1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.500 2007/04/11 15:06:05 steve 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.
207 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
208 fixed by 4.65/MH/01 (is this a record?) are fixed:
210 (i) An empty string was always treated as zero by the numeric comparison
211 operators. This behaviour has been restored.
213 (ii) It is documented that the numeric comparison operators always treat
214 their arguments as decimal numbers. This was broken in that numbers
215 starting with 0 were being interpreted as octal.
217 While fixing these problems I realized that there was another issue that
218 hadn't been noticed. Values of message_size_limit (both the global option
219 and the transport option) were treated as octal if they started with 0.
220 The documentation was vague. These values are now always treated as
221 decimal, and I will make that clear in the documentation.
227 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
228 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
231 MH/01 Don't check that the operands of numeric comparison operators are
232 integers when their expansion is in "skipping" mode (fixes bug
233 introduced by 4.64-PH/07).
235 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
236 child addresses, Exim now panics and dies. Previously, because the count
237 is held in a short int, deliveries were likely to be lost. As such a
238 large number of recipients for a single message is ridiculous
239 (performance will be very, very poor), I have chosen to impose a limit
240 rather than extend the field.
246 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
247 leftover -K file (the existence of which was triggered by #402).
248 While we were at it, introduced process PID as part of the -K
249 filename. This should rule out race conditions when creating
252 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
253 processing considerably. Previous code took too long for large mails,
254 triggering a timeout which in turn triggers #401.
256 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
257 in the DK code in transports.c. sendfile() is not really portable,
258 hence the _LINUX specificness.
260 TF/01 In the add_headers option to the mail command in an Exim filter,
261 there was a bug that Exim would claim a syntax error in any
262 header after the first one which had an odd number of characters
265 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
266 callout verification, Exim cached a "reject" for the entire domain. This
267 is correct for most verifications, but it is not correct for a recipient
268 verification with use_sender or use_postmaster set, because in that case
269 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
270 case of MAIL FROM:<> rejection from other early rejections (e.g.
271 rejection of HELO). When verifying a recipient using a non-null MAIL
272 address, the cache is ignored if it shows MAIL FROM:<> rejection.
273 Whatever the result of the callout, the value of the domain cache is
274 left unchanged (for any other kind of callout, getting as far as trying
275 RCPT means that the domain itself is ok).
277 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
280 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
281 manifest itself as EPIPE rather than ECONNECT. When tidying away a
282 session, the daemon ignores ECONNECT errors and logs others; it now
283 ignores EPIPE as well.
285 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
286 (quoted-printable decoding).
288 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
289 later the small subsequent patch to fix an introduced bug.
291 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
293 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
295 PH/08 An error is now given if message_size_limit is specified negative.
297 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
298 to be given (somewhat) arbitrary names.
300 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
303 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
304 miscellaneous code fixes
306 PH/10 Added the log_reject_target ACL modifier to specify where to log
309 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
310 hostname. This is wrong, because it relates to the incoming message (and
311 probably the interface on which it is arriving) and not to the outgoing
312 callout (which could be using a different interface). This has been
313 changed to use the value of the helo_data option from the smtp transport
314 instead - this is what is used when a message is actually being sent. If
315 there is no remote transport (possible with a router that sets up host
316 addresses), $smtp_active_hostname is used.
318 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
319 tweaks were necessary in order to get it to work (see also 21 below):
320 (a) The code assumed that strncpy() returns a negative number on buffer
321 overflow, which isn't the case. Replaced with Exim's string_format()
323 (b) There were several signed/unsigned issues. I just did the minimum
324 hacking in of casts. There is scope for a larger refactoring.
325 (c) The code used strcasecmp() which is not a standard C function.
326 Replaced with Exim's strcmpic() function.
327 (d) The code set only $1; it now sets $auth1 as well.
328 (e) A simple test gave the error "authentication client didn't specify
329 service in request". It would seem that Dovecot has changed its
330 interface. Fortunately there's a specification; I followed it and
331 changed what the client sends and it appears to be working now.
333 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
336 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
337 address A is aliased to B and C, where B exists and C does not. Without
338 -v the output is "A verified" because verification stops after a
339 successful redirection if more than one address is generated. However,
340 with -v the child addresses are also verified. Exim was outputting "A
341 failed to verify" and then showing the successful verification for C,
342 with its parentage. It now outputs "B failed to verify", showing B's
343 parentage before showing the successful verification of C.
345 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
346 look up a TXT record in a specific list after matching in a combined
349 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
350 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
351 they consult the DNS. I had assumed they would set it the way they
352 wanted; and indeed my experiments on Linux seem to show that in some
353 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
354 To be on the safe side, however, I have now made the interface to
355 host_find_byname() similar to host_find_bydns(), with an argument
356 containing the DNS resolver options. The host_find_byname() function now
357 sets these options at its start, just as host_find_bydns() does. The smtp
358 transport options dns_qualify_single and dns_search_parents are passed to
359 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
360 of host_find_byname() use the default settings of RES_DEFNAMES
361 (qualify_single) but not RES_DNSRCH (search_parents).
363 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
364 spool_read_header() do less string testing, by means of a preliminary
365 switch on the second character of optional "-foo" lines. (This is
366 overdue, caused by the large number of possibilities that now exist.
367 Originally there were few.) While I was there, I also converted the
368 str(n)cmp tests so they don't re-test the leading "-" and the first
369 character, in the hope this might squeeze out yet more improvement.
371 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
372 flag allowing group syntax was set by the header_syntax check but not
373 turned off, possible causing trouble later; (2) The flag was not being
374 set at all for the header_verify test, causing "group"-style headers to
375 be rejected. I have now set it in this case, and also caused header_
376 verify to ignore an empty address taken from a group. While doing this, I
377 came across some other cases where the code for allowing group syntax
378 while scanning a header line wasn't quite right (mostly, not resetting
379 the flag correctly in the right place). These bugs could have caused
380 trouble for malformed header lines. I hope it is now all correct.
382 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
383 with the "reply" argument non-NULL. The code, however (which originally
384 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
385 but it didn't always do it. This confused somebody who was copying the
386 code for some other use. I have removed all the tests.
388 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
389 feature that was used to support insecure browsers during the U.S. crypto
390 embargo. It requires special client support, and Exim is probably the
391 only MTA that supported it -- and would never use it because real RSA is
392 always available. This code has been removed, because it had the bad
393 effect of slowing Exim down by computing (never used) parameters for the
394 RSA_EXPORT functionality.
396 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
397 authenticator to fail if there's a tab character in the incoming data
398 (there should never be unless someone is messing about, as it's supposed
399 to be base64-encoded). Also added, on Timo's advice, the "secured" option
400 if the connection is using TLS or if the remote IP is the same as the
401 local IP, and the "valid-client-cert option" if a client certificate has
404 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
405 authenticators. This can be used for authorization after authentication
406 succeeds. (In the case of plaintext, it servers for both authentication
409 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
410 if any retry times were supplied.
412 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
413 connection, where there is no client IP address to check. In this
414 situation, the verify now always succeeds.
416 PH/25 Applied John Jetmore's -Mset patch.
418 PH/26 Added -bem to be like -Mset, but loading a message from a file.
420 PH/27 In a string expansion for a processed (not raw) header when multiple
421 headers of the same name were present, leading whitespace was being
422 removed from all of them, but trailing whitespace was being removed only
423 from the last one. Now trailing whitespace is removed from each header
424 before concatenation. Completely empty headers in a concatenation (as
427 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
428 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
430 PH/29 [Removed. This was a change that I later backed out, and forgot to
431 correct the ChangeLog entry (that I had efficiently created) before
432 committing the later change.]
434 PH/30 Exim was sometimes attempting to deliver messages that had suffered
435 address errors (4xx response to RCPT) over the same connection as other
436 messages routed to the same hosts. Such deliveries are always "forced",
437 so retry times are not inspected. This resulted in far too many retries
438 for the affected addresses. The effect occurred only when there were more
439 hosts than the hosts_max_try setting in the smtp transport when it had
440 the 4xx errors. Those hosts that it had tried were not added to the list
441 of hosts for which the message was waiting, so if all were tried, there
442 was no problem. Two fixes have been applied:
444 (i) If there are any address or message errors in an SMTP delivery, none
445 of the hosts (tried or untried) are now added to the list of hosts
446 for which the message is waiting, so the message should not be a
447 candidate for sending over the same connection that was used for a
448 successful delivery of some other message. This seems entirely
449 reasonable: after all the message is NOT "waiting for some host".
450 This is so "obvious" that I'm not sure why it wasn't done
451 previously. Hope I haven't missed anything, but it can't do any
452 harm, as the worst effect is to miss an optimization.
454 (ii) If, despite (i), such a delivery is accidentally attempted, the
455 routing retry time is respected, so at least it doesn't keep
456 hammering the server.
458 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
459 in ${readsocket because some servers need this prod.
461 PH/32 Added some extra debug output when updating a wait-xxx database.
463 PH/33 The hint "could be header name not terminated by colon", which has been
464 given for certain expansion errors for a long time, was not being given
465 for the ${if def:h_colon_omitted{... case.
467 PH/34 The spec says: "With one important exception, whenever a domain list is
468 being scanned, $domain contains the subject domain." There was at least
469 one case where this was not true.
471 PH/35 The error "getsockname() failed: connection reset by peer" was being
472 written to the panic log as well as the main log, but it isn't really
473 panic-worthy as it just means the connection died rather early on. I have
474 removed the panic log writing for the ECONNRESET error when getsockname()
477 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
478 runs only) independently of the message's sender address. This meant
479 that, if the 4xx error was in fact related to the sender, a different
480 message to the same recipient with a different sender could confuse
481 things. In particualar, this can happen when sending to a greylisting
482 server, but other circumstances could also provoke similar problems.
483 I have changed the default so that the retry time for these errors is now
484 based a combination of the sender and recipient addresses. This change
485 can be overridden by setting address_retry_include_sender=false in the
488 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
489 remote server are returned as part of bounce messages. This was not
490 happening for LMTP over a pipe (the lmtp transport), but now it is the
491 same for both kinds of LMTP.
493 PH/38 Despite being documented as not happening, Exim was rewriting addresses
494 in header lines that were in fact CNAMEs. This is no longer the case.
496 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
497 and queue runs started by the daemon processed all messages. This has
498 been fixed so that -R and -S can now usefully be given with -q<time>.
500 PH/40 Import PCRE release 6.7 (fixes some bugs).
502 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
504 PH/42 Give an error if -q is specified more than once.
506 PH/43 Renamed the variables $interface_address and $interface_port as
507 $received_ip_address and $received_port, to make it clear that these
508 values apply to message reception, and not to the outgoing interface when
509 a message is delivered. (The old names remain recognized, of course.)
511 PH/44 There was no timeout on the connect() call when using a Unix domain
512 socket in the ${readsocket expansion. There now is.
514 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
515 be meaningful with "accept".
517 SC/01 Eximstats V1.43
518 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
520 SC/02 Eximstats V1.44
521 Use a glob alias rather than an array ref in the generated
522 parser. This improves both readability and performance.
524 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
525 Collect SpamAssassin and rejection statistics.
526 Don't display local sender or destination tables unless
527 there is data to show.
528 Added average volumes into the top table text output.
530 SC/04 Eximstats V1.46
531 Collect data on the number of addresses (recipients)
532 as well as the number of messages.
534 SC/05 Eximstats V1.47
535 Added 'Message too big' to the list of mail rejection
536 reasons (thanks to Marco Gaiarin).
538 SC/06 Eximstats V1.48
539 Mainlog lines which have GMT offsets and are too short to
540 have a flag are now skipped.
542 SC/07 Eximstats V1.49 (Alain Williams)
543 Added the -emptyok flag.
545 SC/08 Eximstats V1.50
546 Fixes for obtaining the IP address from reject messages.
548 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
549 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
550 whitesspace changes from 4.64-PH/27
552 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
555 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
556 are found, allow negative numbers in numeric criteria)
558 JJ/06 exipick.20061117.2, added new $message_body_missing variable
560 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
561 to match changes made in 4.64-PH/43
563 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
565 PH/47 Put in an explicit test for a DNS lookup of an address record where the
566 "domain" is actually an IP address, and force a failure. This locks out
567 those revolvers/nameservers that support "A-for-A" lookups, in
568 contravention of the specifications.
570 PH/48 When a host name was looked up from an IP address, and the subsequent
571 forward lookup of the name timed out, the host name was left in
572 $sender_host_name, contrary to the specification.
574 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
575 restricted to single-key lookups, Exim was not diagnosing an error if
576 * or *@ was used with a query-style lookup.
578 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
580 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
581 long ago, but noone interested enough thought of it. Let's just say that
582 the "1.1" means that there are some new functions that weren't there at
583 some point in the past.
585 PH/51 Error processing for expansion failure of helo_data from an smtp
586 transport during callout processing was broken.
588 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
589 tested/used via the -bh/-bhc/-bs options.
591 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
592 bug, fixed in subsequent PCRE releases).
594 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
595 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
597 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
603 SC/01 Use a glob alias rather than an array ref in eximstats generated
604 parser. This improves both readability and performance.
606 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
607 Don't display local sender or destination tables in eximstats unless
608 there is data to show.
609 Added average volumes into the eximstats top table text output.
611 SC/03 Collect data on the number of addresses (recipients) as well
612 as the number of messages in eximstats.
614 TF/01 Correct an error in the documentation for the redirect router. Exim
615 does (usually) call initgroups() when daemonizing.
617 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
618 with consistent privilege compared to when running as a daemon.
620 TF/03 Note in the spec that $authenticated_id is not set for local
621 submissions from trusted users.
623 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
624 Thanks to Dean Brooks <dean@iglou.com> for the patch.
626 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
627 by adding some example configuration directives to the default
628 configuration file. A little bit of work is required to uncomment the
629 directives and define how usernames and passwords are checked, but
630 there is now a framework to start from.
632 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
633 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
634 without this. I don't know how relevant this is to other LDAP libraries.
636 PH/02 Add the verb name to the "unknown ACL verb" error.
638 PH/03 Magnus Holmgren's patch for filter_prepend_home.
640 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
642 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
643 directory not expanded when it should be if an expanded home directory
644 was set for the address (which is overridden by the transport).
646 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
649 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
650 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
651 because it is too late at that time, and has no effect.
653 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
654 security issue with \' (bugzilla #107). I could not use the
655 PQescapeStringConn() function, because it needs a PGconn value as one of
658 PH/08 When testing addresses using -bt, indicate those final addresses that
659 are duplicates that would not cause an additional delivery. At least one
660 person was confused, thinking that -bt output corresponded to deliveries.
661 (Suppressing duplicates isn't a good idea as you lose the information
662 about possibly different redirections that led to the duplicates.)
664 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
665 systems where poll() doesn't work, in particular OS X.
667 PH/10 Added more information to debugging output for retry time not reached.
669 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
670 operations in malware.c.
672 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
675 PH/13 If write_rejectlog was set false when logging was sent to syslog with
676 syslog_duplication set false, log lines that would normally be written
677 both the the main log and to the reject log were not written to syslog at
680 PH/14 In the default configuration, change the use of "message" in ACL warn
681 statements to "add_header".
683 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
684 not followed by a command (e.g. "seen endif").
686 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
687 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
690 PH/17 Added extra conditions to the default value of delay_warning_condition
694 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
695 { match{$h_precedence:}{(?i)bulk|list|junk} } \
696 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
699 The Auto-Submitted: and various List- headers are standardised, whereas I
700 don't think Precedence: ever was.
702 PH/18 Refactored debugging code in route_finduser() to show more information,
703 in particular, the error code if getpwnam() issues one.
705 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
706 This is apparently needed in addition to the PH/07 change above to avoid
707 any possible encoding problems.
709 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
710 but not after initializing Perl.
712 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
713 output them only if debugging. By default they are written stderr,
714 apparently, which is not desirable.
716 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
719 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
722 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
724 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
725 authenticated or an ident call has been made. Suppress the default
726 values for $authenticated_id and $authenticated_sender (but permit -oMai
727 and -oMas) when testing with -bh.
729 PH/24 Re-jigged the order of the tests in the default configuration so that the
730 tests for valid domains and recipients precede the DNS black list and CSA
731 tests, on the grounds that those ones are more expensive.
733 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
734 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
735 This bug exists in every version of Exim that I still have, right back to
738 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
739 However, an attempt to turn on case-sensitivity in a regex key by
740 including (?-i) didn't work because the subject string was already
741 lowercased, and the effects were non-intuitive. It turns out that a
742 one-line patch can be used to allow (?-i) to work as expected.
748 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
749 other effects) broke the use of negated acl sub-conditions.
751 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
754 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
755 "Deny" causes Exim to reject the incoming connection with a 554 error.
756 Unfortunately, if there is a major crisis, such as a disk failure,
757 tcp-wrappers gives "deny", whereas what one would like would be some
758 kind of temporary error. A kludge has been added to help with this.
759 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
760 554 error is used if errno is still zero or contains ENOENT (which occurs
761 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
764 PH/03 Add -lutil to the default FreeBSD LIBS setting.
766 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
767 errors. Otherwise a message that provokes a temporary error (when other
768 messages do not) can cause a whole host to time out.
770 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
771 addresses were routed directly to files or pipes from a redirect router.
772 File deliveries just didn't batch; pipe deliveries might have suffered
775 PH/06 A failure to get a lock for a hints database would erroneously always say
776 "Failed to get write lock", even when it was really a read lock.
778 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
779 of 0600. This has been changed to use the value of the lockfile_mode
780 option (which defaults to 0600).
782 PH/08 Applied small patch from the Sieve maintainer.
784 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
785 folder from quota calculations, a direct delivery into this folder messed
786 up the contents of the maildirsize file. This was because the regex was
787 used only to exclude .Trash (or whatever) when the size of the mailbox
788 was calculated. There was no check that a delivery was happening into an
789 excluded directory. This bug has been fixed by ignoring all quota
790 processing for deliveries into excluded directories.
792 PH/10 Added the maildirfolder_create_regex option to appendfile.
798 PH/01 The code for finding all the local interface addresses on a FreeBSD
799 system running IPv6 was broken. This may well have applied to all BSD
800 systems, as well as to others that have similar system calls. The broken
801 code found IPv4 interfaces correctly, but gave incorrect values for the
802 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
803 that it would not match correctly against @[] and not recognize the IPv6
806 PH/02 The ipliteral router was not recognizing addresses of the form user@
807 [ipv6:....] because it didn't know about the "ipv6:" prefix.
809 PH/03 Added disable_ipv6.
811 PH/04 Changed $reply_address to use the raw form of the headers instead of the
812 decoded form, because it is most often used to construct To: headers
813 lines in autoreplies, and the decoded form may well be syntactically
814 invalid. However, $reply_address has leading white space removed, and all
815 newlines turned into spaces so that the autoreply transport does not
818 PH/05 If group was specified without a user on a router, and no group or user
819 was specified on a transport, the group from the router was ignored.
821 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
822 for visible compile-time settings that can be used to change these
823 numbers, for those that want even more. Backwards compatibility with old
824 spool files has been maintained. However, going back to a previous Exim
825 release will lost any variables that are in spool files.
827 PH/07 Two small changes when running in the test harness: increase delay when
828 passing a TCP/IP connection to a new process, in case the original
829 process has to generate a bounce, and remove special handling of
830 127.0.0.2 (sic), which is no longer necessary.
832 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
833 be the same on different OS.
835 PH/09 Moved a debug statement in filter processing to avoid a race problem when
838 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
839 whether --show-vars was specified or not
841 JJ/02 exipick: Added support for new ACL variable spool format introduced
844 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
845 syntactically invalid From: or Reply-to: line, and a filter used this to
846 generate an autoreply, and therefore failed to obtain an address for the
847 autoreply, Exim could try to deliver to a non-existent relative file
848 name, causing unrelated and misleading errors. What now happens is that
849 it logs this as a hard delivery error, but does not attempt to create a
852 PH/11 The exinext utility has a -C option for testing purposes, but although
853 the given file was scanned by exinext itself; it wasn't being passed on
854 when Exim was called.
856 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
857 an end-of-file indication when reading a command response.
859 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
860 compiled. In many other places in Exim, IPv6 addresses are always
861 recognized, so I have changed this. It also means that IPv4 domain
862 literals of the form [IPV4:n.n.n.n] are now always recognized.
864 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
865 used if the router is not running as root, for example, when verifying at
866 ACL time, or when using -bh. The debugging output from this situation was
867 non-existent - all you got was a failure to exec. I have made two
870 (a) Failures to set uid/gid, the current directory, or a process leader
871 in a subprocess such as that created by queryprogram now generate
872 suitable debugging ouput when -d is set.
874 (b) The queryprogram router detects when it is not running as root,
875 outputs suitable debugging information if -d is set, and then runs
876 the subprocess without attempting to change uid/gid.
878 PH/15 Minor change to Makefile for building test_host (undocumented testing
881 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
882 additional section of a DNS packet that returns MX or SRV records.
883 Instead, it always explicitly searches for A/AAAA records. This avoids
884 major problems that occur when a DNS server includes only records of one
885 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
886 fixed another bug: if SRV records were looked up and the corresponding
887 address records were *not* found in the additional section, the port
888 values from the SRV records were lost.
890 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
891 using the correct key (the original address) when searching the retry
892 rules in order to find which one to use for generating the retry hint.
894 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
895 adding the default one. Similarly, if it contains a Reply-To: header, the
896 errors_reply_to option, if set, is not used.
898 PH/19 When calculating a retry time, Exim used to measure the "time since
899 failure" by looking at the "first failed" field in the retry record. Now
900 it does not use this if it is later than than the arrival time of the
901 message. Instead it uses the arrival time. This makes for better
902 behaviour in cases where some deliveries succeed, thus re-setting the
903 "first failed" field. An example is a quota failure for a huge message
904 when small messages continue to be delivered. Without this change, the
905 "time since failure" will always be short, possible causing more frequent
906 delivery attempts for the huge message than are intended.
907 [Note: This change was subsequently modified - see PH/04 for 4.62.]
909 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
910 $1, $2, $3) because the numerical variables can be reset during some
911 expansion items (e.g. "match"), thereby losing the authentication data.
913 PH/21 Make -bV show the size of off_t variables so that the test suite can
914 decide whether to run tests for quotas > 2G.
916 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
917 mailbox_size, and mailbox_filecount in the appendfile transport. If a
918 filecount value is greater than 2G or if a quota value is greater than 2G
919 on a system where the size of off_t is not greater than 4, a panic error
922 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
923 never match. The debug and -bh output now contains an explicit error
924 message indicating a malformed IPv4 address or mask.
926 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
927 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
930 PH/25 Do not write to syslog when running in the test harness. The only
931 occasion when this arises is a failure to open the main or panic logs
932 (for which there is an explicit test).
934 PH/26 Added the /no_tell option to "control=freeze".
936 PH/27 If a host name lookup failed very early in a connection, for example, if
937 the IP address matched host_lookup and the reverse lookup yielded a name
938 that did not have a forward lookup, an error message of the form "no IP
939 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
940 could be logged. Now it outputs the IP address instead of "NULL".
942 PH/28 An enabling patch from MH: add new function child_open_exim2() which
943 allows the sender and the authenticated sender to be set when
944 submitting a message from within Exim. Since child_open_exim() is
945 documented for local_scan(), the new function should be too.
947 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
948 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
949 results in an empty string is now treated as unset.
951 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
953 PH/31 Added sender_verify_fail logging option.
955 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
956 needed by RFC 822 but not by RFC 2822 was commented out. I have now
957 tidied the source and removed it altogether.
959 PH/33 When a queue run was abandoned because the load average was too high, a
960 log line was always written; now it is written only if the queue_run log
961 selector is set. In addition, the log line for abandonment now contains
962 information about the queue run such as the pid. This is always present
963 in "start" and "stop" lines but was omitted from the "abandon" line.
965 PH/34 Omit spaces between a header name and the colon in the error message that
966 is given when verify = headers_syntax fails (if there are lots of them,
967 the message gets confusing).
969 PH/35 Change the default for dns_check_names_pattern to allow slashes within
970 names, as there are now some PTR records that contain slashes. This check
971 is only to protect against broken name servers that fall over on strange
972 characters, so the fact that it applies to all lookups doesn't matter.
974 PH/36 Now that the new test suite is complete, we can remove some of the
975 special code in Exim that was needed for the old test suite. For example,
976 sorting DNS records because real resolvers return them in an arbitrary
977 order. The new test suite's fake resolver always returns records in the
980 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
981 bounces) except when queue_only is set, to avoid logging races between
982 the different processes.
984 PH/38 Panic-die if .include specifies a non-absolute path.
986 PH/39 A tweak to the "H" retry rule from its user.
988 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
989 a label. They prevented compilation on older perls.
991 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
992 a warning to be raised on newish perls.
994 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
995 on queue. Changes to match documented behaviour of showing count of
996 messages matching specified criteria.
998 PH/40 Changed the default ident timeout from 30s to 5s.
1000 PH/41 Added support for the use of login_cap features, on those BSD systems
1001 that have them, for controlling the resources used by pipe deliveries.
1003 PH/42 The content-scanning code uses fopen() to create files in which to put
1004 message data. Previously it was not paying any attention to the mode of
1005 the files. Exim runs with umask(0) because the rest of the code creates
1006 files with open(), and sets the required mode explicitly. Thus, these
1007 files were ending up world-writeable. This was not a big issue, because,
1008 being within the spool directory, they were not world-accessible. I have
1009 created a function called modefopen, which takes an additional mode
1010 argument. It sets umask(777), creates the file, chmods it to the required
1011 mode, then resets the umask. All the relevant calls to fopen() in the
1012 content scanning code have been changed to use this function.
1014 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1015 to 24 hours. This avoids potential overflow problems when processing G
1016 and H retry rules. I suspect nobody ever tinkers with this value.
1018 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1020 PH/45 When the plaintext authenticator is running as a client, the server's
1021 challenges are checked to ensure they are valid base64 strings. By
1022 default, the authentication attempt is cancelled if an invalid string is
1023 received. Setting client_ignore_invalid_base64 true ignores these errors.
1024 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1025 they are received. Thus, the responses can be made to depend on the
1026 challenges. If an invalid string is ignored, an empty string is placed in
1029 PH/46 Messages that are created by the autoreply transport now contains a
1030 References: header, in accordance with RFCs 2822 and 3834.
1032 PH/47 Added authenticated_sender_force to the smtp transport.
1034 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1036 PH/49 Installed latest patch from the Sieve maintainer.
1038 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1039 was also set, the appendfile transport was unnecessarily scanning a
1040 directory of message files (e.g. for maildir delivery) to find the count
1041 of files (along with the size), even though it did not need this
1042 information. It now does the scan only if it needs to find either the
1043 size of the count of files.
1045 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1047 PH/52 Two bugs concerned with error handling when the smtp transport is
1050 (i) Exim was not creating retry information for temporary errors given
1051 for individual recipients after the DATA command when the smtp transport
1052 was used in LMTP mode. This meant that they could be retried too
1053 frequently, and not timed out correctly.
1055 (ii) Exim was setting the flag that allows error details to be returned
1056 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1057 recipients that were returned after the DATA command.
1059 PH/53 This is related to PH/52, but is more general: for any failing address,
1060 when detailed error information was permitted to be returned to the
1061 sender, but the error was temporary, then after the final timeout, only
1062 "retry timeout exceeded" was returned. Now it returns the full error as
1063 well as "retry timeout exceeded".
1065 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1066 do this, and (what is worse) MTAs that accept it.
1068 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1069 will now be deprecated.
1071 PH/56 New os.c-cygwin from the Cygwin maintainer.
1073 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1074 formats (previously only available in exim formats via -bpr, -bpru,
1075 and -bpra. Now also available in native and exiqgrep formats)
1077 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1078 with very large, slow to parse queues
1080 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1082 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1084 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1085 responses to authentication challenges, though it was showing the
1086 challenges; (ii) I've removed the CR characters from the debug output for
1089 PH/58 Allow for the insertion of a newline as well as a space when a string
1090 is turned into more than one encoded-word during RFC 2047 encoding. The
1091 Sieve code now uses this.
1093 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1094 data_4xx, lost_connection, tls_required.
1096 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1097 message was being sent as an SMTP response.
1099 PH/61 Add -l and -k options to exicyclog.
1101 PH/62 When verifying, if an address was redirected to one new address, so that
1102 verification continued, and the new address failed or deferred after
1103 having set something in $address_data, the value of $address_data was not
1104 passed back to the ACL. This was different to the case when no
1105 redirection occurred. The value is now passed back in both cases.
1107 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1108 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1109 use login_cap.h, so on its own it isn't the distinguishing feature. The
1110 new name refers directly to the setclassresources() function.
1112 PH/65 Added configuration files for NetBSD3.
1114 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1116 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1117 is preferred over IPv4.
1119 PH/68 The bounce_return_message and bounce_return_body options were not being
1120 honoured for bounces generated during the reception of non-SMTP messages.
1121 In particular, this applied to messages rejected by the ACL. This bug has
1122 been fixed. However, if bounce_return_message is true and bounce_return_
1123 body is false, the headers that are returned for a non-SMTP message
1124 include only those that have been read before the error was detected.
1125 (In the case of an ACL rejection, they have all been read.)
1127 PH/69 The HTML version of the specification is now built in a directory called
1128 spec_html instead of spec.html, because the latter looks like a path with
1129 a MIME-type, and this confuses some software.
1131 PH/70 Catch two compiler warnings in sieve.c.
1133 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1134 function verify_get_ident() calls ip_connect() to connect a socket, but
1135 if the "connect()" function timed out, ip_connect() used to close the
1136 socket. However, verify_get_ident() also closes the socket later, and in
1137 between Exim writes to the log, which may get opened at this point. When
1138 the socket was closed in ip_connect(), the log could get the same file
1139 descriptor number as the socket. This naturally causes chaos. The fix is
1140 not to close the socket in ip_connect(); the socket should be closed by
1141 the function that creates it. There was only one place in the code where
1142 this was missing, in the iplookup router, which I don't think anybody now
1143 uses, but I've fixed it anyway.
1145 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1146 well as to direct DNS lookups. Otherwise the handling of names in host
1147 lists is inconsistent and therefore confusing.
1153 PH/01 Two changes to the default runtime configuration:
1155 (1) Move the checks for relay_from_hosts and authenticated clients from
1156 after to before the (commented out) DNS black list checks.
1158 (2) Add control=submission to the relay_from_hosts and authenticated
1159 clients checks, on the grounds that messages accepted by these
1160 statements are most likely to be submissions.
1162 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1164 (1) Generate an error if the third argument for the ${prvs expansion is
1167 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1170 (3) Reset the variables that are obtained from the first argument of
1171 ${prvscheck and used in the second argument before leaving the code,
1172 because their memory is reclaimed, so using them afterwards may do
1175 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1176 one (it's much easier than Tom thought :-).
1178 (5) Because of (4), we can now allow for the use of $prvscheck_result
1179 inside the third argument.
1181 PH/03 For some reason, the default setting of PATH when running a command from
1182 a pipe transport was just "/usr/bin". I have changed it to
1185 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1186 anything to be listed in the output from -bV.
1188 PH/05 When a filter generated an autoreply, the entire To: header line was
1189 quoted in the delivery log line, like this:
1191 => >A.N.Other <ano@some.domain> <original@ddress> ...
1193 This has been changed so that it extracts the operative address. There
1194 may be more than one such address. If so, they are comma-separated, like
1197 => >ano@some.domain,ona@other.domain <original@ddress> ...
1199 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1200 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1201 not being looked up in the rDNS to get a host name, Exim was showing the
1202 IP address twice in Received: lines, even though the IP addresses were
1203 identical. For example:
1205 Received: from [1.2.3.4] (helo=[1.2.3.4])
1207 However, if the real host name was known, it was omitting the HELO data
1208 if it matched the actual IP address. This has been tidied up so that it
1209 doesn't show the same IP address twice.
1211 PH/07 When both +timestamp and +memory debugging was on, the value given by
1212 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1213 called by the debug printing, thereby overwriting the timestamp buffer.
1214 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1216 PH/08 When the original message was included in an autoreply transport, it
1217 always said "this is a copy of the message, including all the headers",
1218 even if body_only or headers_only was set. It now gives an appropriate
1221 PH/09 Applied a patch from the Sieve maintainer which:
1223 o fixes some comments
1224 o adds the (disabled) notify extension core
1225 o adds some debug output for the result of if/elsif tests
1226 o points to the current vacation draft in the documentation
1227 and documents the missing references header update
1231 o fixes a bug in processing the envelope test (when testing
1232 multiple envelope elements, the last element determinted the
1235 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1236 Electronic Mail") by including:
1238 Auto-submitted: auto-generated
1240 in the messages that it generates (bounce messages and others, such as
1241 warnings). In the case of bounce messages for non-SMTP mesages, there was
1242 also a typo: it was using "Auto_submitted" (underscore instead of
1243 hyphen). Since every message generated by Exim is necessarily in response
1244 to another message, thes have all been changed to:
1246 Auto-Submitted: auto-replied
1248 in accordance with these statements in the RFC:
1250 The auto-replied keyword:
1252 - SHOULD be used on messages sent in direct response to another
1253 message by an automatic process,
1255 - MUST NOT be used on manually-generated messages,
1257 - MAY be used on Delivery Status Notifications (DSNs) and Message
1258 Disposition Notifications (MDNs),
1260 - MUST NOT be used on messages generated by automatic or periodic
1261 processes, except for messages which are automatic responses to
1264 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1265 to the default Received: header definition.
1267 PH/12 Added log selector acl_warn_skipped (default on).
1269 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1270 variables because (a) they are in the wrong storage pool and (b) even if
1271 they were copied, it wouldn't work properly because of the caching.
1273 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1274 checking when decoding. Apparently there are clients that generate
1275 overlong encoded strings. Why am I not surprised?
1277 PH/15 If the first argument of "${if match_address" was not empty, but did not
1278 contain an "@" character, Exim crashed. Now it writes a panic log message
1279 and treats the condition as false.
1281 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1283 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1284 extension "envelope-auth". The code is finished and in agreement with
1285 other implementations, but there is no documentation so far and in fact,
1286 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1287 not changing the active code.
1289 Print executed "if" and "elsif" statements when debugging is used. This
1290 helps a great deal to understand what a filter does.
1292 Document more things not specified clearly in RFC3028. I had all this
1293 sorted out, when out of a sudden new issues came to my mind. Oops."
1295 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1298 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1299 canonical form (as documented). However, after a host name lookup from
1300 the IP address, check_host() was doing a simple string comparison with
1301 addresses acquired from the DNS when checking that the found name did
1302 have the original IP as one of its addresses. Since any found IPv6
1303 addresses are likely to be in abbreviated form, the comparison could
1304 fail. Luckily, there already exists a function for doing the comparison
1305 by converting both addresses to binary, so now that is used instead of
1306 the text comparison.
1308 PH/20 There was another similar case to PH/19, when a complete host name was
1309 given in a host list; looking up its IP address could give an abbreviated
1310 form, whereas the current host's name might or might not be abbreviated.
1311 The same fix has been applied.
1317 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1318 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1321 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1322 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1324 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1326 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1327 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1328 Also, for "Auto-Submitted", treat anything other than "no" as
1329 non-personal, in accordance with RFC 3834. (Previously it treated
1330 anything starting "auto-" as non-personal.)
1332 TF/01 The control=submission/name=... option had a problem with syntax
1333 errors if the name included a slash character. The /name= option
1334 now slurps the rest of the string, so it can include any characters
1335 but it must come last in the list of options (after /sender_retain
1338 PH/05 Some modifications to the interface to the fake nameserver for the new
1346 TK/01 Added the "success_on_redirect" address verification option. See
1347 NewStuff for rationale and an example.
1349 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1351 PH/02 Patch to exigrep to allow it to work on syslog lines.
1353 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1354 fread() to skip over the body file's header line, because in Cygwin the
1355 header line is locked and is inaccessible.
1357 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1358 co-exist for some time) to make it clear that it is the Exim ID that is
1359 referenced, not the Message-ID: header line.
1361 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1362 string_format() function, because snprintf() does not exist on all
1365 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1366 ${stat: expansion item.
1368 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1369 protocol synchronization error", to keep the pedants happy.
1371 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1372 well as for IRIX systems, when gcc is being used. See the host.c source
1375 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1377 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1380 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1381 transport and to the smtp transport in LMTP mode.
1383 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1385 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1386 run a filter in a subprocess. This could lead to confusion in subsequent
1387 lookups in the parent process. There should also be a search_tidyup() at
1388 the end of the subprocess.
1390 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1391 only if the host matched helo_try_verify_hosts, which caused the
1392 verification to occur when the EHLO/HELO command was issued. The ACL just
1393 tested the remembered result. Now, if a previous verification attempt has
1394 not happened, "verify = helo" does it there and then.
1396 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1398 TK/03 Fix log output including CR from clamd.
1400 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1401 exist provoked a memory error which could cause a segfault.
1403 PH/15 Installed PCRE 6.2
1405 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1407 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1408 of the problem. Specifically, suggested +O2 rather than +O1 for the
1411 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1413 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1414 record, the port was not correctly logged when the outgoing_port log
1415 selector was set (it logged the transort's default port).
1417 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1418 fallback_hosts, and "hosts" in the smtp transport.
1420 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1421 host errors such as "Connection refused".
1423 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1424 authentication with radiusclient 0.4.9:
1426 - Error returned from rc_read_config was caught wrongly
1427 - Username/password not passed on to radius server due to wrong length.
1429 The presumption is that some radiusclient API changes for 4.51/PH/17
1430 were not taken care of correctly. The code is still untested by me (my
1431 Linux distribution still has 0.3.2 of radiusclient), but it was
1432 contributed by a Radius user.
1434 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1435 expanding the "port" option of the smtp transport.
1437 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1438 while reading a MIME header. Thanks to Tom Hughes for a patch.
1440 PH/24 Include config.h inside local_scan.h so that configuration settings are
1443 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1444 that in an ACL for RCPT (for example), you can examine exactly what was
1447 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1448 commands, but it was not correctly comparing the address with the actual
1449 client host address. Thus, it would show the EHLO address in Received:
1450 header lines when this was not necessary.
1452 PH/27 Added the % operator to ${eval:}.
1454 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1455 it should be ignoring failures (because with -C, for example, it has lost
1456 privilege). It wasn't ignoring creation failures other than "already
1459 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1462 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1463 because an input line was too long, either on its own, or by virtue of
1464 too many continuations, the temporary file was not being removed, and the
1465 return code was incorrect.
1467 PH/31 Missing "BOOL" in function definition in filtertest.c.
1469 PH/32 Applied Sieve patches from the maintainer.
1471 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1473 PH/33 Added "verify = not_blind".
1475 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1476 Local/Makefile (with some defaults set). These are used in built scripts
1477 such as exicyclog, but they have never been used in the exim_install
1478 script (though there are many overriding facilities there). I have
1479 arranged that the exim_install script now takes note of these two
1482 PH/35 Installed configuration files for Dragonfly.
1484 PH/36 When a locally submitted message by a trusted user did not contain a
1485 From: header, and the sender address was obtained from -f or from an SMTP
1486 MAIL command, and the trusted user did not use -F to supply a sender
1487 name, $originator_name was incorrectly used when constructing a From:
1488 header. Furthermore, $originator_name was used for submission mode
1489 messages from external hosts without From: headers in a similar way,
1490 which is clearly wrong.
1492 PH/37 Added control=suppress_local_fixups.
1494 PH/38 When log_selector = +received_sender was set, and the addition of the
1495 sender made the log line's construction buffer exactly full, or one byte
1496 less than full, an overflow happened when the terminating "\n" was
1499 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1500 when the result of a list match is failure because a DNS lookup failed.
1502 PH/40 RM_COMMAND is now used in the building process.
1504 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1505 the "build-* directories that it finds.
1507 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1508 address in a domain literal was a prefix of an interface address.
1510 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1511 when verifying a sender address, unless rewrite_headers is false.
1513 PH/44 Wrote a long comment about why errors_to addresses are verified as
1514 recipients, not senders.
1516 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1517 the ratelimit ACL was added.
1519 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1521 PH/46 Added extra information about PostgreSQL errors to the error string.
1523 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1524 avoiding the need to install special zones in a real server. This is
1525 backwards compatible; if it can't find the fake resolver, it drops back.
1526 Thus, both old and new test suites can be run.
1528 TF/02 Added util/ratelimit.pl
1530 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1531 clock is set back in time.
1533 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1534 Candler <B.Candler@pobox.com>.
1536 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1537 OK for addresses that are the result of redirections.
1539 PH/48 A number of further additions for the benefit of the new test suite,
1540 including a fake gethostbyname() that interfaces to the fake DNS resolver
1543 TF/06 The fix for widen_domains has also been applied to qualify_single and
1544 search_parents which are the other dnslookup options that can cause
1547 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1550 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1552 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1553 service" if it encountered an MX record with an empty target hostname.
1554 The message is now "an MX or SRV record indicated no SMTP service".
1556 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1557 if the DNS of the sending site is misconfigured. This is quite a
1558 common situation. This change restores the behaviour of treating a
1559 helo verification defer as a failure.
1561 PH/51 If self=fail was set on a router, the bounce message did not include the
1562 actual error message.
1568 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1570 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1571 command itself ended in error, the underlying message about the transport
1572 filter timeout was being overwritten with the pipe command error. Now the
1573 underlying error message should be appended to the second error message.
1575 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1577 PH/02 Reduce the amount of output that "make" produces by default. Full output
1578 can still be requested.
1580 PH/03 The warning log line about a condition test deferring for a "warn" verb
1581 was being output only once per connection, rather than after each
1582 occurrence (because it was using the same function as for successful
1583 "warn" verbs). This seems wrong, so I have changed it.
1585 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1586 it should not have, which might have caused a crash in the right
1587 circumstances, but probably never did.
1589 PH/04 Installed a modified version of Tony Finch's patch to make submission
1590 mode fix the return path as well as the Sender: header line, and to
1591 add a /name= option so that you can make the user's friendly name appear
1594 TF/03 Added the control = fakedefer ACL modifier.
1596 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1597 Mark Lowes for thorough testing.
1599 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1601 TK/03 Merged latest SRS patch from Miles Wilton.
1603 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1604 with the definition in sysexits.h (which is #included earlier).
1605 Fortunately, Exim does not actually use EX_OK. The code used to try to
1606 preserve the sysexits.h value, by assumimg that macro definitions were
1607 scanned for macro replacements. I have been disabused of this notion,
1608 so now the code just undefines EX_OK before #including unistd.h.
1610 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1611 in the smtp transport. When a block could not be written in a single
1612 write() function, the timeout was being re-applied to each part-write.
1613 This seems wrong - if the receiver was accepting one byte at a time it
1614 would take for ever. The timeout is now adjusted when this happens. It
1615 doesn't have to be particularly precise.
1617 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1618 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1620 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1621 without a domain if the check to <postmaster@domain> fails.
1623 SC/01 Eximstats: added -xls and the ability to specify output files
1624 (patch written by Frank Heydlauf).
1626 SC/02 Eximstats: use FileHandles for outputing results.
1628 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1630 SC/04 Eximstats: fixed display of large numbers with -nvr option
1632 SC/05 Eximstats: fixed merging of reports with empty tables.
1634 SC/06 Eximstats: added the -include_original_destination flag
1636 SC/07 Eximstats: removed tabs and trailing whitespace.
1638 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1640 TK/06 MBOX spool code: Add real "From " MBOX separator line
1641 so the .eml file is really in mbox format (even though
1642 most programs do not really care). Patch from Alex Miller.
1644 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1645 The latter is generated from $received_to and is only set if the
1646 message has one envelope recipient. SA can use these headers,
1647 obviously out-of-the-box. Patch from Alex Miller.
1649 PH/08 The ${def test on a variable was returning false if the variable's
1650 value was "0", contrary to what the specification has always said!
1651 The result should be true unless the variable is empty.
1653 PH/09 The syntax error of a character other than { following "${if
1654 def:variable_name" (after optional whitespace) was not being diagnosed.
1655 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1656 accidental colon was present, for example, could give incorrect results.
1658 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1659 result is used (not including appendfile, where other changes are about
1662 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1663 This involved changing a lot of size variables from int to off_t. It
1664 should work with maildirs and everything.
1666 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1667 spamd dying while we are connected to it.
1669 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1672 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1673 function for BDB 4.3.
1675 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1677 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1678 the sources for PCRE have been re-arranged and more files are now
1681 PH/15 The code I had for printing potentially long long variables in PH/11
1682 above was not the best (it lost precision). The length of off_t variables
1683 is now inspected at build time, and an appropriate printing format (%ld
1684 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1685 to be "long long int" or "long int". This is needed for the internal
1686 formatting function string_vformat().
1688 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1689 the configuration file to be ":syslog", then the script "guesses" where
1690 the logs files are, rather than using the compiled in default. In our
1691 case the guess is not the same as the compiled default, so the script
1692 suddenly stopped working when I started to use syslog. The patch checks
1693 to see if log_file_path is "". If so, it attempts to read it from exim
1694 with no configuration file to get the compiled in version, before it
1695 falls back to the previous guessing code."
1697 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1698 implementing BATV in an Exim configuration. See NewStuff for the gory
1701 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1702 Makefile that are specific to HP-UX.
1704 PH/18 If the "use_postmaster" option was set for a recipient callout together
1705 with the "random" option, the postmaster address was used as the MAIL
1706 FROM address for the random test, but not for the subsequent recipient
1707 test. It is now used for both.
1709 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1710 patch removes a few documentation additions to RFC 3028, because the
1711 latest draft now contains them. It adds the new en;ascii-case comparator
1712 and a new error check for 8bit text in MIME parts. Comparator and
1713 require names are now matched exactly. I enabled the subaddress
1714 extension, but it is not well tested yet (read: it works for me)."
1716 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1717 rework some of the code of TK/09 above to avoid the hardwired use of
1718 "%lld" and "long long". Replaced the call to snprintf() with a call to
1721 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1722 records point to non-existent hosts", "retry timeout exceeded", and
1723 "retry time not reached for any host after a long failure period".
1725 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1726 experimental DomainKeys support:
1728 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1729 (2) On an error such as an illegally used "control", the wrong name for
1730 the control was given.
1732 These problems did NOT occur unless DomainKeys support was compiled.
1734 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1736 PH/24 Added ${if match_ip condition.
1738 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1739 databases so that it will be absolutely obvious if a crash occurs in the
1740 DB library. This is a regular occurrence (often caused by mis-matched
1743 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1744 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1745 picked up on a user's system that detects such things. There doesn't seem
1746 to be a gcc warning option for this - only an attribute that has to be
1747 put on the function's prototype. It seems that in Fedora Core 4 they have
1748 set this on a number of new functions. No doubt there will be more in due
1751 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1752 specify a transport. However, if an address that was verified by such a
1753 router was the subject of a callout, Exim crashed because it tried to
1754 read the rcpt_include_affixes from the non-existent transport. Now it
1755 just assumes that the setting of that option is false. This bug was
1756 introduced by 4.51/PH/31.
1758 PH/28 Changed -d+all to exclude +memory, because that information is very
1759 rarely of interest, but it makes the output a lot bigger. People tend to
1760 do -d+all out of habit.
1762 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1763 code in os-type was giving problems when libc.so lives in lib64, like on
1766 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1767 aren't the modern standard, and it seems that some systems' include files
1768 don't always have them. Exim was already checking for some of the newer
1769 ones like T_AAAA, and defining it itself. I've added checks for all the
1770 record types that Exim uses.
1772 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1773 not automatically generating a new one, as it is supposed to. This
1774 prevented TLS from working. If the file did exist, but contained invalid
1775 data, a new version was generated, as expected. It was only the case of a
1776 non-existent file that was broken.
1778 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1779 with a change in libdomainkeys > 0.64.
1781 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1782 from DNS. If the selector record carries the flag, it now has
1783 precedence over the domain-wide flag.
1785 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1787 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1788 the use of a port name that isn't defined in /etc/services) occurred, the
1789 message was deferred as in a normal delivery, and thus remained on the
1790 spool, instead of being failed because of the mua_wrapper setting. This
1791 is now fixed, and I tidied up some of the mua_wrapper messages at the
1794 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1795 the messages in a hash of arrays rather than using individual hashes.
1796 This is a bit cleaner and results in dramatic memory savings, albeit
1797 at a slight CPU cost.
1799 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1800 as requested by Marc Sherman.
1802 SC/10 Eximstats: added histograms for user specified patterns as requested
1805 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1807 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1808 fopen() in the content-scanning modules that did not already have it.
1814 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1815 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1817 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1819 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1821 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1822 to test Sieve filters that use "vacation".
1824 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1825 that changes the way the GnuTLS parameters are stored in the cache file.
1826 The new format can be generated externally. For backward compatibility,
1827 if the data in the cache doesn't make sense, Exim assumes it has read an
1828 old-format file, and it generates new data and writes a new file. This
1829 means that you can't go back to an older release without removing the
1832 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1833 work if there are any delivery delays because "one_time" forces the
1834 parent to be marked "delivered", so its unseen clone is never tried
1835 again. For this reason, Exim now forbids the simultaneous setting of
1838 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1839 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1840 entry if you want to know the details. The fix, however, neglected to
1841 consider the case when local delivery batching is involved. The test for
1842 "previously delivered" was not happening when checking to see if an
1843 address could be batched with a previous (undelivered) one; under
1844 certain circumstances this could lead to multiple deliveries to the same
1847 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1848 in its include files, and this causes problems building Exim.
1850 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1851 header_syntax) but Exim was just ignoring anything given after a slash.
1852 In particular, this caused confusion with an attempt to use "verify =
1853 reverse_host_lookup/defer_ok". An error is now given when options are
1854 supplied for verify items that do not have them. (Maybe reverse_host_
1855 lookup should have a defer_ok option, but that's a different point.)
1857 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1858 defined by RFC 821) to 2048, because there were problems with some AUTH
1859 commands, and RFC 1869 says the size should be increased for extended
1860 SMTP commands that take arguments.
1862 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1865 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1866 "unknown" error; now it says that the functionality isn't in the binary.
1868 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1869 an address' error message when a string expansion fails (syntax or
1870 whatever). Otherwise the password may appear in the log. Following change
1871 PH/42 below, there is no longer a chance of it appearing in a bounce
1874 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1876 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1877 of its addresses was ever tried. (Bugzilla bug #2.)
1879 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1880 the result incorrectly in the debug output. (It correctly added a newline
1881 to what was transported.)
1883 TF/01 Added $received_time.
1885 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1886 commented out examples of how to interface to a virus scanner and to
1887 SpamAssassin. Also added commented examples of av_scanner and
1888 spamd_address settings.
1890 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1891 and controls are allowed in which ACLs. There were a couple of minor
1892 errors. Some of the entries in the conditions table (which is a table of
1893 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1894 negation of where the condition IS allowed.
1896 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1898 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1899 header file does not have a version number, so I've had to invent a new
1900 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1901 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1902 radiusclient), but it was contributed by a Radius user.
1904 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1905 files or directories, for OpenSSL.
1907 PH/19 When an Exim process that is running as root has to create an Exim log
1908 file, it does so in a subprocess that runs as exim:exim so as to get the
1909 ownership right at creation (otherwise, other Exim processes might see
1910 the file with the wrong ownership). There was no test for failure of this
1911 fork() call, which would lead to the process getting stuck as it waited
1912 for a non-existent subprocess. Forks do occasionally fail when resources
1913 run out. I reviewed all the other calls to fork(); they all seem to check
1916 PH/20 When checking for unexpected SMTP input at connect time (before writing
1917 the banner), Exim was not dealing correctly with a non-positive return
1918 from the read() function. If the client had disconnected by this time,
1919 the result was a log entry for a synchronization error with an empty
1920 string after "input=" when read() returned zero. If read() returned -1
1921 (an event I could not check), uninitialized data bytes were printed.
1922 There were reports of junk text (parts of files, etc) appearing after
1925 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1927 PH/22 Added support for macro redefinition, and (re)definition in between
1928 driver and ACL definitions.
1930 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1931 forgetting to use the resulting value; it was using the unexpanded value.
1933 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1934 hadn't been configured. The fix is from Juergen Kreileder, who
1935 understands it better than I do:
1937 "Here's what I see happening with three configured cyrus_sasl
1938 authenticators configured (plain, login, cram-md5):
1940 On startup auth_cyrus_sasl_init() gets called for each of these.
1941 This means three calls to sasl_listmech() without a specified mech_list.
1942 => SASL tests which mechs of all available mechs actually work
1943 => three warnings about OTP not working
1944 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1946 With the patch, sasl_listmech() also gets called three times. But now
1947 SASL's mech_list option is set to the server_mech specified in the the
1948 authenticator. Or in other words, the answer from sasl_listmech()
1949 gets limited to just the mech you're testing for (which is different
1951 => the return list contains just 'plain' or 'login', 'cram-md5' or
1952 nothing depending on the value of ob->server_mech.
1954 I've just tested the patch: Authentication still works fine,
1955 unavailable mechs specified in the exim configuration are still
1956 caught, and the auth.log warnings about OTP are gone."
1958 PH/25 When debugging is enabled, the contents of the command line are added
1959 to the debugging output, even when log_selector=+arguments is not
1962 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1963 answer is "GNU", and only if the return is "GNU/something" is the answer
1966 PH/27 $acl_verify_message is now set immediately after the failure of a
1967 verification in an ACL, and so is available in subsequent modifiers. In
1968 particular, the message can be preserved by coding like this:
1970 warn !verify = sender
1971 set acl_m0 = $acl_verify_message
1973 Previously, $acl_verify_message was set only while expanding "message"
1974 and "log_message" when a very denied access.
1976 PH/28 Modified OS/os.c-Linux with
1978 -#ifndef OS_LOAD_AVERAGE
1979 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1981 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1982 nomenclature these days.)
1984 PH/29 Installed patch from the Sieve maintainer that adds the options
1985 sieve_useraddress and sieve_subaddress to the redirect router.
1987 PH/30 In these circumstances:
1988 . Two addresses routed to the same list of hosts;
1989 . First host does not offer TLS;
1990 . First host accepts first address;
1991 . First host gives temporary error to second address;
1992 . Second host offers TLS and a TLS session is established;
1993 . Second host accepts second address.
1994 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1995 and peerdn, if requested) that were in fact used only for the second
1998 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1999 attention to any local part prefix or suffix that was matched by the
2000 router that accepted the address. It now behaves in the same way as it
2001 does for delivery: the affixes are removed from the local part unless
2002 rcpt_include_affixes is set on the transport.
2004 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2005 timeout during the DATA phase of an incoming message.
2007 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2008 applied a patch sanctioned by the Sieve maintainer.
2010 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2011 the uid or gid is negative. A case of a negative gid caused this to be
2012 noticed. The fix allows for either to be negative.
2014 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2015 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2018 PH/36 The change PH/12 above was broken. Fixed it.
2020 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2021 the grounds that routing the same address twice would always produce the
2022 same answer. This might have been true once, but it is certainly no
2023 longer true now. Routing a child address may depend on the previous
2024 routing that produced that child. Some complicated redirection strategies
2025 went wrong when messages had multiple recipients, and made Exim's
2026 behaviour dependent on the order in which the addresses were given.
2028 I have moved the duplicate checking until after the routing is complete.
2029 Exim scans the addresses that are assigned to local and remote
2030 transports, and removes any duplicates. This means that more work will be
2031 done, as duplicates will always all be routed, but duplicates are
2032 presumably rare, so I don't expect this is of any significance.
2034 For deliveries to pipes, files, and autoreplies, the duplicate checking
2035 still happens during the routing process, since they are not going to be
2038 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2039 It corrects a timeout issue with spamd. This is Ian's comment: "The
2040 background is that sometimes spamd either never reads data from a
2041 connection it has accepted, or it never writes response data. The exiscan
2042 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2043 blindly assumes that writes won't block so it may never time out."
2045 PH/39 Allow G after quota size as well as K and M.
2047 PH/40 The value set for $authenticated_id in an authenticator may not contain
2048 binary zeroes or newlines because the value is written to log lines and
2049 to spool files. There was no check on this. Now the value is run through
2050 the string_printing() function so that such characters are converted to
2051 printable escape sequences.
2053 PH/41 $message_linecount is a new variable that contains the total number of
2054 lines in the message. Compare $body_linecount, which is the count for the
2057 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2058 bounce and delay warning messages, except in certain special cases, which
2061 (a) An SMTP error message from a remote host;
2062 (b) A message specified in a :fail: redirection;
2063 (c) A message specified in a "fail" command in a system filter;
2064 (d) A message specified in a FAIL return from the queryprogram router;
2065 (e) A message specified by the cannot_route_message router option.
2067 In these cases only, Exim does include the error details in bounce and
2068 warning messages. There are also a few cases where bland messages such
2069 as "unrouteable address" or "local delivery error" are given.
2071 PH/43 $value is now also set for the "else" part of a ${run expansion.
2073 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2074 being worked on, but at least Exim now implements the latest version to
2077 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2078 process to complete was treated as a delivery failure, a timeout while
2079 writing the message to the pipe was logged, but erroneously treated as a
2080 successful delivery. Such timeouts include transport filter timeouts. For
2081 consistency with the overall process timeout, these timeouts are now
2082 treated as errors, giving rise to delivery failures by default. However,
2083 there is now a new Boolean option for the pipe transport called
2084 timeout_defer, which, if set TRUE, converts the failures into defers for
2085 both kinds of timeout. A transport filter timeout is now identified in
2088 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2089 systems where "make" and "gmake" are different, calling "gmake" at top
2090 level broke things. I've arranged for the value of $(MAKE) to be passed
2091 from the Makefile to this script so that it can call the same version of
2095 A note about Exim versions 4.44 and 4.50
2096 ----------------------------------------
2098 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2099 changes of various kinds. As a consequence, a big documentation update was
2100 needed. This delayed the release for rather longer than seemed good, especially
2101 in the light of a couple of (minor) security issues. Therefore, the changes
2102 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2103 release. So 4.44 and 4.50 are in effect two different branches that both start
2106 I have left the 4.50 change log unchanged; it contains all the changes since
2107 4.43. The change log for 4.44 is below; many of its items are identical to
2108 those for 4.50. This seems to be the most sensible way to preserve the
2109 historical information.
2115 1. Minor wording change to the doc/README.SIEVE file.
2117 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2118 computation of the current number of files was incorrect.
2120 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2121 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2124 4. Give more explanation in the error message when the command for a transport
2125 filter fails to execute.
2127 5. There are several places where Exim runs a non-Exim command in a
2128 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2129 was being done only for the command run by the queryprogram router. It is
2130 now done for all such subprocesses. The other cases are: ${run, transport
2131 filters, and the commands run by the lmtp and pipe transports.
2133 6. Added CONFIGURE_GROUP build-time option.
2135 7. Some older OS have a limit of 256 on the maximum number of file
2136 descriptors. Exim was using setrlimit() to set 1000 as a large value
2137 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2138 systems. I've change it so that if it can't get 1000, it tries for 256.
2140 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2141 was an oversight, and furthermore, ever since the addition of extra
2142 controls (e.g. 4.43/32), the checks on when to allow different forms of
2143 "control" were broken. There should now be diagnostics for all cases when a
2144 control that does not make sense is encountered.
2146 9. Added the /retain_sender option to "control=submission".
2148 10. $recipients is now available in the predata ACL (oversight).
2150 11. Tidy the search cache before the fork to do a delivery from a message
2151 received from the command line. Otherwise the child will trigger a lookup
2152 failure and thereby defer the delivery if it tries to use (for example) a
2153 cached ldap connection that the parent has called unbind on.
2155 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2156 of $address_data from the recipient verification was clobbered by the
2157 sender verification.
2159 13. The value of address_data from a sender verification is now available in
2160 $sender_address_data in subsequent conditions in the ACL statement.
2162 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2164 15. Added a new option "connect=<time>" to callout options, to set a different
2167 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2168 was its contents. (It was OK if the option was not defined at all.)
2170 17. A "Completed" log line is now written for messages that are removed from
2171 the spool by the -Mrm option.
2173 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2174 information about exactly what failed.
2176 19. Added -dd to debug only the daemon process.
2178 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2179 handles timeouts, both on the server side and network timeouts. Renamed the
2180 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2182 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2183 It is now set to "smtps".
2185 22. $host_address is now set to the target address during the checking of
2186 ignore_target_hosts.
2188 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2189 being passed; this would have caused $sender_host_name to have been used if
2190 matching the list had actually called for a host name (not very likely,
2191 since this list is usually IP addresses). A host name is now passed as
2194 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2195 code that specifies a non-restarting handler (typically sigaction() in
2196 modern systems) in an attempt to fix a rare and obscure crash bug.
2198 25. Narrowed the window for a race in the daemon that could cause it to ignore
2199 SIGCHLD signals. This is not a major problem, because they are used only to
2200 wake it up if nothing else does.
2202 26. A malformed maildirsize file could cause Exim to calculate negative values
2203 for the mailbox size or file count. Odd effects could occur as a result.
2204 The maildirsize information is now recalculated if the size or filecount
2207 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2208 support for a long time. Removed HAVE_SYS_VFS_H.
2210 28. Installed the latest version of exipick from John Jetmore.
2212 29. In an address list, if the pattern was not a regular expression, an empty
2213 subject address (from a bounce message) matched only if the pattern was an
2214 empty string. Non-empty patterns were not even tested. This was the wrong
2215 because it is perfectly reasonable to use an empty address as part of a
2216 database query. An empty address is now tested by patterns that are
2217 lookups. However, all the other forms of pattern expect the subject to
2218 contain a local part and a domain, and therefore, for them, an empty
2219 address still always fails if the pattern is not itself empty.
2221 30. Exim went into a mad DNS loop when attempting to do a callout where the
2222 host was specified on an smtp transport, and looking it up yielded more
2223 than one IP address.
2225 31. Re-factored the code for checking spool and log partition space into a
2226 function that finds that data and another that does the check. The former
2227 is then used to implement four new variables: $spool_space, $log_space,
2228 $spool_inodes, and $log_inodes.
2230 32. The RFC2047 encoding function was originally intended for short strings
2231 such as real names; it was not keeping to the 75-character limit for
2232 encoded words that the RFC imposes. It now respects the limit, and
2233 generates multiple encoded words if necessary. To be on the safe side, I
2234 have increased the buffer size for the ${rfc2047: expansion operator from
2237 33. It is now permitted to omit both strings after an "if" condition; if the
2238 condition is true, the result is "true". As before, when the second string
2239 is omitted, a false condition yields an empty string. This makes it less
2240 cumbersome to write custom ACL and router conditions.
2242 34. Failure to deliver a bounce message always caused it to be frozen, even if
2243 there was an errors_to setting on the router. The errors_to setting is now
2246 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2247 canonical form (fully expanded) before being placed in
2248 $sender_host_address.
2250 36. The table in the code that translates DNS record types into text (T_A to
2251 "A" for instance) was missing entries for NS and CNAME. It is just possible
2252 that this could have caused confusion if both these types were looked up
2253 for the same domain, because the text type is used as part of Exim's
2254 per-process caching. But the chance of anyone hitting this buglet seems
2257 37. The dnsdb lookup has been extended in a number of ways.
2259 (1) There is a new type, "zns", which walks up the domain tree until it
2260 finds some nameserver records. It should be used with care.
2262 (2) There is a new type, "mxh", which is like "mx" except that it returns
2263 just the host names, not the priorities.
2265 (3) It is now possible to give a list of domains (or IP addresses) to be
2266 looked up. The behaviour when one of the lookups defers can be
2267 controlled by a keyword.
2269 (4) It is now possible to specify the separator character for use when
2270 multiple records are returned.
2272 38. The dnslists ACL condition has been extended: it is now possible to supply
2273 a list of IP addresses and/or domains to be looked up in a particular DNS
2276 39. Added log_selector=+queue_time_overall.
2278 40. When running the queue in the test harness, wait just a tad after forking a
2279 delivery process, to get repeatability of debugging output.
2281 41. Include certificate and key file names in error message when GnuTLS fails
2282 to set them up, because the GnuTLS error message doesn't include the name
2283 of the failing file when there is a problem reading it.
2285 42. Allow both -bf and -bF in the same test run.
2287 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2289 44. The "Exiscan patch" is now merged into the mainline Exim source.
2291 45. Sometimes the final signoff response after QUIT could fail to get
2292 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2293 < 0 before doing a fflush(). This bug looks as though it goes back to the
2294 introduction of TLS in release 3.20, but "sometimes" must have been rare
2295 because the tests only now provoked it.
2297 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2298 (this can affect the format of dates).
2300 47. exim_tidydb, when checking for the continued existence of a message for
2301 which it has found a message-specific retry record, was not finding
2302 messages that were in split spool directories. Consequently, it was
2303 deleting retry records that should have stayed in existence.
2305 48. Steve fixed some bugs in eximstats.
2307 49. The SPA authentication driver was not abandoning authentication and moving
2308 on to the next authenticator when an expansion was forced to fail,
2309 contradicting the general specification for all authenticators. Instead it
2310 was generating a temporary error. It now behaves as specified.
2312 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2313 (the order specifies the preference for clients). The order is now AES256,
2314 AES128, 3DES, ARCFOUR128.
2316 51. Small patch to Sieve code - explicitly set From: when generating an
2319 52. Exim crashed if a remote delivery caused a very long error message to be
2320 recorded - for instance if somebody sent an entire SpamAssassin report back
2321 as a large number of 550 error lines. This bug was coincidentally fixed by
2322 increasing the size of one of Exim's internal buffers (big_buffer) that
2323 happened as part of the Exiscan merge. However, to be on the safe side, I
2324 have made the code more robust (and fixed the comments that describe what
2327 53. Now that there can be additional text after "Completed" in log lines (if
2328 the queue_time_overall log selector is set), a one-byte patch to exigrep
2329 was needed to allow it to recognize "Completed" as not the last thing in
2332 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2333 patch that reportedly fixes this has been added. I am not expert enough to
2334 create a test for it. This is what the patch creator wrote:
2336 "I found a little strange behaviour of ldap code when working with
2337 Windows 2003 AD Domain, where users was placed in more than one
2338 Organization Units. When I tried to give exim partial DN, the exit code
2339 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2340 But simultaneously result of request was absolutely normal ldap result,
2341 so I produce this patch..."
2343 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2344 REFERENCE, so I have modified the code to exclude the patch when that macro
2347 55. Some experimental protocols are using DNS PTR records for new purposes. The
2348 keys for these records are domain names, not reversed IP addresses. The
2349 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2350 leaves it alone. Component reversal etc. now happens only for IP addresses.
2353 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2355 57. Double the size of the debug message buffer (to 2048) so that more of very
2356 long debug lines gets shown.
2358 58. The exicyclog utility now does better if the number of log files to keep
2359 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2361 59. Two changes related to the smtp_active_hostname option:
2363 (1) $smtp_active_hostname is now available as a variable.
2364 (2) The default for smtp_banner uses $smtp_active_hostname instead
2365 of $primary_hostname.
2367 60. The host_aton() function is supposed to be passed a string that is known
2368 to be a valid IP address. However, in the case of IPv6 addresses, it was
2369 not checking this. This is a hostage to fortune. Exim now panics and dies
2370 if the condition is not met. A case was found where this could be provoked
2371 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2372 components; fortuitously, this particular loophole had already been fixed
2373 by change 4.50/55 above.
2375 If there are any other similar loopholes, the new check in host_aton()
2376 itself should stop them being exploited. The report I received stated that
2377 data on the command line could provoke the exploit when Exim was running as
2378 exim, but did not say which command line option was involved. All I could
2379 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2380 running as the user.
2383 61. There was a buffer overflow vulnerability in the SPA authentication code
2384 (which came originally from the Samba project). I have added a test to the
2385 spa_base64_to_bits() function which I hope fixes it.
2388 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2389 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2391 63. The daemon start-up calls getloadavg() while still root for those OS that
2392 need the first call to be done as root, but it missed one case: when
2393 deliver_queue_load_max is set with deliver_drop_privilege. This is
2394 necessary for the benefit of the queue runner, because there is no re-exec
2395 when deliver_drop_privilege is set.
2397 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2398 This has been fixed.
2400 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2401 was in use, was not putting the data itself into the right store pool;
2402 consequently, it could be overwritten for a subsequent message in the same
2403 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2406 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2408 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2409 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2410 the function were treating the return as a boolean value, which happened to
2411 work because 0=false and not-0=true, but is not correct code.
2413 68. The host_aton() function was not handling scoped IPv6 addresses (those
2414 with, for example, "%eth0" on the end) correctly.
2416 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2417 negated items (that is, ~something) in unsigned ints. Some compilers
2418 apparently mutter when there is no cast.
2420 70. If an address verification called from an ACL failed, and did not produce a
2421 user-specific message (i.e. there was only a "system" message), nothing was
2422 put in $acl_verify_message. In this situation, it now puts the system
2425 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2426 session; change 4.31/43 added the unwanted input to the log line - except
2427 that it did not do this in the start of session case. It now does.
2429 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2430 This is wrong and can cause the other end to generate a synchronization
2431 error if it is another Exim or anything else that does the synchronization
2432 check. A QUIT command is no longer sent after a timeout.
2434 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2435 during host lookups.
2437 74. The defer_ok option of callout verification was not working if it was used
2438 when verifying addresses in header lines, that is, for this case:
2440 verify = header_sender/callout=defer_ok
2442 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2443 those file descriptors could be used for SMTP connections. If anything
2444 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2445 could be sent to the SMTP client, causing chaos. The daemon now opens
2446 stdin, stdout, and stderr to /dev/null when it puts itself into the
2449 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2450 log by default. The user can override this with suitable Perl magic.
2452 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2453 the log message when discard triggers, was not working for the DATA ACL or
2454 for the non-SMTP ACL.
2456 78. Error message wording change in sieve.c.
2458 79. If smtp_accept_max_per_host was set, the number of connections could be
2459 restricted to fewer than expected, because the daemon was trying to set up
2460 a new connection before checking whether the processes handling previous
2461 connections had finished. The check for completed processes is now done
2462 earlier. On busy systems, this bug wouldn't be noticed because something
2463 else would have woken the daemon, and it would have reaped the completed
2466 80. If a message was submitted locally by a user whose login name contained one
2467 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2468 It caused a spool format error. I have fixed the spool reading code. A
2469 related problem was that the "from" clause in the Received: line became
2470 illegal because of the space(s). It is now covered by ${quote_local_part.
2472 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2475 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2477 83. Patch from Sieve maintainer to fix unterminated string problem in
2478 "vacation" handling.
2480 84. Some minor changes to the Linux configuration files to help with other
2481 OS variants using glibc.
2483 85. One more patch for Sieve to update vacation handling to latest spec.
2486 ----------------------------------------------------
2487 See the note above about the 4.44 and 4.50 releases.
2488 ----------------------------------------------------
2494 1. Change 4.43/35 introduced a bug that caused file counts to be
2495 incorrectly computed when quota_filecount was set in an appendfile
2498 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2499 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2502 3. Give more explanation in the error message when the command for a transport
2503 filter fails to execute.
2505 4. There are several places where Exim runs a non-Exim command in a
2506 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2507 was being done only for the command run by the queryprogram router. It is
2508 now done for all such subprocesses. The other cases are: ${run, transport
2509 filters, and the commands run by the lmtp and pipe transports.
2511 5. Some older OS have a limit of 256 on the maximum number of file
2512 descriptors. Exim was using setrlimit() to set 1000 as a large value
2513 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2514 systems. I've change it so that if it can't get 1000, it tries for 256.
2516 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2517 was an oversight, and furthermore, ever since the addition of extra
2518 controls (e.g. 4.43/32), the checks on when to allow different forms of
2519 "control" were broken. There should now be diagnostics for all cases when a
2520 control that does not make sense is encountered.
2522 7. $recipients is now available in the predata ACL (oversight).
2524 8. Tidy the search cache before the fork to do a delivery from a message
2525 received from the command line. Otherwise the child will trigger a lookup
2526 failure and thereby defer the delivery if it tries to use (for example) a
2527 cached ldap connection that the parent has called unbind on.
2529 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2530 of $address_data from the recipient verification was clobbered by the
2531 sender verification.
2533 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2534 was its contents. (It was OK if the option was not defined at all.)
2536 11. A "Completed" log line is now written for messages that are removed from
2537 the spool by the -Mrm option.
2539 12. $host_address is now set to the target address during the checking of
2540 ignore_target_hosts.
2542 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2543 being passed; this would have caused $sender_host_name to have been used if
2544 matching the list had actually called for a host name (not very likely,
2545 since this list is usually IP addresses). A host name is now passed as
2548 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2549 code that specifies a non-restarting handler (typically sigaction() in
2550 modern systems) in an attempt to fix a rare and obscure crash bug.
2552 15. Narrowed the window for a race in the daemon that could cause it to ignore
2553 SIGCHLD signals. This is not a major problem, because they are used only to
2554 wake it up if nothing else does.
2556 16. A malformed maildirsize file could cause Exim to calculate negative values
2557 for the mailbox size or file count. Odd effects could occur as a result.
2558 The maildirsize information is now recalculated if the size or filecount
2561 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2562 support for a long time. Removed HAVE_SYS_VFS_H.
2564 18. Updated exipick to current release from John Jetmore.
2566 19. Allow an empty sender to be matched against a lookup in an address list.
2567 Previously the only cases considered were a regular expression, or an
2570 20. Exim went into a mad DNS lookup loop when doing a callout where the
2571 host was specified on the transport, if the DNS lookup yielded more than
2574 21. The RFC2047 encoding function was originally intended for short strings
2575 such as real names; it was not keeping to the 75-character limit for
2576 encoded words that the RFC imposes. It now respects the limit, and
2577 generates multiple encoded words if necessary. To be on the safe side, I
2578 have increased the buffer size for the ${rfc2047: expansion operator from
2581 22. Failure to deliver a bounce message always caused it to be frozen, even if
2582 there was an errors_to setting on the router. The errors_to setting is now
2585 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2586 canonical form (fully expanded) before being placed in
2587 $sender_host_address.
2589 24. Updated eximstats to version 1.33
2591 25. Include certificate and key file names in error message when GnuTLS fails
2592 to set them up, because the GnuTLS error message doesn't include the name
2593 of the failing file when there is a problem reading it.
2595 26. Expand error message when OpenSSL has problems setting up cert/key files.
2598 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2599 (this can affect the format of dates).
2601 28. exim_tidydb, when checking for the continued existence of a message for
2602 which it has found a message-specific retry record, was not finding
2603 messages that were in split spool directories. Consequently, it was
2604 deleting retry records that should have stayed in existence.
2606 29. eximstats updated to version 1.35
2607 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2608 1.35 - bugfix such that pie charts by volume are generated correctly
2610 30. The SPA authentication driver was not abandoning authentication and moving
2611 on to the next authenticator when an expansion was forced to fail,
2612 contradicting the general specification for all authenticators. Instead it
2613 was generating a temporary error. It now behaves as specified.
2615 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2616 (the order specifies the preference for clients). The order is now AES256,
2617 AES128, 3DES, ARCFOUR128.
2619 31. Small patch to Sieve code - explicitly set From: when generating an
2622 32. Exim crashed if a remote delivery caused a very long error message to be
2623 recorded - for instance if somebody sent an entire SpamAssassin report back
2624 as a large number of 550 error lines. This bug was coincidentally fixed by
2625 increasing the size of one of Exim's internal buffers (big_buffer) that
2626 happened as part of the Exiscan merge. However, to be on the safe side, I
2627 have made the code more robust (and fixed the comments that describe what
2630 33. Some experimental protocols are using DNS PTR records for new purposes. The
2631 keys for these records are domain names, not reversed IP addresses. The
2632 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2633 leaves it alone. Component reversal etc. now happens only for IP addresses.
2636 34. The host_aton() function is supposed to be passed a string that is known
2637 to be a valid IP address. However, in the case of IPv6 addresses, it was
2638 not checking this. This is a hostage to fortune. Exim now panics and dies
2639 if the condition is not met. A case was found where this could be provoked
2640 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2641 components; fortuitously, this particular loophole had already been fixed
2642 by change 4.50/55 or 4.44/33 above.
2644 If there are any other similar loopholes, the new check in host_aton()
2645 itself should stop them being exploited. The report I received stated that
2646 data on the command line could provoke the exploit when Exim was running as
2647 exim, but did not say which command line option was involved. All I could
2648 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2649 running as the user.
2652 35. There was a buffer overflow vulnerability in the SPA authentication code
2653 (which came originally from the Samba project). I have added a test to the
2654 spa_base64_to_bits() function which I hope fixes it.
2657 36. The daemon start-up calls getloadavg() while still root for those OS that
2658 need the first call to be done as root, but it missed one case: when
2659 deliver_queue_load_max is set with deliver_drop_privilege. This is
2660 necessary for the benefit of the queue runner, because there is no re-exec
2661 when deliver_drop_privilege is set.
2663 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2664 was in use, was not putting the data itself into the right store pool;
2665 consequently, it could be overwritten for a subsequent message in the same
2666 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2669 38. Sometimes the final signoff response after QUIT could fail to get
2670 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2671 < 0 before doing a fflush(). This bug looks as though it goes back to the
2672 introduction of TLS in release 3.20, but "sometimes" must have been rare
2673 because the tests only now provoked it.
2679 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2680 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2681 Now it returns an integer. A number of calls were still expecting a T/F
2682 return. Fortuitously, in all cases, the tests worked in OK situations,
2683 which is the norm. However, things would have gone wrong on any write
2684 failures on the smtp file descriptor. This function is used when sending
2685 messages over SMTP and also when doing verify callouts.
2687 2. When Exim is called to do synchronous delivery of a locally submitted
2688 message (the -odf or -odi options), it no longer closes stderr before doing
2691 3. Implemented the mua_wrapper option.
2693 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2695 5. Implemented the functions header_remove(), header_testname(),
2696 header_add_at_position(), and receive_remove_recipient(), and exported them
2699 6. If an ACL "warn" statement specified the addition of headers, Exim already
2700 inserted X-ACL-Warn: at the start if there was no header name. However, it
2701 was not making this test for the second and subsequent header lines if
2702 there were newlines in the string. This meant that an invalid header could
2703 be inserted if Exim was badly configured.
2705 7. Allow an ACL "warn" statement to add header lines at the start or after all
2706 the Received: headers, as well as at the end.
2708 8. Added the rcpt_4xx retry error code.
2710 9. Added postmaster_mailfrom=xxx to callout verification option.
2712 10. Added mailfrom=xxxx to the callout verification option, for verify=
2715 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2716 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2718 12. Inserted some casts to stop certain compilers warning when using pointer
2719 differences as field lengths or precisions in printf-type calls (mostly
2720 affecting debugging statements).
2722 13. Added optional readline() support for -be (dynamically loaded).
2724 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2725 same clock tick as a message's arrival, so that its received time was the
2726 same as the "first fail" time on the retry record, and that message
2727 remained on the queue past the ultimate address timeout, every queue runner
2728 would try a delivery (because it was past the ultimate address timeout) but
2729 after another failure, the ultimate address timeout, which should have then
2730 bounced the address, did not kick in. This was a "< instead of <=" error;
2731 in most cases the first failure would have been in the next clock tick
2732 after the received time, and all would be well.
2734 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2735 being recognized when the domain list was tested by the match_domain
2736 condition in an expansion string.
2738 16. Added the ${str2b64: operator.
2740 17. Exim was always calling setrlimit() to set a large limit for the number of
2741 processes, without checking whether the existing limit was already
2742 adequate. (It did check for the limit on file descriptors.) Furthermore,
2743 errors from getrlimit() and setrlimit() were being ignored. Now they are
2744 logged to the main and panic logs, but Exim does carry on, to try to do its
2745 job under whatever limits there are.
2747 18. Imported PCRE 5.0.
2749 19. Trivial typo in log message " temporarily refused connection" (the leading
2752 20. If the log selector return_path_on_delivery was set and an address was
2753 redirected to /dev/null, the delivery process crashed because it assumed
2754 that a return path would always be set for a "successful" delivery. In this
2755 case, the whole delivery is bypassed as an optimization, and therefore no
2758 21. Internal re-arrangement: the function for sending a challenge and reading
2759 a response while authentication was assuming a zero-terminated challenge
2760 string. It's now changed to take a pointer and a length, to allow for
2761 binary data in such strings.
2763 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2765 23. Exim was not respecting finduser_retries when seeking the login of the
2766 uid under which it was called; it was always trying 10 times. (The default
2767 setting of finduser_retries is zero.) Also, it was sleeping after the final
2768 failure, which is pointless.
2770 24. Implemented tls_on_connect_ports.
2772 25. Implemented acl_smtp_predata.
2774 26. If the domain in control=submission is set empty, Exim assumes that the
2775 authenticated id is a complete email address when it generates From: or
2776 Sender: header lines.
2778 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2779 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2780 chown and chgrp in /bin and hostname in /usr/bin.
2782 28. Exim was keeping the "process log" file open after each use, just as it
2783 does for the main log. This opens the possibility of it remaining open for
2784 long periods when the USR1 signal hits a daemon. Occasional processlog
2785 errors were reported, that could have been caused by this. Anyway, it seems
2786 much more sensible not to leave this file open at all, so that is what now
2789 29. The long-running daemon process does not normally write to the log once it
2790 has entered its main loop, and it closes the log before doing so. This is
2791 so that log files can straightforwardly be renamed and moved. However,
2792 there are a couple of unusual error situations where the daemon does write
2793 log entries, and I had neglected to close the log afterwards.
2795 30. The text of an SMTP error response that was received during a remote
2796 delivery was being truncated at 512 bytes. This is too short for some of
2797 the long messages that one sometimes sees. I've increased the limit to
2800 31. It is now possible to make retry rules that apply only when a message has a
2801 specific sender, in particular, an empty sender.
2803 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2804 it possible to be selective about when SMTP synchronization is enforced.
2806 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2808 32. Implemented hosts_connection_nolog.
2810 33. Added an ACL for QUIT.
2812 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2815 35. Added mailbox_size and mailbox_filecount to appendfile.
2817 36. Added control = no_multiline_responses to ACLs.
2819 37. There was a bug in the logic of the code that waits for the clock to tick
2820 in the case where the clock went backwards by a substantial amount such
2821 that the microsecond fraction of "now" was more than the microsecond
2822 fraction of "then" (but the whole seconds number was less).
2824 38. Added support for the libradius Radius client library this is found on
2825 FreeBSD (previously only the radiusclient library was supported).
2831 1. When certain lookups returned multiple values in the form name=value, the
2832 quoting of the values was not always being done properly. Specifically:
2833 (a) If the value started with a double quote, but contained no whitespace,
2835 (b) If the value contained whitespace other than a space character (i.e.
2836 tabs or newlines or carriage returns) it was not quoted.
2837 This fix has been applied to the mysql and pgsql lookups by writing a
2838 separate quoting function and calling it from the lookup code. The fix
2839 should probably also be applied to nisplus, ibase and oracle lookups, but
2840 since I cannot test any of those, I have not disturbed their existing code.
2842 2. A hit in the callout cache for a specific address caused a log line with no
2843 reason for rejecting RCPT. Now it says "Previous (cached) callout
2844 verification failure".
2846 3. There was an off-by-one bug in the queryprogram router. An over-long
2847 return line was truncated at 256 instead of 255 characters, thereby
2848 overflowing its buffer with the terminating zero. As well as fixing this, I
2849 have increased the buffer size to 1024 (and made a note to document this).
2851 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2852 when Exim is waiting for an SMTP response from a remote server, Exim
2853 restarts its select() call on the socket, thereby resetting its timeout.
2854 This is not a problem when such interrupts are rare. Somebody set up a cron
2855 job to run exiwhat every 2 minutes, which is less than the normal select()
2856 timeout (5 or 10 minutes). This meant that the select() timeout never
2857 kicked in because it was always reset. I have fixed this by comparing the
2858 time when an interrupt arrives with the time at the start of the first call
2859 to select(). If more time than the timeout has elapsed, the interrupt is
2860 treated as a timeout.
2862 5. Some internal re-factoring in preparation for the addition of Sieve
2863 extensions (by MH). In particular, the "personal" test is moved to a
2864 separate function, and given an option for scanning Cc: and Bcc: (which is
2865 not set for Exim filters).
2867 6. When Exim created an email address using the login of the caller as the
2868 local part (e.g. when creating a From: or Sender: header line), it was not
2869 quoting the local part when it contained special characters such as @.
2871 7. Installed new OpenBSD configuration files.
2873 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2874 try to make them clearer.
2876 9. Callout options, other than the timeout value, were being ignored when
2877 verifying sender addresses in header lines. For example, when using
2879 verify = header_sender/callout=no_cache
2881 the cache was (incorrectly) being used.
2883 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2884 only the Cygwin environment.
2886 11. When return_path_on_delivery was set as a log selector, if different remote
2887 addresses in the same message used different return paths and parallel
2888 remote delivery occurred, the wrong values would sometimes be logged.
2889 (Whenever a remote delivery process finished, the return path value from
2890 the most recently started remote delivery process was logged.)
2892 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2893 lines when AUTH and/or TLS are in use. This is the "received protocol"
2894 field. Exim used to use "asmtp" for authenticated SMTP, without any
2895 indication (in the protocol name) for TLS use. Now it follows the RFC and
2896 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2897 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2898 names appear in log lines as well as in Received: header lines.
2900 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2901 extensions, and comparison tests, and to fix some bugs.
2903 14. Changes to the "personal" filter test:
2905 (1) The test was buggy in that it was just doing the equivalent of
2906 "contains" tests on header lines. For example, if a user's address was
2907 anne@some.where, the "personal" test would incorrectly be true for
2909 To: susanne@some.where
2911 This test is now done by extracting each address from the header in turn,
2912 and checking the entire address. Other tests that are part of "personal"
2913 are now done using regular expressions (for example, to check local parts
2914 of addresses in From: header lines).
2916 (2) The list of non-personal local parts in From: addresses has been
2917 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2918 taken from the Sieve specification recommendations.
2920 (3) If the message contains any header line starting with "List-" it is
2921 treated as non-personal.
2923 (4) The test for "circular" in the Subject: header line has been removed
2924 because it now seems ill-conceived.
2926 15. Minor typos in src/EDITME comments corrected.
2928 16. Installed latest exipick from John Jetmore.
2930 17. If headers_add on a router specified a text string that was too long for
2931 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2932 of string_sprintf() is now avoided.
2934 18. $message_body_size was not set (it was always zero) when running the DATA
2935 ACL and the local_scan() function.
2937 19. For the "mail" command in an Exim filter, no default was being set for
2938 the once_repeat time, causing a random time value to be used if "once" was
2939 specified. (If the value happened to be <= 0, no repeat happened.) The
2940 default is now 0s, meaning "never repeat". The "vacation" command was OK
2941 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2942 (I found it when inspecting the code).
2944 20. There is now an overall timeout for performing a callout verification. It
2945 defaults to 4 times the callout timeout, which applies to individual SMTP
2946 commands during the callout. The overall timeout applies when there is more
2947 than one host that can be tried. The timeout is checked before trying the
2948 next host. This prevents very long delays if there are a large number of
2949 hosts and all are timing out (e.g. when the network connections are timing
2950 out). The value of the overall timeout can be changed by specifying an
2951 additional sub-option for "callout", called "maxwait". For example:
2953 verify = sender/callout=5s,maxwait=20s
2955 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2956 to the end before writing, but this should make it even safer).
2958 22. Exim was forgetting that it had advertised PIPELINING for the second and
2959 subsequent messages on an SMTP connection. It was also not resetting its
2960 memory on STARTTLS and an internal HELO.
2962 23. When Exim logs an SMTP synchronization error within a session, it now
2963 records whether PIPELINING has been advertised or not.
2965 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2966 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2967 rather than long int.
2969 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2971 26. Added the never_mail option to autoreply.
2977 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2978 crash if the getsockname() call failed; this can happen if a connection is
2979 closed very soon after it is established. The problem was simply in the
2980 order in which certain operations were done, causing Exim to try to write
2981 to the SMTP stream before it had set up the file descriptor. The bug has
2982 been fixed by making things happen in the correct order.
2988 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2989 before the connection was closed, thus losing the rejection response.
2991 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2992 some early Solaris releases, but causes trouble in current releases where
2993 socklen_t is defined.
2995 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2998 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3001 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3002 that releasing the top of it at the end releases what was used for sub-
3003 expansions (unless the block got too big). However, discard this block if
3004 the first thing is a variable or header, so that we can use its block when
3005 it is dynamic (useful for very large $message_headers, for example).
3007 6. Lookups now cache *every* query, not just the most recent. A new, separate
3008 store pool is used for this. It can be recovered when all lookup caches are
3009 flushed. Lookups now release memory at the end of their result strings.
3010 This has involved some general refactoring of the lookup sources.
3012 7. Some code has been added to the store_xxx() functions to reduce the amount
3013 of flapping under certain conditions.
3015 8. log_incoming_interface used to affect only the <= reception log lines. Now
3016 it causes the local interface and port to be added to several more SMTP log
3017 lines, for example "SMTP connection from", and rejection lines.
3019 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3021 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3023 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3024 could be overwritten at the end of the current message (or the start of a
3025 new message if it was set in a HELO ACL). The value is now preserved for
3026 the duration of the SMTP connection.
3028 12. If a transport had a headers_rewrite setting, and a matching header line
3029 contained an unqualified address, that address was qualified, even if it
3030 did not match any rewriting rules. The underlying bug was that the values
3031 of the flags that permit the existence of unqualified sender and recipient
3032 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3033 non-local messages, and by -bnq for local messages) were not being
3034 preserved with the message after it was received.
3036 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3037 "next input=" as part of the text comprising the host identity instead of
3038 the correct text. The code was using the same buffer for two different
3039 strings. However, depending on which order the printing function evaluated
3040 its arguments, the bug did not always show up. Under Linux, for example, my
3041 test suite worked just fine.
3043 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3044 doesn't work with some older versions of Perl. It has been changed to "my",
3045 which in any case is probably the better facility to use.
3047 15. A really picky compiler found some instances of statements for creating
3048 error messages that either had too many or two few arguments for the format
3051 16. The size of the buffer for calls to the DNS resolver has been increased
3052 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3053 for addresses that have a lot of PTR records. This alleviates a problem; it
3054 does not fully solve it.
3056 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3057 buffer now truncates the list and logs the incident, which is the same
3058 action as happens when Exim is looking up a host name and its aliases.
3059 Previously in this situation something unpredictable would happen;
3060 sometimes it was "internal error: store_reset failed".
3062 18. If a server dropped the connection unexpectedly when an Exim client was
3063 using GnuTLS and trying to read a response, the client delivery process
3064 crashed while trying to generate an error log message.
3066 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3067 string, for example:
3069 warn message = H1: something\nH2: something
3071 the text was added as a single header line from Exim's point of view
3072 though it ended up OK in the delivered message. However, searching for the
3073 second and subsequent header lines using $h_h2: did not work. This has been
3074 fixed. Similarly, if a system filter added multiple headers in this way,
3075 the routers could not see them.
3077 20. Expanded the error message when iplsearch is called with an invalid key to
3078 suggest using net-iplsearch in a host list.
3080 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3081 ACLs are no longer actually imposed (and a message to that effect is
3084 22. If a "gecos" field in a passwd entry contained escaped characters, in
3085 particular, if it contained a \" sequence, Exim got it wrong when building
3086 a From: or a Sender: header from that name. A second bug also caused
3087 incorrect handling when an unquoted " was present following a character
3088 that needed quoting.
3090 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3091 was not being matched caselessly.
3093 24. Arranged for all hyphens in the exim.8 source to be escaped with
3096 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3097 properly. Recipient callout cache records were still being keyed to include
3098 the sender, even when use_sender was set false. This led to far more
3099 callouts that were necessary. The sender is no longer included in the key
3100 when use_sender is false.
3102 26. Added "control = submission" modifier to ACLs.
3104 27. Added the ${base62d: operator to decode base 62 numbers.
3106 28. dnsdb lookups can now access SRV records.
3108 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3109 the configuration file.
3111 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3112 (-v) mode. This makes the output for a verbose queue run more intelligible.
3114 31. Added a use_postmaster feature to recipient callouts.
3116 32. Added the $body_zerocount variable, containing the number of binary zero
3117 bytes in the message body.
3119 33. The time of last modification of the "new" subdirectory is now used as the
3120 "mailbox time last read" when there is a quota error for a maildir
3123 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3125 35. Added +ignore_unknown as a special item in host lists.
3127 36. Code for decoding IPv6 addresses in host lists is now included, even if
3128 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3129 address was recognized as an IP address, but was then not correctly decoded
3130 into binary, causing unexpected and incorrect effects when compared with
3137 1. Very minor rewording of debugging text in manualroute to say "list of
3138 hosts" instead of "hostlist".
3140 2. If verify=header_syntax was set, and a header line with an unqualified
3141 address (no domain) and a large number of spaces between the end of the
3142 name and the colon was received, the reception process suffered a buffer
3143 overflow, and (when I tested it) crashed. This was caused by some obsolete
3144 code that should have been removed. The fix is to remove it!
3146 3. When running in the test harness, delay a bit after writing a bounce
3147 message to get a bit more predictability in the log output.
3149 4. Added a call to search_tidyup() just before forking a reception process. In
3150 theory, someone could use a lookup in the expansion of smtp_accept_max_
3151 per_host which, without the tidyup, could leave open a database connection.
3153 5. Added the variables $recipient_data and $sender_data which get set from a
3154 lookup success in an ACL "recipients" or "senders" condition, or a router
3155 "senders" option, similar to $domain_data and $local_part_data.
3157 6. Moved the writing of debug_print from before to after the "senders" test
3160 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3161 problems for message scanning, either using a data ACL, or using
3162 local_scan() because the Received: header was not generated till after they
3163 were called (in order to set the time as the time of reception completion).
3164 I have revised the way this works. The header is now generated after the
3165 body is received, but before the ACL or local_scan() are called. After they
3166 are run, the timestamp in the header is updated.
3172 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3173 before starting a queue runner without re-exec. This happened only when
3174 deliver_drop_privilege was set or when the Exim user was set to root. The
3175 effect of the bug was that timeouts during subsequent deliveries caused
3176 crashes instead of being properly handled. The handler is now left at its
3177 default (and expected) setting.
3179 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3180 message, again when deliver_drop_privilege is set or Exim is run as root.
3181 The bug described in (1) was not present in this case, but the tidying up
3182 of the other signals was missing. I have made the two cases consistent.
3184 3. The ignore_target_hosts setting on a manualroute router was being ignored
3185 for hosts that were looked up using the /MX notation.
3187 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3190 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3191 operated on the sender address. After changing the $sender_address to <>
3192 for the sender address verify, Exim was re-instated it as the original
3193 (before rewriting) address, but remembering that it had rewritten it, so it
3194 wasn't rewriting it again. This bug also had the effect of breaking the
3195 sender address verification caching when the sender address was rewritten.
3197 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3198 This has been changed so that if the ip literal address matches
3199 ignore_target_hosts, the router declines.
3201 7. Added expansion conditions match_domain, match_address, and match_local_
3202 part (NOT match_host).
3204 8. The placeholder for the Received: header didn't have a length field set.
3206 9. Added code to Exim itself and to exim_lock to test for a specific race
3207 condition that could lead to file corruption when using MBX delivery. The
3208 issue is with the lockfile that is created in /tmp. If this file is removed
3209 after a process has opened it but before that process has acquired a lock,
3210 there is the potential for a second process to recreate the file and also
3211 acquire a lock. This could lead to two Exim processes writing to the file
3212 at the same time. The added code performs the same test as UW imapd; it
3213 checks after acquiring the lock that its file descriptor still refers to
3214 the same named file.
3216 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3217 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3220 11. Added the smtp_active_hostname option. If used, this will typically be made
3221 to depend on the incoming interface address. Because $interface_address is
3222 not set up until the daemon has forked a reception process, error responses
3223 that can happen earlier (such as "too many connections") no longer contain
3226 12. If an expansion in a condition on a "warn" statement fails because a lookup
3227 defers, the "warn" statement is abandoned, and the next ACL statement is
3228 processed. Previously this caused the whole ACL to be aborted.
3230 13. Added the iplsearch lookup type.
3232 14. Added ident_timeout as a log selector.
3234 15. Added tls_certificate_verified as a log selector.
3236 16. Added a global option tls_require_ciphers (compare the smtp transport
3237 option of the same name). This controls incoming TLS connections.
3239 17. I finally figured out how to make tls_require_ciphers do a similar thing
3240 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3241 before starting the TLS session.
3243 18. Tabs are now shown as \t in -bP output.
3245 19. If the log selector return_path_on_delivery was set, Exim crashed when
3246 bouncing a message because it had too many Received: header lines.
3248 20. If two routers both had headers_remove settings, and the first one included
3249 a superfluous trailing colon, the final name in the first list and the
3250 first name in the second list were incorrectly joined into one item (with a
3251 colon in the middle).
3257 1. Added -C and -D options to the exinext utility, mainly to make it easier
3258 to include in the automated testing, but these could be helpful when
3259 multiple configurations are in use.
3261 2. The exinext utility was not formatting the output nicely when there was
3262 an alternate port involved in the retry record key, nor when there was a
3263 message id as well (for retries that were specific to a specific message
3264 and a specific host). It was also confused by IPv6 addresses, because of
3265 the additional colons they contain. I have fixed the IPv4 problem, and
3266 patched it up to do a reasonable job for IPv6.
3268 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3269 delivery, the log line now contains "pipelined" if PIPELINING was used.
3271 4. An SMTP transport process used to panic and die if the bind() call to set
3272 an explicit outgoing interface failed. This has been changed; it is now
3273 treated in the same way as a connect() failure.
3275 5. A reference to $sender_host_name in the part of a conditional expansion
3276 that was being skipped was still causing a DNS lookup. This no longer
3279 6. The def: expansion condition was not recognizing references to header lines
3280 that used bh_ and bheader_.
3282 7. Added the _cache feature to named lists.
3284 8. The code for checking quota_filecount in the appendfile transport was
3285 allowing one more file than it should have been.
3287 9. For compatibility with Sendmail, the command line option
3295 and sets the incoming protocol and host name (for trusted callers). The
3296 host name and its colon can be omitted when only the protocol is to be set.
3297 Note the Exim already has two private options, -pd and -ps, that refer to
3298 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3299 "s", but I don't think that's a major issue.
3301 10. A number of refactoring changes to the code, none of which should affect
3304 (a) The number of logging options was getting close to filling up the
3305 32-bit word that was used as a bit map. I have split them into two classes:
3306 those that are passed in the argument to log_write(), and those that are
3307 only ever tested independently outside of that function. These are now in
3308 separate 32-bit words, so there is plenty of room for expansion again.
3309 There is no change in the user interface or the logging behaviour.
3311 (b) When building, for example, log lines, the code previously used a
3312 macro that called string_cat() twice, in order to add two strings. This is
3313 not really sufficiently general. Furthermore, there was one instance where
3314 it was actually wrong because one of the argument was used twice, and in
3315 one call a function was used. (As it happened, calling the function twice
3316 did not affect the overall behaviour.) The macro has been replaced by a
3317 function that can join an arbitrary number of extra strings onto a growing
3320 (c) The code for expansion conditions now uses a table and a binary chop
3321 instead of a serial search (which was left over from when there were very
3322 few conditions). Also, it now recognizes conditions like "pam" even when
3323 the relevant support is not compiled in: a suitably worded error message is
3324 given if an attempt is made to use such a condition.
3326 11. Added ${time_interval:xxxxx}.
3328 12. A bug was causing one of the ddress fields not to be passed back correctly
3329 from remote delivery subprocesses. The field in question was not being
3330 subsequently used, so this caused to problems in practice.
3332 13. Added new log selectors queue_time and deliver_time.
3334 14. Might have fixed a bug in maildirsizefile handling that threw up
3335 "unexpected character" debug warnings, and recalculated the data
3336 unnecessarily. In any case, I expanded the warning message to give more
3339 15. Added the message "Restricted characters in address" to the statements in
3340 the default ACL that block characters like @ and % in local parts.
3342 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3343 Three changes have been made:
3345 (a) There was a serious bug; a negative response to MAIL caused the whole
3346 recipient domain to be cached as invalid, thereby blocking all messages
3347 to all local parts at the same domain, from all senders. This bug has
3348 been fixed. The domain is no longer cached after a negative response to
3349 MAIL if the sender used is not empty.
3351 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3354 (c) A new callout option, "use_sender" has been added for people who want
3355 the modified behaviour.
3361 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3364 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3365 indeed breaks things for older releases.
3367 3. Added additional logging to the case where there is a problem reading data
3368 from a filter that is running in a subprocess using a pipe, in order to
3369 try to track down a specific problem.
3371 4. Testing facility fudge: when running in the test harness and attempting
3372 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3373 getting "No route to host". Convert this to a timeout.
3375 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3378 6. Some OS don't have socklen_t but use size_t instead. This affects the
3379 fifth argument of getsockopt() amongst other things. This is now
3380 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3381 can be set for individual OS. I have set it for SunOS5, OSF1, and
3382 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3383 some earlier ones do not.
3385 7. Change 4.30/15 was not doing the test caselessly.
3387 8. The standard form for an IPv6 address literal was being rejected by address
3388 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3389 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3390 this, as well as the form without the "IPv6" on the front (but only when
3391 address literals are enabled, of course).
3393 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3395 10. Exim crashed if a message with an empty sender address specified by -f
3396 encountered a router with an errors_to setting. This could be provoked only
3397 by a command such as
3401 where an empty string was supplied; "<>" did not hit this bug.
3403 11. Installed PCRE release 4.5.
3405 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3406 remained set. It is now erased.
3408 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3409 times from message ids (which are base 36 rather than the normal 62).
3411 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3412 were being counted as actual protocol errors, and logged if the log
3413 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3414 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3415 and DATA following a set of rejected RCPTs do not count as protocol errors.
3416 In other words, Exim assumes they were pipelined, though this may not
3417 actually be the case. Of course, in all cases the client gets an
3418 appropriate error code.
3420 15. If a lookup fails in an ACL condition, a message about the failure may
3421 be available; it is used if testing the ACL cannot continue, because most
3422 such messages specify what the cause of the deferral is. However, some
3423 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3424 that caused an old message to be retained and used if a later statement
3425 caused a defer, replacing the real cause of the deferral.
3427 16. If an IP address had so many PTR records that the DNS lookup buffer
3428 was not large enough to hold them, Exim could crash while trying to process
3429 the truncated data. It now detects and logs this case.
3431 17. Further to 4.21/58, another change has been made: if (and only if) the
3432 first line of a message (the first header line) ends with CRLF, a bare LF
3433 in a subsequent header line has a space inserted after it, so as not to
3434 terminate the header.
3436 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3437 unnecessarily, used atoi() instead of strtol(), and didn't check the
3438 termination when getting file sizes from file names by regex.
3440 19. Completely re-implemented the support for maildirsize files, in the light
3441 of a number of problems with the previous contributed implementation
3442 (4.30/29). In particular:
3444 . If the quota is zero, the maildirsize file is maintained, but no quota is
3447 . If the maildir directory does not exist, it is created before any attempt
3448 to write a maildirsize file.
3450 . The quota value in the file is just a cache; if the quota is changed in
3451 the transport, the new value overrides.
3453 . A regular expression is available for excluding directories from the
3456 20. The autoreply transport checks the characters in options that define the
3457 message's headers; it allows continued headers, but it was checking with
3458 isspace() after an embedded newline instead of explicitly looking for a
3461 21. If all the "regular" hosts to which an address was routed had passed their
3462 expiry times, and had not reached their retry times, the address was
3463 bounced, even if fallback hosts were defined. Now Exim should go on to try
3466 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3467 equivalent code in the SMTP transport. Some hosts send humungous responses
3468 to HELO/EHLO, more than 1024 it seems.
3470 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3471 gives compiler warnings in some environments. I've now done it "properly",
3474 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3475 (because of problems with the built-in one) was declared to return uschar *
3476 instead of char *, causing compiler failure.
3478 25. Fixed a file descriptor leak when processing alias/forward files.
3480 26. Fixed a minor format string issue in dbfn.c.
3482 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3484 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3485 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3488 29. When a sender address is verified, it is cached, to save repeating the test
3489 when there is more than one recipient in a message. However, when the
3490 verification involves a callout, it is possible for different callout
3491 options to be set for different recipients. It is too complicated to keep
3492 track of this in the cache, so now Exim always runs a verification when a
3493 callout is required, relying on the callout cache for the optimization.
3494 The overhead is duplication of the address routing, but this should not be
3497 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3498 to be verified with callout=postmaster, and the main callout worked but the
3499 postmaster check failed, the verification correctly failed. However, if a
3500 subsequent RCPT command asked for sender verification *without* the
3501 postmaster check, incorrect caching caused this verification also to fail,
3504 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3505 it was not caching the DNS options (qualify_single, search_parents) that
3506 were used when the lookup failed. A subsequent lookup with different
3507 options therefore always gave the same answer, though there were cases
3508 where it should not have. (Example: a "domains = !$mx_any" option on a
3509 dnslookup router: the "domains" option is always processed without any
3510 widening, but the router might have qualify_single set.) Now Exim uses the
3511 cached value only when the same options are set.
3513 32. Added John Jetmore's "exipick" utility to the distribution.
3515 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3516 than a timeout (e.g. a certificate is required, and is not provided), an
3517 Exim server now closes the connection immediately. Previously it waited for
3518 the client to close - but if the client is SSL, it seems that they each
3519 wait for each other, leading to a delay before one of them times out.
3521 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3522 maintained 0.8.x compatibility because I don't think many are using it, and
3523 it is clearly obsolete.
3525 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3528 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3529 client certificate was expired. A simple patch fixes this, though I don't
3530 understand the full logic of why the verify callback is called multiple
3533 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3534 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3535 which causes problems with some clients (such as the Certicom SSL Plus
3536 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3537 disables the coutermeasure allowing Eudora to connect."
3539 38. Exim was not checking that a write() to a log file succeeded. This could
3540 lead to Bad Things if a log got too big, in particular if it hit a file
3541 size limit. Exim now panics and dies if it cannot write to a log file, just
3542 as it does if it cannot open a log file.
3544 39. Modified OS/Makefile-Linux so that it now contains
3546 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3548 The two -D definitions ensure that Exim is compiled with large file
3549 support, which makes it possible to handle log files that are bigger than
3552 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3553 instance) a domain was checked against a named list that involved a lookup,
3554 causing $domain_data to be set, then another domain was checked against the
3555 same list, then the first domain was re-checked, the value of $domain_data
3556 after the final check could be wrong. In particular, if the second check
3557 failed, it could be set empty. This bug probably also applied to
3560 41. The strip_trailing_dot option was not being applied to the address given
3561 with the -f command-line option.
3563 42. The code for reading a message's header from the spool was incrementing
3564 $received_count, but never initializing it. This meant that the value was
3565 incorrect (doubled) while delivering a message in the same process in which
3566 it was received. In the most common configuration of Exim, this never
3567 happens - a fresh exec is done - but it can happen when
3568 deliver_drop_privilege is set.
3570 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3571 it now includes up to 150 characters of the unexpected data in the log
3574 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3575 and building data strings. The size of both of these buffers was 10 000
3576 bytes - far larger than anybody would *ever* want, thought I. Needless to
3577 say, somebody hit the limit. I have increased the maximum line length to
3578 20 000 and the maximum data length of concatenated lines to 100 000. I have
3579 also fixed two bugs, because there was no checking on these buffers. Tsk,
3580 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3581 buffer is too small.
3583 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3584 lsearch lookups. Now it does.
3586 46. When parsing a route_list item in a manualroute router, a fixed-length
3587 buffer was used for the list of hosts. I made this 1024 bytes long,
3588 thinking that nobody would ever have a list of hosts that long. Wrong.
3589 Somebody had a whole pile of complicated expansion conditions, and the
3590 string was silently truncated, leading to an expansion error. It turns out
3591 that it is easier to change to an unlimited length (owing to other changes
3592 that have happened since this code was originally written) than to build
3593 structure for giving a limitation error. The length of the item that
3594 expands into the list of hosts is now unlimited.
3596 47. The lsearch lookup could not handle data where the length of text line was
3597 more than 4095 characters. Such lines were truncated, leading to shortened
3598 data being returned. It should now handle lines of any length.
3600 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3601 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3604 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3605 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3606 no longer happens in comment lines. A list of replacements is now placed
3607 at the head of all of the source files, except those whose only change is
3608 to replace PERL_COMMAND in the very first #! line.
3610 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3611 messages on the queue, with a bottom-up merge sort, using code contributed
3612 by Michael Haardt. This should make operations like -bp somewhat faster on
3613 large queues. It won't affect queue runners, except when queue_run_in_order
3616 51. Installed eximstats 1.31 in the distribution.
3618 52. Added support for SRV lookups to the dnslookup router.
3620 53. If an ACL referred to $message_body or $message_body_end, the value was not
3621 reset for any messages that followed in the same SMTP session.
3623 54. The store-handling optimization for building very long strings was not
3624 differentiating between the different store pools. I don't think this
3625 actually made any difference in practice, but I've tidied it.
3627 55. While running the routers to verify a sender address, $sender_address
3628 was still set to the sender address. This is wrong, because when routing to
3629 send a bounce to the sender, it would be empty. Therefore, I have changed
3630 it so that, while verifying a sender address, $sender_address is set to <>.
3631 (There is no change to what happens when verifying a recipient address.)
3633 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3634 target A or AAAA records (if not already returned) without resetting the
3635 qualify_single or search_parents options of the DNS resolver. These are
3636 inappropriate in this case because the targets of MX and SRV records must
3637 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3638 target that, when qualified, matched something in the local domain. These
3639 two options are now turned off when doing these lookups.
3641 57. It seems that at least some releases of Reiserfs (which does not have the
3642 concept of a fixed number of inodes) returns zero and not -1 for the
3643 number of available inodes. This interacted badly with check_spool_inodes,
3644 which assumed that -1 was the "no such thing" setting. What I have done is
3645 to check that the total number of inodes is greater than zero before doing
3646 the test of how many are available.
3648 58. When a "warn" ACL statement has a log_message modifier, the message is
3649 remembered, and not repeated. This is to avoid a lot of repetition when a
3650 message has many recipients that cause the same warning to be written.
