.
4572 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4573 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4574 called TMPDIR, and if it finds it is different, it changes its value.
4576 54. The smtp_printf() function is now made available to local_scan() so
4577 additional output lines can be written before returning. There is also an
4578 smtp_fflush() function to enable the detection of a dropped connection.
4579 The variables smtp_input and smtp_batched_input are exported to
4582 55. Changed the default runtime configuration: the message "Unknown user"
4583 has been removed from the ACL, and instead placed on the localuser router,
4584 using the cannot_route_message feature. This means that any verification
4585 failures that generate their own messages won't get overridden. Similarly,
4586 the "Unrouteable address" message that was in the ACL for unverifiable
4587 relay addresses has also been removed.
4589 56. Added hosts_avoid_esmtp to the smtp transport.
4591 57. The exicyclog script was not checking for the esoteric option
4592 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4593 will work only if exicyclog is run under the appropriate euid.
4595 58. Following a discussion on the list, the rules by which Exim recognises line
4596 endings on incoming messages have been changed. The -dropcr and drop_cr
4597 options are now no-ops, retained only for backwards compatibility. The
4598 following line terminators are recognized: LF CRLF CR. However, special
4599 processing applies to CR:
4601 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4602 nor a local message in the state where . is a terminator.
4604 (ii) If a bare CR is encountered in a header line, an extra space is added
4605 after the line terminator so as not to end the header. The reasoning
4606 behind this is that bare CRs in header lines are most likely either
4607 to be mistakes, or people trying to play silly games.
4609 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4610 was being incorrectly given as 18 bytes larger than it should have been.
4611 This is a VOB (very old bug).
4613 60. This may never have affected anything current, but just in case it has:
4614 When the local host is found other than at the start of a list of hosts,
4615 the local host, those with the same MX, and any that follow, are discarded.
4616 When the list in question was part of a longer list of hosts, the following
4617 hosts (not currently being processed) were also being discarded. This no
4618 longer happens. I'm not sure if this situation could ever has previously
4621 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4624 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4625 header. This is something that is recommended in a new Internet Draft, and
4626 is something that is documented as being done by Sendmail. There are two
4627 possible values. For messages generated by the autoreply transport, Exim
4630 Auto-Submitted: auto-replied
4632 whereas for all other generated messages (e.g. bounces) it adds
4634 Auto-Submitted: auto-generated
4636 63. The "personal" condition in filters now includes a test for the
4637 Auto-Submitted: header. If it contains the string "auto-" the message it
4638 not considered personal.
4640 64. Added rcpt_include_affixes as a generic transport option.
4642 65. Added queue_only_override (default true).
4644 66. Added the syslog_duplication option.
4646 67. If what should have been the first header line of a message consisted of
4647 a space followed by a colon, Exim was mis-interpreting it as a header line.
4648 It isn't of course - it is syntactically invalid and should therefore be
4649 treated as the start of the message body. The misbehaviour could have
4650 caused a number of strange effects, including loss of data in subsequent
4651 header lines, and spool format errors.
4653 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4654 client host had authenticated. This control can now be exercised by an ACL
4655 for more flexibility.
4657 69. By default, callouts do not happen when testing with -bh. There is now a
4658 variant, -bhc, which does actually run the callout code, including
4659 consulting and updating the callout cache.
4661 70. Added support for saslauthd authentication, courtesy of Alexander
4664 71. If statvfs() failed on the spool or log directories while checking their
4665 size for availability, Exim confusingly gave the error "space shortage".
4666 Furthermore, in debugging mode it crashed with a floating point exception.
4667 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4668 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4669 serious problem, Exim now writes to the main and panic logs when this
4670 happens, with details of the failure. It then refuses to accept the
4671 incoming message, giving the message "spool directory problem" or "log
4672 directory problem" with a 421 code for SMTP messages.
4674 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4675 0, 1, and 2 exist, because some OS complain for execs without them (see
4676 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4677 descriptors. However, the code omitted to check that the open succeeded,
4678 causing mysterious errors if for some reason the permissions on /dev/null
4679 got screwed. Now Exim writes a message to the main and panic logs, and
4680 bombs out if it can't open /dev/null.
4682 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4683 interact so that it is all more flexible. It is supposed to remain
4684 backwards compatible. Also added extra_local_interfaces.
4686 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4687 to bomb out with an assertion failure - to the client this appears as a
4688 connection drop. This problem occurs in the part of the code that was taken
4689 from the Samba project. Fortunately, the assertion is in a very simple
4690 function, so I have fixed this by reproducing the function inline in the
4691 one place where it is called, and arranging for authentication to fail
4692 instead of killing the process with assert().
4694 75. The SPA client code was not working when the server requested OEM rather
4695 than Unicode encoding.
4697 76. Added code to make require_files with a specific uid setting more usable in
4698 the case where statting the file as root fails - usually a non-root-mounted
4699 NFS file system. When this happens and the failure is EACCES, Exim now
4700 forks a subprocess and does the per-uid checking as the relevant uid.
4702 77. Added process_log_path.
4704 78. If log_file_path was not explicitly set, a setting of check_log_space or
4705 check_log_inodes was ignored.
4707 79. If a space check for the spool or log partitions fails, the incident is now
4708 logged. Of course, in the latter case the data may get lost...
4710 80. Added the %p formatting code to string_format() so that it can be used to
4711 print addresses in debug_print(). Adjusted all the address printing in the
4712 debugging in store.c to use %p rather than %d.
4714 81. There was a concern that a line of code in smtp_in.c could overflow a
4715 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4716 initially expressed, the concern was not well-founded, because trailing
4717 spaces are removed early. However, if the trailing spaces were followed by
4718 a NULL, they did not get removed, so the overflow was possible. Two fixes
4721 (a) I re-wrote the offending code in a cleaner fashion.
4722 (b) If an incoming SMTP command contains a NULL character, it is rejected
4725 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4726 runs initgroups(), so that if the Exim user is in any additional groups,
4727 they will be used during message reception.
4733 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob