1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.501 2007/04/12 09:00:51 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
10 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
13 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
16 PH/02 In an ACL statement such as
18 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
20 if a client was not listed at all, or was listed with a value other than
21 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
22 the condition was not true (as it should be), so access was not denied.
23 The bug was that the ! inversion was incorrectly passed on to the second
24 item. This has been fixed.
26 PH/03 Added additional dnslists conditions == and =& which are different from
27 = and & when the dns lookup returns more than one IP address.
29 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
30 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
32 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
33 FSYNC, which compiles an option called disable_fsync that allows for
34 bypassing fsync(). The documentation is heavily laced with warnings.
36 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
38 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
39 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
40 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
41 including adding "make clean"; (3) Added -fPIC when compiling the test
42 dynamically loaded module, to get rid of a warning.
44 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
45 message fails, move_frozen_messages = true and ignore_bounce_errors_after
46 = 0s. The bug is otherwise harmless.
48 PH/07 There was a bug in the dovecot authenticator such that the value of
49 $auth1 could be overwritten, and so not correctly preserved, after a
50 successful authentication. This usually meant that the value preserved by
51 the server_setid option was incorrect.
53 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
55 PH/09 Installed PCRE release 7.0.
57 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
58 run for batched SMTP input. It is now run at the start of every message
59 in the batch. While fixing this I discovered that the process information
60 (output by running exiwhat) was not always getting set for -bs and -bS
61 input. This is fixed, and it now also says "batched" for BSMTP.
63 PH/11 Added control=no_pipelining.
65 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
66 patch, slightly modified), and move the expansion of helo_data till after
67 the connection is made in the smtp transport (so it can use these
70 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
72 PH/14 Added log_selector = +pid.
74 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
76 PH/16 Add ${if forany and ${if forall.
78 PH/17 Added dsn_from option to vary the From: line in DSNs.
80 PH/18 Flush SMTP output before performing a callout, unless control =
81 no_callout_flush is set.
83 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
84 was true (the default) a successful delivery failed to delete the retry
85 item, thus causing premature timeout of the address. The bug is now
88 PH/20 Added hosts_avoid_pipelining to the smtp transport.
90 PH/21 Long custom messages for fakedefer and fakereject are now split up
91 into multiline reponses in the same way that messages for "deny" and
92 other ACL rejections are.
94 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
95 with slight modification.
97 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
98 draft, changing the syntax and factoring some duplicate code".
100 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
101 for deliveries of the second and subsequent messages over the same SMTP
104 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
105 ${reduce, with only minor "tidies".
107 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
109 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
110 expansion side effects.
112 PH/27 When a message times out after an over-quota error from an Exim-imposed
113 quota, the bounce message says "mailbox is full". This message was not
114 being given when it was a system quota that was exceeded. It now should
117 MH/03 Made $recipients available in local_scan(). local_scan() already has
118 better access to the recipient list through recipients_list[], but
119 $recipients can be useful in postmaster-provided expansion strings.
121 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
122 in the case of a MAIL command with additional options following the
123 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
124 were accidentally chopped off.
126 PH/29 SMTP synchronization checks are implemented when a command is read -
127 there is a check that no more input is waiting when there shouldn't be
128 any. However, for some commands, a delay in an ACL can mean that it is
129 some time before the response is written. In this time, more input might
130 arrive, invalidly. So now there are extra checks after an ACL has run for
131 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
132 pipelining has not been advertised.
134 PH/30 MH's patch to allow iscntrl() characters to be list separators.
136 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
137 returned in the SMTP response when smtp_return_error_details was false.
140 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
141 instead of the C I/O that was originally supplied, because problems were
144 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
145 Exim which did not show up earlier: it was assuming that a call to
146 SSL_CTX_set_info_callback() might give an error value. In fact, there is
147 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
148 was a macro that became an assignment, so it seemed to work. This has
149 changed to a proper function call with a void return, hence the compile
150 error. Exim's code has been fixed.
152 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
155 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
157 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
159 PH/37 If a message is not accepted after it has had an id assigned (e.g.
160 because it turns out to be too big or there is a timeout) there is no
161 "Completed" line in the log. When some messages of this type were
162 selected by exigrep, they were listed as "not completed". Others were
163 picked up by some special patterns. I have improved the selection
164 criteria to be more general.
166 PH/38 The host_find_failed option in the manualroute router can now be set
167 to "ignore", to completely ignore a host whose IP address cannot be
168 found. If all hosts are ignored, the behaviour is controlled by the new
169 host_all_ignored option.
171 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
172 homing or because of multiple MX records with /mx) generated more than
173 one IP address, and the following item turned out to be the local host,
174 all the secondary addresses of the first item were incorrectly removed
175 from the list, along with the local host and any following hosts (which
176 is what is supposed to happen).
178 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
179 whoever called Exim into the -H file. In the case of the daemon it was
180 behaving confusingly. When first started, it used values for whoever
181 started the daemon, but after a SIGHUP it used the Exim user (because it
182 calls itself on a restart). I have changed the code so that it now always
185 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
186 message are rejected with the same error (e.g. no authentication or bad
187 sender address), and a DATA command is nevertheless sent (as can happen
188 with PIPELINING or a stupid MUA), the error message that was given to the
189 RCPT commands is included in the rejection of the DATA command. This is
190 intended to be helpful for MUAs that show only the final error to their
193 PH/42 Another patch from the Sieve maintainer.
195 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
196 Eximstats - Fixed some broken HTML links and added missing column headers
198 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
199 columns for Rejects, Temp Rejects, Ham, and Spam rows.
201 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
203 PH/43 Yet another patch from the Sieve maintainer.
209 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
210 fixed by 4.65/MH/01 (is this a record?) are fixed:
212 (i) An empty string was always treated as zero by the numeric comparison
213 operators. This behaviour has been restored.
215 (ii) It is documented that the numeric comparison operators always treat
216 their arguments as decimal numbers. This was broken in that numbers
217 starting with 0 were being interpreted as octal.
219 While fixing these problems I realized that there was another issue that
220 hadn't been noticed. Values of message_size_limit (both the global option
221 and the transport option) were treated as octal if they started with 0.
222 The documentation was vague. These values are now always treated as
223 decimal, and I will make that clear in the documentation.
229 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
230 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
233 MH/01 Don't check that the operands of numeric comparison operators are
234 integers when their expansion is in "skipping" mode (fixes bug
235 introduced by 4.64-PH/07).
237 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
238 child addresses, Exim now panics and dies. Previously, because the count
239 is held in a short int, deliveries were likely to be lost. As such a
240 large number of recipients for a single message is ridiculous
241 (performance will be very, very poor), I have chosen to impose a limit
242 rather than extend the field.
248 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
249 leftover -K file (the existence of which was triggered by #402).
250 While we were at it, introduced process PID as part of the -K
251 filename. This should rule out race conditions when creating
254 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
255 processing considerably. Previous code took too long for large mails,
256 triggering a timeout which in turn triggers #401.
258 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
259 in the DK code in transports.c. sendfile() is not really portable,
260 hence the _LINUX specificness.
262 TF/01 In the add_headers option to the mail command in an Exim filter,
263 there was a bug that Exim would claim a syntax error in any
264 header after the first one which had an odd number of characters
267 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
268 callout verification, Exim cached a "reject" for the entire domain. This
269 is correct for most verifications, but it is not correct for a recipient
270 verification with use_sender or use_postmaster set, because in that case
271 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
272 case of MAIL FROM:<> rejection from other early rejections (e.g.
273 rejection of HELO). When verifying a recipient using a non-null MAIL
274 address, the cache is ignored if it shows MAIL FROM:<> rejection.
275 Whatever the result of the callout, the value of the domain cache is
276 left unchanged (for any other kind of callout, getting as far as trying
277 RCPT means that the domain itself is ok).
279 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
282 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
283 manifest itself as EPIPE rather than ECONNECT. When tidying away a
284 session, the daemon ignores ECONNECT errors and logs others; it now
285 ignores EPIPE as well.
287 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
288 (quoted-printable decoding).
290 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
291 later the small subsequent patch to fix an introduced bug.
293 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
295 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
297 PH/08 An error is now given if message_size_limit is specified negative.
299 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
300 to be given (somewhat) arbitrary names.
302 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
305 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
306 miscellaneous code fixes
308 PH/10 Added the log_reject_target ACL modifier to specify where to log
311 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
312 hostname. This is wrong, because it relates to the incoming message (and
313 probably the interface on which it is arriving) and not to the outgoing
314 callout (which could be using a different interface). This has been
315 changed to use the value of the helo_data option from the smtp transport
316 instead - this is what is used when a message is actually being sent. If
317 there is no remote transport (possible with a router that sets up host
318 addresses), $smtp_active_hostname is used.
320 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
321 tweaks were necessary in order to get it to work (see also 21 below):
322 (a) The code assumed that strncpy() returns a negative number on buffer
323 overflow, which isn't the case. Replaced with Exim's string_format()
325 (b) There were several signed/unsigned issues. I just did the minimum
326 hacking in of casts. There is scope for a larger refactoring.
327 (c) The code used strcasecmp() which is not a standard C function.
328 Replaced with Exim's strcmpic() function.
329 (d) The code set only $1; it now sets $auth1 as well.
330 (e) A simple test gave the error "authentication client didn't specify
331 service in request". It would seem that Dovecot has changed its
332 interface. Fortunately there's a specification; I followed it and
333 changed what the client sends and it appears to be working now.
335 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
338 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
339 address A is aliased to B and C, where B exists and C does not. Without
340 -v the output is "A verified" because verification stops after a
341 successful redirection if more than one address is generated. However,
342 with -v the child addresses are also verified. Exim was outputting "A
343 failed to verify" and then showing the successful verification for C,
344 with its parentage. It now outputs "B failed to verify", showing B's
345 parentage before showing the successful verification of C.
347 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
348 look up a TXT record in a specific list after matching in a combined
351 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
352 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
353 they consult the DNS. I had assumed they would set it the way they
354 wanted; and indeed my experiments on Linux seem to show that in some
355 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
356 To be on the safe side, however, I have now made the interface to
357 host_find_byname() similar to host_find_bydns(), with an argument
358 containing the DNS resolver options. The host_find_byname() function now
359 sets these options at its start, just as host_find_bydns() does. The smtp
360 transport options dns_qualify_single and dns_search_parents are passed to
361 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
362 of host_find_byname() use the default settings of RES_DEFNAMES
363 (qualify_single) but not RES_DNSRCH (search_parents).
365 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
366 spool_read_header() do less string testing, by means of a preliminary
367 switch on the second character of optional "-foo" lines. (This is
368 overdue, caused by the large number of possibilities that now exist.
369 Originally there were few.) While I was there, I also converted the
370 str(n)cmp tests so they don't re-test the leading "-" and the first
371 character, in the hope this might squeeze out yet more improvement.
373 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
374 flag allowing group syntax was set by the header_syntax check but not
375 turned off, possible causing trouble later; (2) The flag was not being
376 set at all for the header_verify test, causing "group"-style headers to
377 be rejected. I have now set it in this case, and also caused header_
378 verify to ignore an empty address taken from a group. While doing this, I
379 came across some other cases where the code for allowing group syntax
380 while scanning a header line wasn't quite right (mostly, not resetting
381 the flag correctly in the right place). These bugs could have caused
382 trouble for malformed header lines. I hope it is now all correct.
384 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
385 with the "reply" argument non-NULL. The code, however (which originally
386 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
387 but it didn't always do it. This confused somebody who was copying the
388 code for some other use. I have removed all the tests.
390 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
391 feature that was used to support insecure browsers during the U.S. crypto
392 embargo. It requires special client support, and Exim is probably the
393 only MTA that supported it -- and would never use it because real RSA is
394 always available. This code has been removed, because it had the bad
395 effect of slowing Exim down by computing (never used) parameters for the
396 RSA_EXPORT functionality.
398 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
399 authenticator to fail if there's a tab character in the incoming data
400 (there should never be unless someone is messing about, as it's supposed
401 to be base64-encoded). Also added, on Timo's advice, the "secured" option
402 if the connection is using TLS or if the remote IP is the same as the
403 local IP, and the "valid-client-cert option" if a client certificate has
406 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
407 authenticators. This can be used for authorization after authentication
408 succeeds. (In the case of plaintext, it servers for both authentication
411 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
412 if any retry times were supplied.
414 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
415 connection, where there is no client IP address to check. In this
416 situation, the verify now always succeeds.
418 PH/25 Applied John Jetmore's -Mset patch.
420 PH/26 Added -bem to be like -Mset, but loading a message from a file.
422 PH/27 In a string expansion for a processed (not raw) header when multiple
423 headers of the same name were present, leading whitespace was being
424 removed from all of them, but trailing whitespace was being removed only
425 from the last one. Now trailing whitespace is removed from each header
426 before concatenation. Completely empty headers in a concatenation (as
429 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
430 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
432 PH/29 [Removed. This was a change that I later backed out, and forgot to
433 correct the ChangeLog entry (that I had efficiently created) before
434 committing the later change.]
436 PH/30 Exim was sometimes attempting to deliver messages that had suffered
437 address errors (4xx response to RCPT) over the same connection as other
438 messages routed to the same hosts. Such deliveries are always "forced",
439 so retry times are not inspected. This resulted in far too many retries
440 for the affected addresses. The effect occurred only when there were more
441 hosts than the hosts_max_try setting in the smtp transport when it had
442 the 4xx errors. Those hosts that it had tried were not added to the list
443 of hosts for which the message was waiting, so if all were tried, there
444 was no problem. Two fixes have been applied:
446 (i) If there are any address or message errors in an SMTP delivery, none
447 of the hosts (tried or untried) are now added to the list of hosts
448 for which the message is waiting, so the message should not be a
449 candidate for sending over the same connection that was used for a
450 successful delivery of some other message. This seems entirely
451 reasonable: after all the message is NOT "waiting for some host".
452 This is so "obvious" that I'm not sure why it wasn't done
453 previously. Hope I haven't missed anything, but it can't do any
454 harm, as the worst effect is to miss an optimization.
456 (ii) If, despite (i), such a delivery is accidentally attempted, the
457 routing retry time is respected, so at least it doesn't keep
458 hammering the server.
460 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
461 in ${readsocket because some servers need this prod.
463 PH/32 Added some extra debug output when updating a wait-xxx database.
465 PH/33 The hint "could be header name not terminated by colon", which has been
466 given for certain expansion errors for a long time, was not being given
467 for the ${if def:h_colon_omitted{... case.
469 PH/34 The spec says: "With one important exception, whenever a domain list is
470 being scanned, $domain contains the subject domain." There was at least
471 one case where this was not true.
473 PH/35 The error "getsockname() failed: connection reset by peer" was being
474 written to the panic log as well as the main log, but it isn't really
475 panic-worthy as it just means the connection died rather early on. I have
476 removed the panic log writing for the ECONNRESET error when getsockname()
479 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
480 runs only) independently of the message's sender address. This meant
481 that, if the 4xx error was in fact related to the sender, a different
482 message to the same recipient with a different sender could confuse
483 things. In particualar, this can happen when sending to a greylisting
484 server, but other circumstances could also provoke similar problems.
485 I have changed the default so that the retry time for these errors is now
486 based a combination of the sender and recipient addresses. This change
487 can be overridden by setting address_retry_include_sender=false in the
490 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
491 remote server are returned as part of bounce messages. This was not
492 happening for LMTP over a pipe (the lmtp transport), but now it is the
493 same for both kinds of LMTP.
495 PH/38 Despite being documented as not happening, Exim was rewriting addresses
496 in header lines that were in fact CNAMEs. This is no longer the case.
498 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
499 and queue runs started by the daemon processed all messages. This has
500 been fixed so that -R and -S can now usefully be given with -q<time>.
502 PH/40 Import PCRE release 6.7 (fixes some bugs).
504 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
506 PH/42 Give an error if -q is specified more than once.
508 PH/43 Renamed the variables $interface_address and $interface_port as
509 $received_ip_address and $received_port, to make it clear that these
510 values apply to message reception, and not to the outgoing interface when
511 a message is delivered. (The old names remain recognized, of course.)
513 PH/44 There was no timeout on the connect() call when using a Unix domain
514 socket in the ${readsocket expansion. There now is.
516 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
517 be meaningful with "accept".
519 SC/01 Eximstats V1.43
520 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
522 SC/02 Eximstats V1.44
523 Use a glob alias rather than an array ref in the generated
524 parser. This improves both readability and performance.
526 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
527 Collect SpamAssassin and rejection statistics.
528 Don't display local sender or destination tables unless
529 there is data to show.
530 Added average volumes into the top table text output.
532 SC/04 Eximstats V1.46
533 Collect data on the number of addresses (recipients)
534 as well as the number of messages.
536 SC/05 Eximstats V1.47
537 Added 'Message too big' to the list of mail rejection
538 reasons (thanks to Marco Gaiarin).
540 SC/06 Eximstats V1.48
541 Mainlog lines which have GMT offsets and are too short to
542 have a flag are now skipped.
544 SC/07 Eximstats V1.49 (Alain Williams)
545 Added the -emptyok flag.
547 SC/08 Eximstats V1.50
548 Fixes for obtaining the IP address from reject messages.
550 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
551 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
552 whitesspace changes from 4.64-PH/27
554 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
557 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
558 are found, allow negative numbers in numeric criteria)
560 JJ/06 exipick.20061117.2, added new $message_body_missing variable
562 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
563 to match changes made in 4.64-PH/43
565 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
567 PH/47 Put in an explicit test for a DNS lookup of an address record where the
568 "domain" is actually an IP address, and force a failure. This locks out
569 those revolvers/nameservers that support "A-for-A" lookups, in
570 contravention of the specifications.
572 PH/48 When a host name was looked up from an IP address, and the subsequent
573 forward lookup of the name timed out, the host name was left in
574 $sender_host_name, contrary to the specification.
576 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
577 restricted to single-key lookups, Exim was not diagnosing an error if
578 * or *@ was used with a query-style lookup.
580 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
582 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
583 long ago, but noone interested enough thought of it. Let's just say that
584 the "1.1" means that there are some new functions that weren't there at
585 some point in the past.
587 PH/51 Error processing for expansion failure of helo_data from an smtp
588 transport during callout processing was broken.
590 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
591 tested/used via the -bh/-bhc/-bs options.
593 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
594 bug, fixed in subsequent PCRE releases).
596 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
597 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
599 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
605 SC/01 Use a glob alias rather than an array ref in eximstats generated
606 parser. This improves both readability and performance.
608 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
609 Don't display local sender or destination tables in eximstats unless
610 there is data to show.
611 Added average volumes into the eximstats top table text output.
613 SC/03 Collect data on the number of addresses (recipients) as well
614 as the number of messages in eximstats.
616 TF/01 Correct an error in the documentation for the redirect router. Exim
617 does (usually) call initgroups() when daemonizing.
619 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
620 with consistent privilege compared to when running as a daemon.
622 TF/03 Note in the spec that $authenticated_id is not set for local
623 submissions from trusted users.
625 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
626 Thanks to Dean Brooks <dean@iglou.com> for the patch.
628 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
629 by adding some example configuration directives to the default
630 configuration file. A little bit of work is required to uncomment the
631 directives and define how usernames and passwords are checked, but
632 there is now a framework to start from.
634 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
635 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
636 without this. I don't know how relevant this is to other LDAP libraries.
638 PH/02 Add the verb name to the "unknown ACL verb" error.
640 PH/03 Magnus Holmgren's patch for filter_prepend_home.
642 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
644 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
645 directory not expanded when it should be if an expanded home directory
646 was set for the address (which is overridden by the transport).
648 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
651 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
652 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
653 because it is too late at that time, and has no effect.
655 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
656 security issue with \' (bugzilla #107). I could not use the
657 PQescapeStringConn() function, because it needs a PGconn value as one of
660 PH/08 When testing addresses using -bt, indicate those final addresses that
661 are duplicates that would not cause an additional delivery. At least one
662 person was confused, thinking that -bt output corresponded to deliveries.
663 (Suppressing duplicates isn't a good idea as you lose the information
664 about possibly different redirections that led to the duplicates.)
666 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
667 systems where poll() doesn't work, in particular OS X.
669 PH/10 Added more information to debugging output for retry time not reached.
671 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
672 operations in malware.c.
674 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
677 PH/13 If write_rejectlog was set false when logging was sent to syslog with
678 syslog_duplication set false, log lines that would normally be written
679 both the the main log and to the reject log were not written to syslog at
682 PH/14 In the default configuration, change the use of "message" in ACL warn
683 statements to "add_header".
685 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
686 not followed by a command (e.g. "seen endif").
688 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
689 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
692 PH/17 Added extra conditions to the default value of delay_warning_condition
696 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
697 { match{$h_precedence:}{(?i)bulk|list|junk} } \
698 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
701 The Auto-Submitted: and various List- headers are standardised, whereas I
702 don't think Precedence: ever was.
704 PH/18 Refactored debugging code in route_finduser() to show more information,
705 in particular, the error code if getpwnam() issues one.
707 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
708 This is apparently needed in addition to the PH/07 change above to avoid
709 any possible encoding problems.
711 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
712 but not after initializing Perl.
714 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
715 output them only if debugging. By default they are written stderr,
716 apparently, which is not desirable.
718 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
721 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
724 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
726 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
727 authenticated or an ident call has been made. Suppress the default
728 values for $authenticated_id and $authenticated_sender (but permit -oMai
729 and -oMas) when testing with -bh.
731 PH/24 Re-jigged the order of the tests in the default configuration so that the
732 tests for valid domains and recipients precede the DNS black list and CSA
733 tests, on the grounds that those ones are more expensive.
735 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
736 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
737 This bug exists in every version of Exim that I still have, right back to
740 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
741 However, an attempt to turn on case-sensitivity in a regex key by
742 including (?-i) didn't work because the subject string was already
743 lowercased, and the effects were non-intuitive. It turns out that a
744 one-line patch can be used to allow (?-i) to work as expected.
750 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
751 other effects) broke the use of negated acl sub-conditions.
753 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
756 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
757 "Deny" causes Exim to reject the incoming connection with a 554 error.
758 Unfortunately, if there is a major crisis, such as a disk failure,
759 tcp-wrappers gives "deny", whereas what one would like would be some
760 kind of temporary error. A kludge has been added to help with this.
761 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
762 554 error is used if errno is still zero or contains ENOENT (which occurs
763 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
766 PH/03 Add -lutil to the default FreeBSD LIBS setting.
768 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
769 errors. Otherwise a message that provokes a temporary error (when other
770 messages do not) can cause a whole host to time out.
772 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
773 addresses were routed directly to files or pipes from a redirect router.
774 File deliveries just didn't batch; pipe deliveries might have suffered
777 PH/06 A failure to get a lock for a hints database would erroneously always say
778 "Failed to get write lock", even when it was really a read lock.
780 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
781 of 0600. This has been changed to use the value of the lockfile_mode
782 option (which defaults to 0600).
784 PH/08 Applied small patch from the Sieve maintainer.
786 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
787 folder from quota calculations, a direct delivery into this folder messed
788 up the contents of the maildirsize file. This was because the regex was
789 used only to exclude .Trash (or whatever) when the size of the mailbox
790 was calculated. There was no check that a delivery was happening into an
791 excluded directory. This bug has been fixed by ignoring all quota
792 processing for deliveries into excluded directories.
794 PH/10 Added the maildirfolder_create_regex option to appendfile.
800 PH/01 The code for finding all the local interface addresses on a FreeBSD
801 system running IPv6 was broken. This may well have applied to all BSD
802 systems, as well as to others that have similar system calls. The broken
803 code found IPv4 interfaces correctly, but gave incorrect values for the
804 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
805 that it would not match correctly against @[] and not recognize the IPv6
808 PH/02 The ipliteral router was not recognizing addresses of the form user@
809 [ipv6:....] because it didn't know about the "ipv6:" prefix.
811 PH/03 Added disable_ipv6.
813 PH/04 Changed $reply_address to use the raw form of the headers instead of the
814 decoded form, because it is most often used to construct To: headers
815 lines in autoreplies, and the decoded form may well be syntactically
816 invalid. However, $reply_address has leading white space removed, and all
817 newlines turned into spaces so that the autoreply transport does not
820 PH/05 If group was specified without a user on a router, and no group or user
821 was specified on a transport, the group from the router was ignored.
823 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
824 for visible compile-time settings that can be used to change these
825 numbers, for those that want even more. Backwards compatibility with old
826 spool files has been maintained. However, going back to a previous Exim
827 release will lost any variables that are in spool files.
829 PH/07 Two small changes when running in the test harness: increase delay when
830 passing a TCP/IP connection to a new process, in case the original
831 process has to generate a bounce, and remove special handling of
832 127.0.0.2 (sic), which is no longer necessary.
834 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
835 be the same on different OS.
837 PH/09 Moved a debug statement in filter processing to avoid a race problem when
840 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
841 whether --show-vars was specified or not
843 JJ/02 exipick: Added support for new ACL variable spool format introduced
846 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
847 syntactically invalid From: or Reply-to: line, and a filter used this to
848 generate an autoreply, and therefore failed to obtain an address for the
849 autoreply, Exim could try to deliver to a non-existent relative file
850 name, causing unrelated and misleading errors. What now happens is that
851 it logs this as a hard delivery error, but does not attempt to create a
854 PH/11 The exinext utility has a -C option for testing purposes, but although
855 the given file was scanned by exinext itself; it wasn't being passed on
856 when Exim was called.
858 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
859 an end-of-file indication when reading a command response.
861 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
862 compiled. In many other places in Exim, IPv6 addresses are always
863 recognized, so I have changed this. It also means that IPv4 domain
864 literals of the form [IPV4:n.n.n.n] are now always recognized.
866 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
867 used if the router is not running as root, for example, when verifying at
868 ACL time, or when using -bh. The debugging output from this situation was
869 non-existent - all you got was a failure to exec. I have made two
872 (a) Failures to set uid/gid, the current directory, or a process leader
873 in a subprocess such as that created by queryprogram now generate
874 suitable debugging ouput when -d is set.
876 (b) The queryprogram router detects when it is not running as root,
877 outputs suitable debugging information if -d is set, and then runs
878 the subprocess without attempting to change uid/gid.
880 PH/15 Minor change to Makefile for building test_host (undocumented testing
883 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
884 additional section of a DNS packet that returns MX or SRV records.
885 Instead, it always explicitly searches for A/AAAA records. This avoids
886 major problems that occur when a DNS server includes only records of one
887 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
888 fixed another bug: if SRV records were looked up and the corresponding
889 address records were *not* found in the additional section, the port
890 values from the SRV records were lost.
892 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
893 using the correct key (the original address) when searching the retry
894 rules in order to find which one to use for generating the retry hint.
896 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
897 adding the default one. Similarly, if it contains a Reply-To: header, the
898 errors_reply_to option, if set, is not used.
900 PH/19 When calculating a retry time, Exim used to measure the "time since
901 failure" by looking at the "first failed" field in the retry record. Now
902 it does not use this if it is later than than the arrival time of the
903 message. Instead it uses the arrival time. This makes for better
904 behaviour in cases where some deliveries succeed, thus re-setting the
905 "first failed" field. An example is a quota failure for a huge message
906 when small messages continue to be delivered. Without this change, the
907 "time since failure" will always be short, possible causing more frequent
908 delivery attempts for the huge message than are intended.
909 [Note: This change was subsequently modified - see PH/04 for 4.62.]
911 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
912 $1, $2, $3) because the numerical variables can be reset during some
913 expansion items (e.g. "match"), thereby losing the authentication data.
915 PH/21 Make -bV show the size of off_t variables so that the test suite can
916 decide whether to run tests for quotas > 2G.
918 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
919 mailbox_size, and mailbox_filecount in the appendfile transport. If a
920 filecount value is greater than 2G or if a quota value is greater than 2G
921 on a system where the size of off_t is not greater than 4, a panic error
924 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
925 never match. The debug and -bh output now contains an explicit error
926 message indicating a malformed IPv4 address or mask.
928 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
929 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
932 PH/25 Do not write to syslog when running in the test harness. The only
933 occasion when this arises is a failure to open the main or panic logs
934 (for which there is an explicit test).
936 PH/26 Added the /no_tell option to "control=freeze".
938 PH/27 If a host name lookup failed very early in a connection, for example, if
939 the IP address matched host_lookup and the reverse lookup yielded a name
940 that did not have a forward lookup, an error message of the form "no IP
941 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
942 could be logged. Now it outputs the IP address instead of "NULL".
944 PH/28 An enabling patch from MH: add new function child_open_exim2() which
945 allows the sender and the authenticated sender to be set when
946 submitting a message from within Exim. Since child_open_exim() is
947 documented for local_scan(), the new function should be too.
949 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
950 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
951 results in an empty string is now treated as unset.
953 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
955 PH/31 Added sender_verify_fail logging option.
957 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
958 needed by RFC 822 but not by RFC 2822 was commented out. I have now
959 tidied the source and removed it altogether.
961 PH/33 When a queue run was abandoned because the load average was too high, a
962 log line was always written; now it is written only if the queue_run log
963 selector is set. In addition, the log line for abandonment now contains
964 information about the queue run such as the pid. This is always present
965 in "start" and "stop" lines but was omitted from the "abandon" line.
967 PH/34 Omit spaces between a header name and the colon in the error message that
968 is given when verify = headers_syntax fails (if there are lots of them,
969 the message gets confusing).
971 PH/35 Change the default for dns_check_names_pattern to allow slashes within
972 names, as there are now some PTR records that contain slashes. This check
973 is only to protect against broken name servers that fall over on strange
974 characters, so the fact that it applies to all lookups doesn't matter.
976 PH/36 Now that the new test suite is complete, we can remove some of the
977 special code in Exim that was needed for the old test suite. For example,
978 sorting DNS records because real resolvers return them in an arbitrary
979 order. The new test suite's fake resolver always returns records in the
982 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
983 bounces) except when queue_only is set, to avoid logging races between
984 the different processes.
986 PH/38 Panic-die if .include specifies a non-absolute path.
988 PH/39 A tweak to the "H" retry rule from its user.
990 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
991 a label. They prevented compilation on older perls.
993 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
994 a warning to be raised on newish perls.
996 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
997 on queue. Changes to match documented behaviour of showing count of
998 messages matching specified criteria.
1000 PH/40 Changed the default ident timeout from 30s to 5s.
1002 PH/41 Added support for the use of login_cap features, on those BSD systems
1003 that have them, for controlling the resources used by pipe deliveries.
1005 PH/42 The content-scanning code uses fopen() to create files in which to put
1006 message data. Previously it was not paying any attention to the mode of
1007 the files. Exim runs with umask(0) because the rest of the code creates
1008 files with open(), and sets the required mode explicitly. Thus, these
1009 files were ending up world-writeable. This was not a big issue, because,
1010 being within the spool directory, they were not world-accessible. I have
1011 created a function called modefopen, which takes an additional mode
1012 argument. It sets umask(777), creates the file, chmods it to the required
1013 mode, then resets the umask. All the relevant calls to fopen() in the
1014 content scanning code have been changed to use this function.
1016 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1017 to 24 hours. This avoids potential overflow problems when processing G
1018 and H retry rules. I suspect nobody ever tinkers with this value.
1020 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1022 PH/45 When the plaintext authenticator is running as a client, the server's
1023 challenges are checked to ensure they are valid base64 strings. By
1024 default, the authentication attempt is cancelled if an invalid string is
1025 received. Setting client_ignore_invalid_base64 true ignores these errors.
1026 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1027 they are received. Thus, the responses can be made to depend on the
1028 challenges. If an invalid string is ignored, an empty string is placed in
1031 PH/46 Messages that are created by the autoreply transport now contains a
1032 References: header, in accordance with RFCs 2822 and 3834.
1034 PH/47 Added authenticated_sender_force to the smtp transport.
1036 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1038 PH/49 Installed latest patch from the Sieve maintainer.
1040 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1041 was also set, the appendfile transport was unnecessarily scanning a
1042 directory of message files (e.g. for maildir delivery) to find the count
1043 of files (along with the size), even though it did not need this
1044 information. It now does the scan only if it needs to find either the
1045 size of the count of files.
1047 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1049 PH/52 Two bugs concerned with error handling when the smtp transport is
1052 (i) Exim was not creating retry information for temporary errors given
1053 for individual recipients after the DATA command when the smtp transport
1054 was used in LMTP mode. This meant that they could be retried too
1055 frequently, and not timed out correctly.
1057 (ii) Exim was setting the flag that allows error details to be returned
1058 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1059 recipients that were returned after the DATA command.
1061 PH/53 This is related to PH/52, but is more general: for any failing address,
1062 when detailed error information was permitted to be returned to the
1063 sender, but the error was temporary, then after the final timeout, only
1064 "retry timeout exceeded" was returned. Now it returns the full error as
1065 well as "retry timeout exceeded".
1067 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1068 do this, and (what is worse) MTAs that accept it.
1070 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1071 will now be deprecated.
1073 PH/56 New os.c-cygwin from the Cygwin maintainer.
1075 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1076 formats (previously only available in exim formats via -bpr, -bpru,
1077 and -bpra. Now also available in native and exiqgrep formats)
1079 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1080 with very large, slow to parse queues
1082 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1084 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1086 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1087 responses to authentication challenges, though it was showing the
1088 challenges; (ii) I've removed the CR characters from the debug output for
1091 PH/58 Allow for the insertion of a newline as well as a space when a string
1092 is turned into more than one encoded-word during RFC 2047 encoding. The
1093 Sieve code now uses this.
1095 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1096 data_4xx, lost_connection, tls_required.
1098 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1099 message was being sent as an SMTP response.
1101 PH/61 Add -l and -k options to exicyclog.
1103 PH/62 When verifying, if an address was redirected to one new address, so that
1104 verification continued, and the new address failed or deferred after
1105 having set something in $address_data, the value of $address_data was not
1106 passed back to the ACL. This was different to the case when no
1107 redirection occurred. The value is now passed back in both cases.
1109 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1110 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1111 use login_cap.h, so on its own it isn't the distinguishing feature. The
1112 new name refers directly to the setclassresources() function.
1114 PH/65 Added configuration files for NetBSD3.
1116 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1118 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1119 is preferred over IPv4.
1121 PH/68 The bounce_return_message and bounce_return_body options were not being
1122 honoured for bounces generated during the reception of non-SMTP messages.
1123 In particular, this applied to messages rejected by the ACL. This bug has
1124 been fixed. However, if bounce_return_message is true and bounce_return_
1125 body is false, the headers that are returned for a non-SMTP message
1126 include only those that have been read before the error was detected.
1127 (In the case of an ACL rejection, they have all been read.)
1129 PH/69 The HTML version of the specification is now built in a directory called
1130 spec_html instead of spec.html, because the latter looks like a path with
1131 a MIME-type, and this confuses some software.
1133 PH/70 Catch two compiler warnings in sieve.c.
1135 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1136 function verify_get_ident() calls ip_connect() to connect a socket, but
1137 if the "connect()" function timed out, ip_connect() used to close the
1138 socket. However, verify_get_ident() also closes the socket later, and in
1139 between Exim writes to the log, which may get opened at this point. When
1140 the socket was closed in ip_connect(), the log could get the same file
1141 descriptor number as the socket. This naturally causes chaos. The fix is
1142 not to close the socket in ip_connect(); the socket should be closed by
1143 the function that creates it. There was only one place in the code where
1144 this was missing, in the iplookup router, which I don't think anybody now
1145 uses, but I've fixed it anyway.
1147 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1148 well as to direct DNS lookups. Otherwise the handling of names in host
1149 lists is inconsistent and therefore confusing.
1155 PH/01 Two changes to the default runtime configuration:
1157 (1) Move the checks for relay_from_hosts and authenticated clients from
1158 after to before the (commented out) DNS black list checks.
1160 (2) Add control=submission to the relay_from_hosts and authenticated
1161 clients checks, on the grounds that messages accepted by these
1162 statements are most likely to be submissions.
1164 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1166 (1) Generate an error if the third argument for the ${prvs expansion is
1169 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1172 (3) Reset the variables that are obtained from the first argument of
1173 ${prvscheck and used in the second argument before leaving the code,
1174 because their memory is reclaimed, so using them afterwards may do
1177 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1178 one (it's much easier than Tom thought :-).
1180 (5) Because of (4), we can now allow for the use of $prvscheck_result
1181 inside the third argument.
1183 PH/03 For some reason, the default setting of PATH when running a command from
1184 a pipe transport was just "/usr/bin". I have changed it to
1187 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1188 anything to be listed in the output from -bV.
1190 PH/05 When a filter generated an autoreply, the entire To: header line was
1191 quoted in the delivery log line, like this:
1193 => >A.N.Other <ano@some.domain> <original@ddress> ...
1195 This has been changed so that it extracts the operative address. There
1196 may be more than one such address. If so, they are comma-separated, like
1199 => >ano@some.domain,ona@other.domain <original@ddress> ...
1201 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1202 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1203 not being looked up in the rDNS to get a host name, Exim was showing the
1204 IP address twice in Received: lines, even though the IP addresses were
1205 identical. For example:
1207 Received: from [1.2.3.4] (helo=[1.2.3.4])
1209 However, if the real host name was known, it was omitting the HELO data
1210 if it matched the actual IP address. This has been tidied up so that it
1211 doesn't show the same IP address twice.
1213 PH/07 When both +timestamp and +memory debugging was on, the value given by
1214 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1215 called by the debug printing, thereby overwriting the timestamp buffer.
1216 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1218 PH/08 When the original message was included in an autoreply transport, it
1219 always said "this is a copy of the message, including all the headers",
1220 even if body_only or headers_only was set. It now gives an appropriate
1223 PH/09 Applied a patch from the Sieve maintainer which:
1225 o fixes some comments
1226 o adds the (disabled) notify extension core
1227 o adds some debug output for the result of if/elsif tests
1228 o points to the current vacation draft in the documentation
1229 and documents the missing references header update
1233 o fixes a bug in processing the envelope test (when testing
1234 multiple envelope elements, the last element determinted the
1237 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1238 Electronic Mail") by including:
1240 Auto-submitted: auto-generated
1242 in the messages that it generates (bounce messages and others, such as
1243 warnings). In the case of bounce messages for non-SMTP mesages, there was
1244 also a typo: it was using "Auto_submitted" (underscore instead of
1245 hyphen). Since every message generated by Exim is necessarily in response
1246 to another message, thes have all been changed to:
1248 Auto-Submitted: auto-replied
1250 in accordance with these statements in the RFC:
1252 The auto-replied keyword:
1254 - SHOULD be used on messages sent in direct response to another
1255 message by an automatic process,
1257 - MUST NOT be used on manually-generated messages,
1259 - MAY be used on Delivery Status Notifications (DSNs) and Message
1260 Disposition Notifications (MDNs),
1262 - MUST NOT be used on messages generated by automatic or periodic
1263 processes, except for messages which are automatic responses to
1266 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1267 to the default Received: header definition.
1269 PH/12 Added log selector acl_warn_skipped (default on).
1271 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1272 variables because (a) they are in the wrong storage pool and (b) even if
1273 they were copied, it wouldn't work properly because of the caching.
1275 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1276 checking when decoding. Apparently there are clients that generate
1277 overlong encoded strings. Why am I not surprised?
1279 PH/15 If the first argument of "${if match_address" was not empty, but did not
1280 contain an "@" character, Exim crashed. Now it writes a panic log message
1281 and treats the condition as false.
1283 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1285 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1286 extension "envelope-auth". The code is finished and in agreement with
1287 other implementations, but there is no documentation so far and in fact,
1288 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1289 not changing the active code.
1291 Print executed "if" and "elsif" statements when debugging is used. This
1292 helps a great deal to understand what a filter does.
1294 Document more things not specified clearly in RFC3028. I had all this
1295 sorted out, when out of a sudden new issues came to my mind. Oops."
1297 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1300 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1301 canonical form (as documented). However, after a host name lookup from
1302 the IP address, check_host() was doing a simple string comparison with
1303 addresses acquired from the DNS when checking that the found name did
1304 have the original IP as one of its addresses. Since any found IPv6
1305 addresses are likely to be in abbreviated form, the comparison could
1306 fail. Luckily, there already exists a function for doing the comparison
1307 by converting both addresses to binary, so now that is used instead of
1308 the text comparison.
1310 PH/20 There was another similar case to PH/19, when a complete host name was
1311 given in a host list; looking up its IP address could give an abbreviated
1312 form, whereas the current host's name might or might not be abbreviated.
1313 The same fix has been applied.
1319 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1320 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1323 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1324 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1326 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1328 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1329 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1330 Also, for "Auto-Submitted", treat anything other than "no" as
1331 non-personal, in accordance with RFC 3834. (Previously it treated
1332 anything starting "auto-" as non-personal.)
1334 TF/01 The control=submission/name=... option had a problem with syntax
1335 errors if the name included a slash character. The /name= option
1336 now slurps the rest of the string, so it can include any characters
1337 but it must come last in the list of options (after /sender_retain
1340 PH/05 Some modifications to the interface to the fake nameserver for the new
1348 TK/01 Added the "success_on_redirect" address verification option. See
1349 NewStuff for rationale and an example.
1351 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1353 PH/02 Patch to exigrep to allow it to work on syslog lines.
1355 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1356 fread() to skip over the body file's header line, because in Cygwin the
1357 header line is locked and is inaccessible.
1359 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1360 co-exist for some time) to make it clear that it is the Exim ID that is
1361 referenced, not the Message-ID: header line.
1363 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1364 string_format() function, because snprintf() does not exist on all
1367 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1368 ${stat: expansion item.
1370 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1371 protocol synchronization error", to keep the pedants happy.
1373 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1374 well as for IRIX systems, when gcc is being used. See the host.c source
1377 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1379 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1382 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1383 transport and to the smtp transport in LMTP mode.
1385 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1387 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1388 run a filter in a subprocess. This could lead to confusion in subsequent
1389 lookups in the parent process. There should also be a search_tidyup() at
1390 the end of the subprocess.
1392 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1393 only if the host matched helo_try_verify_hosts, which caused the
1394 verification to occur when the EHLO/HELO command was issued. The ACL just
1395 tested the remembered result. Now, if a previous verification attempt has
1396 not happened, "verify = helo" does it there and then.
1398 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1400 TK/03 Fix log output including CR from clamd.
1402 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1403 exist provoked a memory error which could cause a segfault.
1405 PH/15 Installed PCRE 6.2
1407 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1409 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1410 of the problem. Specifically, suggested +O2 rather than +O1 for the
1413 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1415 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1416 record, the port was not correctly logged when the outgoing_port log
1417 selector was set (it logged the transort's default port).
1419 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1420 fallback_hosts, and "hosts" in the smtp transport.
1422 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1423 host errors such as "Connection refused".
1425 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1426 authentication with radiusclient 0.4.9:
1428 - Error returned from rc_read_config was caught wrongly
1429 - Username/password not passed on to radius server due to wrong length.
1431 The presumption is that some radiusclient API changes for 4.51/PH/17
1432 were not taken care of correctly. The code is still untested by me (my
1433 Linux distribution still has 0.3.2 of radiusclient), but it was
1434 contributed by a Radius user.
1436 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1437 expanding the "port" option of the smtp transport.
1439 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1440 while reading a MIME header. Thanks to Tom Hughes for a patch.
1442 PH/24 Include config.h inside local_scan.h so that configuration settings are
1445 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1446 that in an ACL for RCPT (for example), you can examine exactly what was
1449 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1450 commands, but it was not correctly comparing the address with the actual
1451 client host address. Thus, it would show the EHLO address in Received:
1452 header lines when this was not necessary.
1454 PH/27 Added the % operator to ${eval:}.
1456 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1457 it should be ignoring failures (because with -C, for example, it has lost
1458 privilege). It wasn't ignoring creation failures other than "already
1461 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1464 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1465 because an input line was too long, either on its own, or by virtue of
1466 too many continuations, the temporary file was not being removed, and the
1467 return code was incorrect.
1469 PH/31 Missing "BOOL" in function definition in filtertest.c.
1471 PH/32 Applied Sieve patches from the maintainer.
1473 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1475 PH/33 Added "verify = not_blind".
1477 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1478 Local/Makefile (with some defaults set). These are used in built scripts
1479 such as exicyclog, but they have never been used in the exim_install
1480 script (though there are many overriding facilities there). I have
1481 arranged that the exim_install script now takes note of these two
1484 PH/35 Installed configuration files for Dragonfly.
1486 PH/36 When a locally submitted message by a trusted user did not contain a
1487 From: header, and the sender address was obtained from -f or from an SMTP
1488 MAIL command, and the trusted user did not use -F to supply a sender
1489 name, $originator_name was incorrectly used when constructing a From:
1490 header. Furthermore, $originator_name was used for submission mode
1491 messages from external hosts without From: headers in a similar way,
1492 which is clearly wrong.
1494 PH/37 Added control=suppress_local_fixups.
1496 PH/38 When log_selector = +received_sender was set, and the addition of the
1497 sender made the log line's construction buffer exactly full, or one byte
1498 less than full, an overflow happened when the terminating "\n" was
1501 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1502 when the result of a list match is failure because a DNS lookup failed.
1504 PH/40 RM_COMMAND is now used in the building process.
1506 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1507 the "build-* directories that it finds.
1509 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1510 address in a domain literal was a prefix of an interface address.
1512 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1513 when verifying a sender address, unless rewrite_headers is false.
1515 PH/44 Wrote a long comment about why errors_to addresses are verified as
1516 recipients, not senders.
1518 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1519 the ratelimit ACL was added.
1521 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1523 PH/46 Added extra information about PostgreSQL errors to the error string.
1525 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1526 avoiding the need to install special zones in a real server. This is
1527 backwards compatible; if it can't find the fake resolver, it drops back.
1528 Thus, both old and new test suites can be run.
1530 TF/02 Added util/ratelimit.pl
1532 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1533 clock is set back in time.
1535 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1536 Candler <B.Candler@pobox.com>.
1538 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1539 OK for addresses that are the result of redirections.
1541 PH/48 A number of further additions for the benefit of the new test suite,
1542 including a fake gethostbyname() that interfaces to the fake DNS resolver
1545 TF/06 The fix for widen_domains has also been applied to qualify_single and
1546 search_parents which are the other dnslookup options that can cause
1549 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1552 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1554 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1555 service" if it encountered an MX record with an empty target hostname.
1556 The message is now "an MX or SRV record indicated no SMTP service".
1558 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1559 if the DNS of the sending site is misconfigured. This is quite a
1560 common situation. This change restores the behaviour of treating a
1561 helo verification defer as a failure.
1563 PH/51 If self=fail was set on a router, the bounce message did not include the
1564 actual error message.
1570 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1572 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1573 command itself ended in error, the underlying message about the transport
1574 filter timeout was being overwritten with the pipe command error. Now the
1575 underlying error message should be appended to the second error message.
1577 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1579 PH/02 Reduce the amount of output that "make" produces by default. Full output
1580 can still be requested.
1582 PH/03 The warning log line about a condition test deferring for a "warn" verb
1583 was being output only once per connection, rather than after each
1584 occurrence (because it was using the same function as for successful
1585 "warn" verbs). This seems wrong, so I have changed it.
1587 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1588 it should not have, which might have caused a crash in the right
1589 circumstances, but probably never did.
1591 PH/04 Installed a modified version of Tony Finch's patch to make submission
1592 mode fix the return path as well as the Sender: header line, and to
1593 add a /name= option so that you can make the user's friendly name appear
1596 TF/03 Added the control = fakedefer ACL modifier.
1598 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1599 Mark Lowes for thorough testing.
1601 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1603 TK/03 Merged latest SRS patch from Miles Wilton.
1605 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1606 with the definition in sysexits.h (which is #included earlier).
1607 Fortunately, Exim does not actually use EX_OK. The code used to try to
1608 preserve the sysexits.h value, by assumimg that macro definitions were
1609 scanned for macro replacements. I have been disabused of this notion,
1610 so now the code just undefines EX_OK before #including unistd.h.
1612 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1613 in the smtp transport. When a block could not be written in a single
1614 write() function, the timeout was being re-applied to each part-write.
1615 This seems wrong - if the receiver was accepting one byte at a time it
1616 would take for ever. The timeout is now adjusted when this happens. It
1617 doesn't have to be particularly precise.
1619 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1620 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1622 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1623 without a domain if the check to <postmaster@domain> fails.
1625 SC/01 Eximstats: added -xls and the ability to specify output files
1626 (patch written by Frank Heydlauf).
1628 SC/02 Eximstats: use FileHandles for outputing results.
1630 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1632 SC/04 Eximstats: fixed display of large numbers with -nvr option
1634 SC/05 Eximstats: fixed merging of reports with empty tables.
1636 SC/06 Eximstats: added the -include_original_destination flag
1638 SC/07 Eximstats: removed tabs and trailing whitespace.
1640 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1642 TK/06 MBOX spool code: Add real "From " MBOX separator line
1643 so the .eml file is really in mbox format (even though
1644 most programs do not really care). Patch from Alex Miller.
1646 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1647 The latter is generated from $received_to and is only set if the
1648 message has one envelope recipient. SA can use these headers,
1649 obviously out-of-the-box. Patch from Alex Miller.
1651 PH/08 The ${def test on a variable was returning false if the variable's
1652 value was "0", contrary to what the specification has always said!
1653 The result should be true unless the variable is empty.
1655 PH/09 The syntax error of a character other than { following "${if
1656 def:variable_name" (after optional whitespace) was not being diagnosed.
1657 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1658 accidental colon was present, for example, could give incorrect results.
1660 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1661 result is used (not including appendfile, where other changes are about
1664 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1665 This involved changing a lot of size variables from int to off_t. It
1666 should work with maildirs and everything.
1668 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1669 spamd dying while we are connected to it.
1671 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1674 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1675 function for BDB 4.3.
1677 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1679 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1680 the sources for PCRE have been re-arranged and more files are now
1683 PH/15 The code I had for printing potentially long long variables in PH/11
1684 above was not the best (it lost precision). The length of off_t variables
1685 is now inspected at build time, and an appropriate printing format (%ld
1686 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1687 to be "long long int" or "long int". This is needed for the internal
1688 formatting function string_vformat().
1690 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1691 the configuration file to be ":syslog", then the script "guesses" where
1692 the logs files are, rather than using the compiled in default. In our
1693 case the guess is not the same as the compiled default, so the script
1694 suddenly stopped working when I started to use syslog. The patch checks
1695 to see if log_file_path is "". If so, it attempts to read it from exim
1696 with no configuration file to get the compiled in version, before it
1697 falls back to the previous guessing code."
1699 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1700 implementing BATV in an Exim configuration. See NewStuff for the gory
1703 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1704 Makefile that are specific to HP-UX.
1706 PH/18 If the "use_postmaster" option was set for a recipient callout together
1707 with the "random" option, the postmaster address was used as the MAIL
1708 FROM address for the random test, but not for the subsequent recipient
1709 test. It is now used for both.
1711 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1712 patch removes a few documentation additions to RFC 3028, because the
1713 latest draft now contains them. It adds the new en;ascii-case comparator
1714 and a new error check for 8bit text in MIME parts. Comparator and
1715 require names are now matched exactly. I enabled the subaddress
1716 extension, but it is not well tested yet (read: it works for me)."
1718 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1719 rework some of the code of TK/09 above to avoid the hardwired use of
1720 "%lld" and "long long". Replaced the call to snprintf() with a call to
1723 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1724 records point to non-existent hosts", "retry timeout exceeded", and
1725 "retry time not reached for any host after a long failure period".
1727 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1728 experimental DomainKeys support:
1730 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1731 (2) On an error such as an illegally used "control", the wrong name for
1732 the control was given.
1734 These problems did NOT occur unless DomainKeys support was compiled.
1736 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1738 PH/24 Added ${if match_ip condition.
1740 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1741 databases so that it will be absolutely obvious if a crash occurs in the
1742 DB library. This is a regular occurrence (often caused by mis-matched
1745 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1746 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1747 picked up on a user's system that detects such things. There doesn't seem
1748 to be a gcc warning option for this - only an attribute that has to be
1749 put on the function's prototype. It seems that in Fedora Core 4 they have
1750 set this on a number of new functions. No doubt there will be more in due
1753 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1754 specify a transport. However, if an address that was verified by such a
1755 router was the subject of a callout, Exim crashed because it tried to
1756 read the rcpt_include_affixes from the non-existent transport. Now it
1757 just assumes that the setting of that option is false. This bug was
1758 introduced by 4.51/PH/31.
1760 PH/28 Changed -d+all to exclude +memory, because that information is very
1761 rarely of interest, but it makes the output a lot bigger. People tend to
1762 do -d+all out of habit.
1764 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1765 code in os-type was giving problems when libc.so lives in lib64, like on
1768 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1769 aren't the modern standard, and it seems that some systems' include files
1770 don't always have them. Exim was already checking for some of the newer
1771 ones like T_AAAA, and defining it itself. I've added checks for all the
1772 record types that Exim uses.
1774 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1775 not automatically generating a new one, as it is supposed to. This
1776 prevented TLS from working. If the file did exist, but contained invalid
1777 data, a new version was generated, as expected. It was only the case of a
1778 non-existent file that was broken.
1780 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1781 with a change in libdomainkeys > 0.64.
1783 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1784 from DNS. If the selector record carries the flag, it now has
1785 precedence over the domain-wide flag.
1787 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1789 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1790 the use of a port name that isn't defined in /etc/services) occurred, the
1791 message was deferred as in a normal delivery, and thus remained on the
1792 spool, instead of being failed because of the mua_wrapper setting. This
1793 is now fixed, and I tidied up some of the mua_wrapper messages at the
1796 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1797 the messages in a hash of arrays rather than using individual hashes.
1798 This is a bit cleaner and results in dramatic memory savings, albeit
1799 at a slight CPU cost.
1801 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1802 as requested by Marc Sherman.
1804 SC/10 Eximstats: added histograms for user specified patterns as requested
1807 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1809 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1810 fopen() in the content-scanning modules that did not already have it.
1816 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1817 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1819 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1821 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1823 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1824 to test Sieve filters that use "vacation".
1826 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1827 that changes the way the GnuTLS parameters are stored in the cache file.
1828 The new format can be generated externally. For backward compatibility,
1829 if the data in the cache doesn't make sense, Exim assumes it has read an
1830 old-format file, and it generates new data and writes a new file. This
1831 means that you can't go back to an older release without removing the
1834 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1835 work if there are any delivery delays because "one_time" forces the
1836 parent to be marked "delivered", so its unseen clone is never tried
1837 again. For this reason, Exim now forbids the simultaneous setting of
1840 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1841 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1842 entry if you want to know the details. The fix, however, neglected to
1843 consider the case when local delivery batching is involved. The test for
1844 "previously delivered" was not happening when checking to see if an
1845 address could be batched with a previous (undelivered) one; under
1846 certain circumstances this could lead to multiple deliveries to the same
1849 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1850 in its include files, and this causes problems building Exim.
1852 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1853 header_syntax) but Exim was just ignoring anything given after a slash.
1854 In particular, this caused confusion with an attempt to use "verify =
1855 reverse_host_lookup/defer_ok". An error is now given when options are
1856 supplied for verify items that do not have them. (Maybe reverse_host_
1857 lookup should have a defer_ok option, but that's a different point.)
1859 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1860 defined by RFC 821) to 2048, because there were problems with some AUTH
1861 commands, and RFC 1869 says the size should be increased for extended
1862 SMTP commands that take arguments.
1864 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1867 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1868 "unknown" error; now it says that the functionality isn't in the binary.
1870 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1871 an address' error message when a string expansion fails (syntax or
1872 whatever). Otherwise the password may appear in the log. Following change
1873 PH/42 below, there is no longer a chance of it appearing in a bounce
1876 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1878 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1879 of its addresses was ever tried. (Bugzilla bug #2.)
1881 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1882 the result incorrectly in the debug output. (It correctly added a newline
1883 to what was transported.)
1885 TF/01 Added $received_time.
1887 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1888 commented out examples of how to interface to a virus scanner and to
1889 SpamAssassin. Also added commented examples of av_scanner and
1890 spamd_address settings.
1892 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1893 and controls are allowed in which ACLs. There were a couple of minor
1894 errors. Some of the entries in the conditions table (which is a table of
1895 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1896 negation of where the condition IS allowed.
1898 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1900 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1901 header file does not have a version number, so I've had to invent a new
1902 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1903 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1904 radiusclient), but it was contributed by a Radius user.
1906 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1907 files or directories, for OpenSSL.
1909 PH/19 When an Exim process that is running as root has to create an Exim log
1910 file, it does so in a subprocess that runs as exim:exim so as to get the
1911 ownership right at creation (otherwise, other Exim processes might see
1912 the file with the wrong ownership). There was no test for failure of this
1913 fork() call, which would lead to the process getting stuck as it waited
1914 for a non-existent subprocess. Forks do occasionally fail when resources
1915 run out. I reviewed all the other calls to fork(); they all seem to check
1918 PH/20 When checking for unexpected SMTP input at connect time (before writing
1919 the banner), Exim was not dealing correctly with a non-positive return
1920 from the read() function. If the client had disconnected by this time,
1921 the result was a log entry for a synchronization error with an empty
1922 string after "input=" when read() returned zero. If read() returned -1
1923 (an event I could not check), uninitialized data bytes were printed.
1924 There were reports of junk text (parts of files, etc) appearing after
1927 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1929 PH/22 Added support for macro redefinition, and (re)definition in between
1930 driver and ACL definitions.
1932 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1933 forgetting to use the resulting value; it was using the unexpanded value.
1935 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1936 hadn't been configured. The fix is from Juergen Kreileder, who
1937 understands it better than I do:
1939 "Here's what I see happening with three configured cyrus_sasl
1940 authenticators configured (plain, login, cram-md5):
1942 On startup auth_cyrus_sasl_init() gets called for each of these.
1943 This means three calls to sasl_listmech() without a specified mech_list.
1944 => SASL tests which mechs of all available mechs actually work
1945 => three warnings about OTP not working
1946 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1948 With the patch, sasl_listmech() also gets called three times. But now
1949 SASL's mech_list option is set to the server_mech specified in the the
1950 authenticator. Or in other words, the answer from sasl_listmech()
1951 gets limited to just the mech you're testing for (which is different
1953 => the return list contains just 'plain' or 'login', 'cram-md5' or
1954 nothing depending on the value of ob->server_mech.
1956 I've just tested the patch: Authentication still works fine,
1957 unavailable mechs specified in the exim configuration are still
1958 caught, and the auth.log warnings about OTP are gone."
1960 PH/25 When debugging is enabled, the contents of the command line are added
1961 to the debugging output, even when log_selector=+arguments is not
1964 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1965 answer is "GNU", and only if the return is "GNU/something" is the answer
1968 PH/27 $acl_verify_message is now set immediately after the failure of a
1969 verification in an ACL, and so is available in subsequent modifiers. In
1970 particular, the message can be preserved by coding like this:
1972 warn !verify = sender
1973 set acl_m0 = $acl_verify_message
1975 Previously, $acl_verify_message was set only while expanding "message"
1976 and "log_message" when a very denied access.
1978 PH/28 Modified OS/os.c-Linux with
1980 -#ifndef OS_LOAD_AVERAGE
1981 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1983 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1984 nomenclature these days.)
1986 PH/29 Installed patch from the Sieve maintainer that adds the options
1987 sieve_useraddress and sieve_subaddress to the redirect router.
1989 PH/30 In these circumstances:
1990 . Two addresses routed to the same list of hosts;
1991 . First host does not offer TLS;
1992 . First host accepts first address;
1993 . First host gives temporary error to second address;
1994 . Second host offers TLS and a TLS session is established;
1995 . Second host accepts second address.
1996 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1997 and peerdn, if requested) that were in fact used only for the second
2000 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2001 attention to any local part prefix or suffix that was matched by the
2002 router that accepted the address. It now behaves in the same way as it
2003 does for delivery: the affixes are removed from the local part unless
2004 rcpt_include_affixes is set on the transport.
2006 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2007 timeout during the DATA phase of an incoming message.
2009 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2010 applied a patch sanctioned by the Sieve maintainer.
2012 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2013 the uid or gid is negative. A case of a negative gid caused this to be
2014 noticed. The fix allows for either to be negative.
2016 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2017 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2020 PH/36 The change PH/12 above was broken. Fixed it.
2022 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2023 the grounds that routing the same address twice would always produce the
2024 same answer. This might have been true once, but it is certainly no
2025 longer true now. Routing a child address may depend on the previous
2026 routing that produced that child. Some complicated redirection strategies
2027 went wrong when messages had multiple recipients, and made Exim's
2028 behaviour dependent on the order in which the addresses were given.
2030 I have moved the duplicate checking until after the routing is complete.
2031 Exim scans the addresses that are assigned to local and remote
2032 transports, and removes any duplicates. This means that more work will be
2033 done, as duplicates will always all be routed, but duplicates are
2034 presumably rare, so I don't expect this is of any significance.
2036 For deliveries to pipes, files, and autoreplies, the duplicate checking
2037 still happens during the routing process, since they are not going to be
2040 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2041 It corrects a timeout issue with spamd. This is Ian's comment: "The
2042 background is that sometimes spamd either never reads data from a
2043 connection it has accepted, or it never writes response data. The exiscan
2044 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2045 blindly assumes that writes won't block so it may never time out."
2047 PH/39 Allow G after quota size as well as K and M.
2049 PH/40 The value set for $authenticated_id in an authenticator may not contain
2050 binary zeroes or newlines because the value is written to log lines and
2051 to spool files. There was no check on this. Now the value is run through
2052 the string_printing() function so that such characters are converted to
2053 printable escape sequences.
2055 PH/41 $message_linecount is a new variable that contains the total number of
2056 lines in the message. Compare $body_linecount, which is the count for the
2059 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2060 bounce and delay warning messages, except in certain special cases, which
2063 (a) An SMTP error message from a remote host;
2064 (b) A message specified in a :fail: redirection;
2065 (c) A message specified in a "fail" command in a system filter;
2066 (d) A message specified in a FAIL return from the queryprogram router;
2067 (e) A message specified by the cannot_route_message router option.
2069 In these cases only, Exim does include the error details in bounce and
2070 warning messages. There are also a few cases where bland messages such
2071 as "unrouteable address" or "local delivery error" are given.
2073 PH/43 $value is now also set for the "else" part of a ${run expansion.
2075 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2076 being worked on, but at least Exim now implements the latest version to
2079 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2080 process to complete was treated as a delivery failure, a timeout while
2081 writing the message to the pipe was logged, but erroneously treated as a
2082 successful delivery. Such timeouts include transport filter timeouts. For
2083 consistency with the overall process timeout, these timeouts are now
2084 treated as errors, giving rise to delivery failures by default. However,
2085 there is now a new Boolean option for the pipe transport called
2086 timeout_defer, which, if set TRUE, converts the failures into defers for
2087 both kinds of timeout. A transport filter timeout is now identified in
2090 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2091 systems where "make" and "gmake" are different, calling "gmake" at top
2092 level broke things. I've arranged for the value of $(MAKE) to be passed
2093 from the Makefile to this script so that it can call the same version of
2097 A note about Exim versions 4.44 and 4.50
2098 ----------------------------------------
2100 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2101 changes of various kinds. As a consequence, a big documentation update was
2102 needed. This delayed the release for rather longer than seemed good, especially
2103 in the light of a couple of (minor) security issues. Therefore, the changes
2104 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2105 release. So 4.44 and 4.50 are in effect two different branches that both start
2108 I have left the 4.50 change log unchanged; it contains all the changes since
2109 4.43. The change log for 4.44 is below; many of its items are identical to
2110 those for 4.50. This seems to be the most sensible way to preserve the
2111 historical information.
2117 1. Minor wording change to the doc/README.SIEVE file.
2119 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2120 computation of the current number of files was incorrect.
2122 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2123 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2126 4. Give more explanation in the error message when the command for a transport
2127 filter fails to execute.
2129 5. There are several places where Exim runs a non-Exim command in a
2130 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2131 was being done only for the command run by the queryprogram router. It is
2132 now done for all such subprocesses. The other cases are: ${run, transport
2133 filters, and the commands run by the lmtp and pipe transports.
2135 6. Added CONFIGURE_GROUP build-time option.
2137 7. Some older OS have a limit of 256 on the maximum number of file
2138 descriptors. Exim was using setrlimit() to set 1000 as a large value
2139 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2140 systems. I've change it so that if it can't get 1000, it tries for 256.
2142 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2143 was an oversight, and furthermore, ever since the addition of extra
2144 controls (e.g. 4.43/32), the checks on when to allow different forms of
2145 "control" were broken. There should now be diagnostics for all cases when a
2146 control that does not make sense is encountered.
2148 9. Added the /retain_sender option to "control=submission".
2150 10. $recipients is now available in the predata ACL (oversight).
2152 11. Tidy the search cache before the fork to do a delivery from a message
2153 received from the command line. Otherwise the child will trigger a lookup
2154 failure and thereby defer the delivery if it tries to use (for example) a
2155 cached ldap connection that the parent has called unbind on.
2157 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2158 of $address_data from the recipient verification was clobbered by the
2159 sender verification.
2161 13. The value of address_data from a sender verification is now available in
2162 $sender_address_data in subsequent conditions in the ACL statement.
2164 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2166 15. Added a new option "connect=<time>" to callout options, to set a different
2169 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2170 was its contents. (It was OK if the option was not defined at all.)
2172 17. A "Completed" log line is now written for messages that are removed from
2173 the spool by the -Mrm option.
2175 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2176 information about exactly what failed.
2178 19. Added -dd to debug only the daemon process.
2180 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2181 handles timeouts, both on the server side and network timeouts. Renamed the
2182 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2184 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2185 It is now set to "smtps".
2187 22. $host_address is now set to the target address during the checking of
2188 ignore_target_hosts.
2190 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2191 being passed; this would have caused $sender_host_name to have been used if
2192 matching the list had actually called for a host name (not very likely,
2193 since this list is usually IP addresses). A host name is now passed as
2196 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2197 code that specifies a non-restarting handler (typically sigaction() in
2198 modern systems) in an attempt to fix a rare and obscure crash bug.
2200 25. Narrowed the window for a race in the daemon that could cause it to ignore
2201 SIGCHLD signals. This is not a major problem, because they are used only to
2202 wake it up if nothing else does.
2204 26. A malformed maildirsize file could cause Exim to calculate negative values
2205 for the mailbox size or file count. Odd effects could occur as a result.
2206 The maildirsize information is now recalculated if the size or filecount
2209 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2210 support for a long time. Removed HAVE_SYS_VFS_H.
2212 28. Installed the latest version of exipick from John Jetmore.
2214 29. In an address list, if the pattern was not a regular expression, an empty
2215 subject address (