3651 Howewer, Exim was preserving the list of already written lines for an
3652 entire SMTP session, which doesn't seem right. The memory is now reset if a
3653 new message is started.
3655 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3656 debugging output unless log_rewrite was also set.
3658 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3659 of (int)(handle) when we know that handle contains (void *)(-1).
3661 61. The Exim daemon panic-logs an error return when it closes the incoming
3662 connection. However "connection reset by peer" seems to be common, and
3663 isn't really an error worthy of noting specially, so that particular error
3666 62. When Exim is trying to find all the local interfaces, it used to panic and
3667 die if the ioctl to get the interface flags failed. However, it seems that
3668 on at least one OS (Solaris 9) it is possible to have an interface that is
3669 included in the list of interfaces, but for which you get a failure error
3670 for this call. This happens when the interface is not "plumbed" into a
3671 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3672 failure of the "get flags" call assumes that the interface is down.
3674 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3675 makes life easier for people who are doing arithmetic on fields extracted
3676 from dates, where you often get leading zeros that should not be
3677 interpreted as octal.
3679 64. Added qualify_domain to the redirect router, to override the global
3682 65. If a pathologically long header line contained very many addresses (the
3683 report of this problem mentioned 10 000) and each of them was rewritten,
3684 Exim could use up a very large amount of memory. (It kept on making new
3685 copies of the header line as it rewrote, and never released the old ones.)
3686 At the expense of a bit more processing, the header rewriting function has
3687 been changed so that it no longer eats memory in this way.
3689 66. The generation of the Received: header has been moved from the time that a
3690 message starts to be received, to the time that it finishes. The timestamp
3691 in the Received: header should now be very close to that of the <= log
3692 line. There are two side-effects of this change:
3694 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3695 logged header lines no longer include the local Received: line, because
3696 it has not yet been created. The same applies to a copy of the message
3697 that is returned to a non-SMTP sender when a message is rejected.
3699 (b) When a filter file is tested using -bf, no additional Received: header
3700 is added to the test message. After some thought, I decided that this
3703 This change does not affect the value of $received_for. It is still set
3704 after address rewriting, but before local_scan() is called.
3706 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3708 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3709 gave an unhelpful panic error message, and a defer error. I have managed to
3710 change this behaviour so that it now rejects any supplied certificate,
3711 which seems right, as the list of acceptable certificates is empty.
3713 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3714 gave an unhelpful defer error. I have not managed to make this reject any
3715 supplied certificates, but the error message it gives is "no certificate
3716 supplied", which is not helpful.
3718 70. exigrep's output now also includes lines that are not associated with any
3719 message, but which match the given pattern. Implemented by a patch from
3720 Martin Sluka, which also tidied up the Perl a bit.
3722 71. Recipient callout verification, like sender verification, was using <> in
3723 the MAIL FROM command. This isn't really the right thing, since the actual
3724 sender may affect whether the remote host accepts the recipient or not. I
3725 have changed it to use the actual sender in the callout; this means that
3726 the cache record is now keyed on a recipient/sender pair, not just the
3727 recipient address. There doesn't seem to be a real danger of callout loops,
3728 since a callout by the remote host to check the sender would use <>.
3729 [SEE ABOVE: changed after hitting problems.]
3731 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3732 temporary errors. However, in the case of such a code being given after
3733 the end of a data transmission (i.e. after ".") Exim was failing to write
3734 a retry record for the message. (Yes, there was some broken host that was
3735 actually sending 8xx at this point.)
3737 73. An unknown lookup type in a host list could cause Exim to panic-die when
3738 the list was checked. (An example that provoked this was putting <; in the
3739 middle of a list instead of at the start.) If this happened during a DATA
3740 ACL check, a -D file could be left lying around. This kind of configuration
3741 error no longer causes Exim to die; instead it causes a defer errror. The
3742 incident is still logged to the main and panic logs.
3744 74. Buglet left over from Exim 3 conversion. The message "too many messages
3745 in one connection" was written to the rejectlog but not the mainlog, except
3746 when address rewriting (yes!) was being logged.
3748 75. Added write_rejectlog option.
3750 76. When a system filter was run not as root (that is, when system_filter_user
3751 was set), the values of the $n variables were not being returned to the
3752 main process; thus, they were not subsequently available in the $sn
3755 77. Added +return_path_on_delivery log selector.
3757 78. A connection timeout was being treated differently from recipients deferred
3758 when testing hosts_max_try with a message that was older than the host's
3759 retry timeout. (The host should not be counted, thus allowing all hosts to
3760 be tried at least once before bouncing.) This may have been the cause of an
3761 occasionally reported bug whereby a message would remain on the queue
3762 longer than the retry timeout, but would be bounced if a delivery was
3763 forced. I say "may" because I never totally pinned down the problem;
3764 setting up timeout/retry tests is difficult. See also the next item.
3766 79. The ultimate address timeout was not being applied to errors that involved
3767 a combination of host plus message (for example, a timeout on a MAIL
3768 command). When an address resolved to a number of possible hosts, and they
3769 were not all tried for each delivery (e.g. because of hosts_max_try), a
3770 message could remain on the queue longer than the retry timeout.
3772 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3775 81. Fixed an obscure SMTP outgoing bug which required at least the following
3776 conditions: (a) there was another message waiting for the same server;
3777 (b) the server returned 5xx to all RCPT commands in the first message so
3778 that the message was not completed; (c) the server dropped the connection
3779 or gave a negative response to the RSET that Exim sends to abort the
3780 transaction. The observed case was a dropped connection after DATA that had
3781 been sent in pipelining mode. That is, the server had advertised PIPELINING
3782 but was not implementing it correctly. The effect of the bug was incorrect
3783 behaviour, such as trying another host, and this could lead to a crash.
3789 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3790 and daemon.c were passed as pointers to ints; they should have been
3791 pointers to socklen_t variables (which are typically unsigned ints).
3793 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3796 3. Fixed a really odd bug that affected only the testing scheme; patching a
3797 certain fixed string in the binary changed the value of another string that
3798 happened to be identical to the end of the original first string.
3800 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3801 name", it returns that address as the IP address. On some operating
3802 systems (e.g. Solaris), it also passes back the IP address string as the
3803 "host name". However, on others (e.g. Linux), it passes back an empty
3804 string. Exim wasn't checking for this, and was changing the host name to an
3805 empty string, assuming it had been canonicized.
3807 5. Although rare, it is permitted to have more than one PTR record for a given
3808 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3809 all the names associated with an address, because they do in Solaris.
3810 However, it seems that they do not in Linux for data that comes from the
3811 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3812 I found this out when I moved to a new Linux workstation and tried to run
3813 the Exim test suite.
3815 To get round this problem I have changed the code so that it now does its
3816 own call to the DNS to look up PTR records when searching for a host name.
3817 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3818 addresses that are only in /etc/hosts are still found.
3820 This behaviour is, however, controlled by an option called host_lookup_
3821 order, which defaults to "bydns:byaddr". If people want to use the other
3822 order, or indeed, just use one or the other means of lookup, they can
3823 specify it in this variable.
3825 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3826 some operating systems, this comes back from gethostbyaddr() as an empty
3827 string, and this is what Exim used to test for. However, it seems that in
3828 other systems, "." is yielded. Exim now tests for this case too.
3830 7. The values of check_spool_space and check_log_space are now held internally
3831 as a number of kilobytes instead of an absolute number of bytes. If a
3832 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3833 kilobyte. This means that much larger values can be stored.
3835 8. Exim monitor: an attempt to get the action menu when not actually pointing
3836 at a message produces an empty menu entitled "No message selected". This
3837 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3838 no entries in it ("Shell widget menu has zero width and/or height"). So I
3839 have added a single, blank menu entry in this case.
3841 9. Added ${quote_local_part.
3843 10. MIME decoding is now applied to the contents of Subject: header lines when
3846 11. Now that a reference to $sender_host_address automatically causes a reverse
3847 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3848 host lookup before query-style lookups in lists that might use this
3849 variable. This has therefore been abolished, and the "net-" prefix is no
3850 longer necessary for query-style lookups.
3852 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3853 have been a typo for LFLAGS, so it has been changed.
3855 13. The install script calls Exim with "-C /dev/null" in order to find the
3856 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3857 to be output. Howeve, since Exim outputs its version number before the
3858 error, it didn't break the script. It just looked ugly. I fixed this by
3859 always allowing "-C /dev/null" if the caller is root.
3861 14. Ignore overlarge ACL variable number when reading spool file - insurance
3862 against a later release with more variables having written the file.
3864 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3865 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3866 this, as well as the form without the "IPv6" on the front.
3868 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3869 OS/Makefile-Darwin file.
3871 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3872 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3874 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3875 variable. However, some OS use ints for inodes. I've added cast to long int
3876 to get rid of the compiler warning.
3878 19. I had forgotten to lock out "/../" in configuration file names when
3879 ALT_CONFIG_PREFIX was set.
3881 20. Routers used for verification do not need to specify transports. However,
3882 if such a router generated a host list, and callout was configured, Exim
3883 crashed, because it could not find a port number from the (non-existent)
3884 transport. It now assumes port 25 in this circumstance.
3886 21. Added the -t option to exigrep.
3888 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3889 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3890 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3891 them is set, LOOKUP_LSEARCH is forced.
3893 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3895 24. Added sender and host information to the "rejected by local_scan()" log
3896 line; previously there was no indication of these.
3898 25. Added .include_if_exists.
3900 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3901 Linux. It turns out that not all file systems support this. Apparently some
3902 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3903 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3904 not supported on the file descriptor, is now ignored when Exim is trying to
3905 sync a directory. This applies only to Linux.
3907 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3909 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3910 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3913 29. Added support for maildirsize files from supplied patch (modified a bit).
3915 30. The use of :fail: followed by an empty string could lead Exim to respond to
3916 sender verification failures with (e.g.):
3918 550 Verification failed for <xxx>
3919 550 Sender verify failed
3921 where the first response line was missing the '-' that indicates it is not
3922 the final line of the response.
3924 31. The loop for finding the name of the user that called Exim had a hardwired
3925 limit of 10; it now uses the value of finduser_retries, which is used for
3926 all other user lookups.
3928 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3931 33. Exim was neglecting to zero errno before one call of strtol() when
3932 expanding a string and expecting an integer value. On some systems this
3933 resulted in spurious "integer overflow" errors. Also, it was casting the
3934 result into an int without checking.
3936 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3937 did not work. The code looks as if it has *never* worked, though it appears
3938 to have been documented since at least releast 1.62. I have made it work.
3940 35. The "timeout_DNS" error in retry rules, also documented since at least
3941 1.62, also never worked. As it isn't clear exactly what this means, and
3942 clearly it isn't a major issue, I have abolished the feature by treating it
3943 as "timeout", and writing a warning to the main and panic logs.
3945 36. The display of retry rules for -brt wasn't always showing the error code
3948 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3949 timeout_connect_A, timeout_connect_MX.
3951 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3952 to the empty sender.
3954 39. The daemon was not analysing the content of -oX till after it had closed
3955 stderr and disconnected from the controlling terminal. This meant that any
3956 syntax errors were only noted on the panic log, and the return code from
3957 the command was 0. By re-arranging the code a little, I've made the
3958 decoding happen first, so such errors now appear on stderr, and the return
3959 code is 1. However, the actual setting up of the sockets still happens in
3960 the disconnected process, so errors there are still only recorded on the
3963 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3964 connections (as happens on some IP stacks) was logged at start up time as
3965 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3966 it from "IPv6 and IPv4", which means that two separate sockets are being
3969 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3970 says whether AF_INET or AF_INET6 was passed as an argument.
3972 42. Exiwhat output was messed up when time zones were included in log
3975 43. Exiwhat now gives more information about the daemon's listening ports,
3976 and whether -tls-on-connect was used.
3978 44. The "port" option of the smtp transport is now expanded.
3980 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3981 silently ignored. Now an error message is written to the main and panic
3984 46. There's a new ACL modifier called "logwrite" which writes to a log file
3985 as soon as it is encountered.
3987 47. Added $local_user_uid and $local_user_gid at routing time.
3989 48. Exim crashed when trying to verify a sender address that was being
3992 49. Exim was recognizing only a space character after ".include". It now also
3993 recognizes a tab character.
3995 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3996 extracting the relevant information from the specification. The man page no
3997 longer contains scrambled data for the -d option, and I've added a section
3998 at the front about calling Exim under different names.
4000 51. Added "extra_headers" argument to the "mail" command in filter files.
4002 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4005 53. Installed eximstats 1.29.
4007 54. Added transport_filter_timeout as a generic transport option.
4009 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4010 Cc: header lines. This was required by RFC 822, but it not required by RFC
4013 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4014 incoming messages that did not have them. Now it does so only if the
4015 message originates locally, that is, if there is no associated remote host
4016 address. When Resent- header lines are present, this applies to the Resent-
4017 lines rather than the non-Resent- lines.
4019 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4020 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4022 58. Messages for configuration errors now include the name of the main
4023 configuration file - useful now that there may be more than one file in a
4024 list (.included file names were always shown).
4026 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4027 for those rare installations that do not start the daemon as root or run it
4028 setuid root. I've cut out the call to initgroups() if the daemon is not
4031 60. The Exim user and group can now be bound into the binary as text strings
4032 that are looked up at the start of Exim's processing.
4034 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4036 62. Added $mailstore_basename variable.
4038 63. Installed patch to sieve.c from Michael Haardt.
4040 64. When Exim failed to open the panic log after failing to open the main log,
4041 the original message it was trying to log was written to stderr and debug
4042 output, but if they were not available (the usual case in production), it
4043 was lost. Now it is written to syslog before the two lines that record the
4044 failures to open the logs.
4046 65. Users' Exim filters run in subprocesses under the user's uid. It is
4047 possible for a "deliver" command or an alias in a "personal" command to
4048 provoke an address rewrite. If logging of address rewriting is configured,
4049 this fails because the process is not running as root or exim. There may be
4050 a better way of dealing with this, but for the moment (because 4.30 needs
4051 to be released), I have disabled address rewrite logging when running a
4052 filter in a non-root, non-exim process.
4058 1. The buildconfig auxiliary program wasn't quoting the value set for
4059 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4060 not defined. This bug was masked in 4.22 by the effect that was fixed in
4063 2. Some messages that were rejected after a message id was allocated were
4064 shown as "incomplete" by exigrep. It no longer does this for messages that
4065 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4067 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4068 have allow_domain_literals set, the ID did not get logged in the <= line.
4069 Domain literals are now always recognized in Message-ID: header lines.
4071 4. The first argument for a ${extract expansion item is the key name or field
4072 number. Leading and trailing spaces in this item were not being ignored,
4073 causing some misleading effects.
4075 5. When deliver_drop_privilege was set, single queue runner processes started
4076 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4077 same command in the process it spins off) were not dropping privilege.
4079 6. When the daemon running as "exim" started a queue runner, it always
4080 re-executed Exim in the spun-off process. This is a waste of effort when
4081 deliver_drop_privilege is set. The new process now just calls the
4082 queue-runner function directly.
4088 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4091 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4092 ignored. Though the use of -oP was forcing the writing of a pid file, it
4093 was always written to the default place.
4095 3. If the message "no IP address found for host xxxx" is generated during
4096 incoming verification, it is now followed by identification of the incoming
4097 connection (so you can more easily find what provoked it).
4099 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4101 5. Added some features to "harden" Exim a bit more against certain attacks:
4103 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4104 be put in Local/Makefile. This is like the never_users runtime option,
4105 but it cannot be overridden. The default setting is "root".
4107 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4108 prefix string with which any file named in a -C command line option
4111 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4112 is retained for -C and -D only if the caller of Exim is root. Without
4113 it, the exim user may also use -C and -D and retain privilege.
4115 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4116 command line option is disabled.
4118 6. Macro names set by the -D option must start with an upper case letter, just
4119 like macro names defined in the configuration file.
4121 7. Added "dereference=" facility to LDAP.
4123 8. Two instances of the typo "uknown" in the source files are fixed.
4125 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4126 the Configure-Makefile script screwed up while processing it.
4128 10. Incorporated PCRE 4.4.
4130 11. The SMTP synchronization check was not operating right at the start of an
4131 SMTP session. For example, it could not catch a HELO sent before the client
4132 waited for the greeting. There is now a check for outstanding input at the
4133 point when the greeting is written. Because of the duplex, asynchronous
4134 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4135 way, but not yet received, when the check is performed.
4137 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4138 on TCP/IP sockets, because this apparently causes some broken clients to
4141 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4142 unchanged) from the Cygwin maintainer.
4144 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4145 was supported instead of testing for mbx. Effectively a typo.
4147 15. The spa authenticator server code was not checking that the input it
4148 received was valid base64.
4150 16. The debug output line for the "set" modifier in ACLs was not showing the
4151 name of the variable that was being set.
4153 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4155 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4156 lookup on its own. Something else was needed to trigger the lookup. For
4157 example, a match in host_lookup or the need for a host name in a host list.
4158 Now, if $sender_host_name is referenced and the host name has not yet been
4159 looked up, a lookup is performed. If the lookup fails, the variable remains
4160 empty, and $host_lookup_failed is set to "1".
4162 19. Added "eqi" as a case-independent comparison operator.
4164 20. The saslauthd authentication condition could segfault if neither service
4165 nor realm was specified.
4167 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4168 error message that was logged was misleading, and incoming SMTP
4169 connections were dropped. The message is now more accurate, and temporary
4170 errors are given to SMTP connections.
4172 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4173 (except RSET) with a 421 error, until QUIT is received. However, it was
4174 failing to send a response to QUIT.
4176 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4177 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4178 ACL is now run after the helo_try_verify_hosts code.
4180 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4181 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4182 case-independent, so other case variants are also recognized). Apparently
4183 some systems use these upper case variants.
4185 25. If more than two messages were waiting for the same host, and a transport
4186 filter was specified for the transport, Exim sent two messages over the
4187 same TCP/IP connection, and then failed with "socket operation on non-
4188 socket" when it tried to send the third.
4190 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4192 27. The extern definition of crypt16() in expand.c was not being excluded when
4193 the OS had its own crypt16() function.
4195 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4196 as a preferred synonym for return_size_limit, both as an option and as an
4199 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4201 30. Changed the default configuration ACL to relax the local part checking rule
4202 for addresses that are not in any local domains. For these addresses,
4203 slashes and pipe symbols are allowed within local parts, but the sequence
4204 /../ is explicitly forbidden.
4206 31. SPA server authentication was not clearing the challenge buffer before
4209 32. log_message in a "warn" ACL statement was writing to the reject log as
4210 well as to the main log, which contradicts the documentation and doesn't
4211 seem right (because no rejection is happening). So I have stopped it.
4213 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4214 However, I am unable to do any testing of this.
4216 34. Fixed an infelicity in the appendfile transport. When checking directories
4217 for a mailbox, to see if any needed to be created, it was accidentally
4218 using path names with one or more superfluous leading slashes; tracing
4219 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4221 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4222 contents are added to the log line that is written for every discarded
4223 recipient. (Previously a log_message setting was ignored.)
4225 36. The ${quote: operator now quotes the string if it is empty.
4227 37. The install script runs exim in order to find its version number. If for
4228 some reason other than non-existence or emptiness, which it checks, it
4229 could not run './exim', it was installing it with an empty version number,
4230 i.e. as "exim-". This error state is now caught, and the installation is
4233 38. An argument was missing from the function that creates an error message
4234 when Exim fails to connect to the socket for saslauthd authentication.
4235 This could cause Exim to crash, or give a corrupted message.
4237 39. Added isip, isip4, and isip6 to ${if conditions.
4239 40. The ACL variables $acl_xx are now saved with the message, and can be
4240 accessed later in routers, transports, and filters.
4242 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4243 strings in the file are not string-expanded.
4245 42. If a MAIL command specified a SIZE value that was too large to fit into an
4246 int variable, the check against message_size_limit failed. Such values are
4247 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4248 day this will have to be increased, but I don't think I want to be around
4249 when emails are that large.
4256 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4257 iconv() is not standard in FreeBSD.
4259 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4260 IPv6 enabled. The observed symptom was a segmentation fault on return from
4261 the function os_common_find_running_interfaces() in src/os.c.
4263 3. In the check_special_case() function in daemon.c I had used "errno" as an
4264 argument name, which causes warnings on some systems. This was basically a
4265 typo, since it was named "eno" in the comments!
4267 4. The code that waits for the clock to tick (at a resolution of some fraction
4268 of a second) so as to ensure message-id uniqueness was always waiting for
4269 at least one whole tick, when it could have waited for less. [This is
4270 almost certainly not relevant at current processor speeds, where it is
4271 unlikely to ever wait at all. But we try to future-proof.]
4273 5. The function that sleeps for a time interval that includes fractions of a
4274 second contained a race. It did not block SIGALRM between setting the
4275 timer, and suspending (a couple of lines later). If the interval was short
4276 and the sigsuspend() was delayed until after it had expired, the suspension
4277 never ended. On busy systems this could lead to processes getting stuck for
4280 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4281 process, before it forks any delivery processes. The open lookup caching
4282 mechanism meant that the open file or database connection was passed into
4283 the delivery process. The problem was that delivery processes always tidy
4284 up cached lookup data. This could cause a problem for the next delivery
4285 process started by the queue runner, because the external queue runner
4286 process does not know about the closure. So the next delivery process
4287 still has data in the lookup cache. In the case of a file lookup, there was
4288 no problem because closing a file descriptor in a subprocess doesn't affect
4289 the parent. However, if the lookup was caching a connection to a database,
4290 the connection was closed, and the second delivery process was likely to
4291 see errors such as "PGSQL: query failed: server closed the connection
4292 unexpectedly". The problem has been fixed by closing all cached lookups
4293 in a queue runner before running a delivery process.
4295 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4296 seem to have the "const" qualifier which it has on other OS. I've
4299 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4300 *of the same type* (client or server) with the same public name that an
4301 error should be diagnosed.
4303 9. When Exim looked up a host name for an IP address, but failed to find the
4304 original IP address when looking up the host name (a safety check), it
4305 output the message "<ip address> does not match any IP for NULL", which was
4306 confusing, to say the least. The bug was that the host name should have
4307 appeared instead of "NULL".
4309 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4310 user that is built into the binary, and the -C or -D options is used, root
4311 privilege is dropped before the configuration file is read. In addition,
4312 logging is switched to stderr instead of the normal log files. If the
4313 configuration then re-defines the Exim user, the unprivileged environment
4314 is probably not what is expected, so Exim logs a panic warning message (but
4317 However, if deliver_drop_privilege is set, the unprivileged state may well
4318 be exactly what is intended, so the warning has been cut out in that case,
4319 and Exim is allowed to try to write to its normal log files.
4325 1. smtp_return_error_details was not giving details for temporary sender
4326 or receiver verification errors.
4328 2. Diagnose a configuration error if two authenticators have the same public
4331 3. Exim used not to create the message log file for a message until the first
4332 delivery attempt. This could be confusing when incoming messages were held
4333 for policy or load reasons. The message log file is now created at the time
4334 the message is received, and an initial "Received" line is written to it.
4336 4. The automatically generated man page for command line options had a minor
4337 bug that caused no ill effects; however, a more serious problem was that
4338 the procedure for building the man page automatically didn't always
4339 operate. Consequently, release 4.20 contains an out-of-date version. This
4340 shouldn't happen again.
4342 5. When building Exim with embedded Perl support, the script that builds the
4343 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4344 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4346 6. The freeze_tell option was not being used for messages that were frozen on
4347 arrival, either by an ACL or by local_scan().
4349 7. Added the smtp_incomplete_transaction log selector.
4351 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4352 was accepting AUTH without a new EHLO.
4354 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4355 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4358 10. Logging of TCP/IP connections (when configured) now happens in the main
4359 daemon process instead of the child process, so that the TCP/IP connection
4360 count is more accurate (but it can never be perfect).
4362 11. The use of "drop" in a nested ACL was not being handled correctly in the
4363 outer ACL. Now, if condition failure induced by the nested "drop" causes
4364 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4365 or "require"), the connection is dropped.
4367 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4368 that yield "discard" can now be used with an "accept" or a "discard" verb,
4369 but an error is generated for any others (because I can't see a useful way
4370 to define what should happen).
4372 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4373 are now processed in the same way as lines in the Exim configuration file.
4374 In particular, continuation lines are supported.
4376 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4378 15. Added -ti meaning -t -i.
4380 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4381 domains, and warn by logging if others are found.
4383 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4384 returned from GIFCONF when Exim is trying to find the list of interfaces on
4385 a host. The code in os.c has been modified to copy each ifreq to an aligned
4386 structure in all cases.
4388 Also, in some cases, the returned ifreq's were being copied to a 'struct
4389 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4390 means the last couple of bytes of an IPv6 address could be chopped if the
4391 ifreq contained only a normal sockaddr (14 bytes storage).
4393 18. Named domain lists were not supported in the hosts_treat_as_local option.
4394 An entry such as +xxxx was not recognized, and was treated as a literal
4397 19. Ensure that header lines added by a DATA ACL are included in the reject log
4398 if the ACL subsequently rejects the message.
4400 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4401 MD5 (which is deprecated).
4403 21. When testing a filter file using -bf, Exim was writing a message when it
4404 took the sender from a "From " line in the message, but it was not doing so
4405 when it took $return_path from a Return-Path: header line. It now does.
4407 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4408 with a valid header line field name (a series of printing characters
4409 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4411 23. Changed "disc" in the source to "disk" to conform to the documentation and
4412 the book and for uniformity.
4414 24. Ignore Sendmail's -Ooption=value command line item.
4416 25. When execve() failed while trying to run a command in a pipe transport,
4417 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4418 could be confused with a return value of 69 from the command itself. This
4419 has been changed to 127, the value the shell returns if it is asked to run
4420 a non-existent command. The wording for the related log line suggests a
4421 non-existent command as the problem.
4423 26. If received_header_text expands to an empty string, do not add a Received:
4424 header line to the message. (Well, it adds a token one on the spool, but
4425 marks it "old" so that it doesn't get used or transmitted.)
4427 27. Installed eximstats 1.28 (addition of -nt option).
4429 28. There was no check for failure on the call to getsockname() in the daemon
4430 code. This can fail if there is a shortage of resources on the system, with
4431 ENOMEM, for example. A temporary error is now given on failure.
4433 29. Contrary to the C standard, it seems that in some environments, the
4434 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4435 program. Exim now does this explicitly; it affects the formatting of
4436 timestamps using strftime().
4438 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4439 complaints. I've now initialized all the variables, to avoid this.
4441 32. Header lines added by a system filter were not being "seen" during
4442 transport-time rewrites.
4444 33. The info_callback() function passed to OpenSSL is set up with type void
4445 (*)(SSL *, int, int), as described somewhere. However, when calling the
4446 function (actually a macro) that sets it up, the type void(*)() is
4447 expected. I've put in a cast to prevent warnings from picky compilers.
4449 34. If a DNS black list lookup found a CNAME record, but there were no A
4450 records associated with the domain it pointed at, Exim crashed.
4452 35. If a DNS black list lookup returned more than one A record, Exim ignored
4453 all but the first. It now scans all returned addresses if a particular IP
4454 value is being sought. In this situation, the contents of the
4455 $dnslist_value variable are a list of all the addresses, separated by a
4458 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4459 to accept a host name and all its aliases if the forward lookup for any of
4460 them yielded the IP address of the incoming connection. Now it accepts only
4461 those names whose forward lookup yields the correct IP address. Any other
4462 names are discarded. This closes a loophole whereby a rogue DNS
4463 administrator could create reverse DNS records to break through a
4464 wildcarded host restriction in an ACL.
4466 37. If a user filter or a system filter that ran in a subprocess used any of
4467 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4468 the wrong values were passed to the pipe command ($thisaddress had the
4469 value of $0, $0 had the value of $1, etc). This bug was introduced by
4470 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4472 38. Improved the line breaking for long SMTP error messages from ACLs.
4473 Previously, if there was no break point between 40 and 75 characters, Exim
4474 left the rest of the message alone. Two changes have been made: (a) I've
4475 reduced the minimum length to 35 characters; (b) if it can't find a break
4476 point between 35 and 75 characters, it looks ahead and uses the first one
4477 that it finds. This may give the occasional overlong line, but at least the
4478 remaining text gets split now.
4480 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4481 file descriptors might be low, and that setting 1000 would always raise it.
4482 It turns out that in some environments, the limit is already over 1000 and
4483 that lowering it causes trouble. So now Exim takes care not to decrease it.
4485 40. When delivering a message, the value of $return_path is set to $sender_
4486 address at the start of routing (routers may change the value). By an
4487 oversight, this default was not being set up when an address was tested by
4488 -bt or -bv, which affected the outcome if any router or filter referred to
4491 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4492 log only when "message" or "log_message" are set. However, if one of the
4493 conditions was an address verification, or a call to a nested ACL, the
4494 messages generated by the underlying test were being passed through. This
4495 no longer happens. The underlying message is available in $acl_verify_
4496 message for both "message" and "log_message" expansions, so it can be
4497 passed through if needed.
4499 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4500 new expansion $bh_ to give the encoded byte string without charset
4501 translation. Translation happens only if iconv() is available; HAVE_ICONV
4502 indicates this at build time. HEADERS_CHARSET gives the charset to
4503 translate to; headers_charset can change it in the configuration, and
4504 "headers charset" can change it in an individual filter file.
4506 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4507 that creates RFC 2047 encoded "words" labels them as that charset instead
4508 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4509 expansion operator; (ii) when Exim creates a From: line for a local
4510 message; (iii) when a header line is rewritten to include a "phrase" part.
4512 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4513 buggy, causing it to skip the first lines of messages whose message ID
4514 ended in 'D'. This would not have bitten before Exim release 4.14, because
4515 message IDs were unlikely to end in 'D' before then. The effect was to have
4516 incorrect size information for certain domains.
4518 45. #include "config.h" was missing at the start of the crypt16.c module. This
4519 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4522 46. If there was a timeout during a "random" callout check, Exim treated it as
4523 a failure of the random address, and carried on sending RSET and the real
4524 address. If the delay was just some slowness somewhere, the response to the
4525 original RCPT would be taken as a response to RSET and so on, causing
4526 mayhem of various kinds.
4528 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4529 when I implemented it. It didn't allow for the fact that some option values
4530 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4531 the right test for positive values.
4533 48. Domain names in DNS records are case-independent. Exim always looks them up
4534 in lower case. Some resolvers return domain names in exactly the case they
4535 appear in the zone file, that is, they may contain uppercase letters. Not
4536 all resolvers do this - some return always lower case. Exim was treating a
4537 change of case by a resolver as a change of domain, similar to a widening
4538 of a domain abbreviation. This triggered its re-routing code and so it was
4539 trying to route what was effectively the same domain again. This normally
4540 caused routing to fail (because the router wouldn't handle the domain
4541 twice). Now Exim checks for this case specially, and just changes the
4542 casing of the domain that it ultimately uses when it transmits the message
4545 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4548 50. If a filter generated a file delivery with a non-absolute name (possible if
4549 no home directory exists for the router), the forbid_file option was not
4552 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4553 the existing equality matching.
4555 52. Exim was using ints instead of ino_t variables in some places where it was
4556 dealing with inode numbers.
4558 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4559 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4560 called TMPDIR, and if it finds it is different, it changes its value.
4562 54. The smtp_printf() function is now made available to local_scan() so
4563 additional output lines can be written before returning. There is also an
4564 smtp_fflush() function to enable the detection of a dropped connection.
4565 The variables smtp_input and smtp_batched_input are exported to
4568 55. Changed the default runtime configuration: the message "Unknown user"
4569 has been removed from the ACL, and instead placed on the localuser router,
4570 using the cannot_route_message feature. This means that any verification
4571 failures that generate their own messages won't get overridden. Similarly,
4572 the "Unrouteable address" message that was in the ACL for unverifiable
4573 relay addresses has also been removed.
4575 56. Added hosts_avoid_esmtp to the smtp transport.
4577 57. The exicyclog script was not checking for the esoteric option
4578 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4579 will work only if exicyclog is run under the appropriate euid.
4581 58. Following a discussion on the list, the rules by which Exim recognises line
4582 endings on incoming messages have been changed. The -dropcr and drop_cr
4583 options are now no-ops, retained only for backwards compatibility. The
4584 following line terminators are recognized: LF CRLF CR. However, special
4585 processing applies to CR:
4587 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4588 nor a local message in the state where . is a terminator.
4590 (ii) If a bare CR is encountered in a header line, an extra space is added
4591 after the line terminator so as not to end the header. The reasoning
4592 behind this is that bare CRs in header lines are most likely either
4593 to be mistakes, or people trying to play silly games.
4595 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4596 was being incorrectly given as 18 bytes larger than it should have been.
4597 This is a VOB (very old bug).
4599 60. This may never have affected anything current, but just in case it has:
4600 When the local host is found other than at the start of a list of hosts,
4601 the local host, those with the same MX, and any that follow, are discarded.
4602 When the list in question was part of a longer list of hosts, the following
4603 hosts (not currently being processed) were also being discarded. This no
4604 longer happens. I'm not sure if this situation could ever has previously
4607 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4610 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4611 header. This is something that is recommended in a new Internet Draft, and
4612 is something that is documented as being done by Sendmail. There are two
4613 possible values. For messages generated by the autoreply transport, Exim
4616 Auto-Submitted: auto-replied
4618 whereas for all other generated messages (e.g. bounces) it adds
4620 Auto-Submitted: auto-generated
4622 63. The "personal" condition in filters now includes a test for the
4623 Auto-Submitted: header. If it contains the string "auto-" the message it
4624 not considered personal.
4626 64. Added rcpt_include_affixes as a generic transport option.
4628 65. Added queue_only_override (default true).
4630 66. Added the syslog_duplication option.
4632 67. If what should have been the first header line of a message consisted of
4633 a space followed by a colon, Exim was mis-interpreting it as a header line.
4634 It isn't of course - it is syntactically invalid and should therefore be
4635 treated as the start of the message body. The misbehaviour could have
4636 caused a number of strange effects, including loss of data in subsequent
4637 header lines, and spool format errors.
4639 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4640 client host had authenticated. This control can now be exercised by an ACL
4641 for more flexibility.
4643 69. By default, callouts do not happen when testing with -bh. There is now a
4644 variant, -bhc, which does actually run the callout code, including
4645 consulting and updating the callout cache.
4647 70. Added support for saslauthd authentication, courtesy of Alexander
4650 71. If statvfs() failed on the spool or log directories while checking their
4651 size for availability, Exim confusingly gave the error "space shortage".
4652 Furthermore, in debugging mode it crashed with a floating point exception.
4653 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4654 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4655 serious problem, Exim now writes to the main and panic logs when this
4656 happens, with details of the failure. It then refuses to accept the
4657 incoming message, giving the message "spool directory problem" or "log
4658 directory problem" with a 421 code for SMTP messages.
4660 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4661 0, 1, and 2 exist, because some OS complain for execs without them (see
4662 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4663 descriptors. However, the code omitted to check that the open succeeded,
4664 causing mysterious errors if for some reason the permissions on /dev/null
4665 got screwed. Now Exim writes a message to the main and panic logs, and
4666 bombs out if it can't open /dev/null.
4668 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4669 interact so that it is all more flexible. It is supposed to remain
4670 backwards compatible. Also added extra_local_interfaces.
4672 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4673 to bomb out with an assertion failure - to the client this appears as a
4674 connection drop. This problem occurs in the part of the code that was taken
4675 from the Samba project. Fortunately, the assertion is in a very simple
4676 function, so I have fixed this by reproducing the function inline in the
4677 one place where it is called, and arranging for authentication to fail
4678 instead of killing the process with assert().
4680 75. The SPA client code was not working when the server requested OEM rather
4681 than Unicode encoding.
4683 76. Added code to make require_files with a specific uid setting more usable in
4684 the case where statting the file as root fails - usually a non-root-mounted
4685 NFS file system. When this happens and the failure is EACCES, Exim now
4686 forks a subprocess and does the per-uid checking as the relevant uid.
4688 77. Added process_log_path.
4690 78. If log_file_path was not explicitly set, a setting of check_log_space or
4691 check_log_inodes was ignored.
4693 79. If a space check for the spool or log partitions fails, the incident is now
4694 logged. Of course, in the latter case the data may get lost...
4696 80. Added the %p formatting code to string_format() so that it can be used to
4697 print addresses in debug_print(). Adjusted all the address printing in the
4698 debugging in store.c to use %p rather than %d.
4700 81. There was a concern that a line of code in smtp_in.c could overflow a
4701 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4702 initially expressed, the concern was not well-founded, because trailing
4703 spaces are removed early. However, if the trailing spaces were followed by
4704 a NULL, they did not get removed, so the overflow was possible. Two fixes
4707 (a) I re-wrote the offending code in a cleaner fashion.
4708 (b) If an incoming SMTP command contains a NULL character, it is rejected
4711 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4712 runs initgroups(), so that if the Exim user is in any additional groups,
4713 they will be used during message reception.
4719 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob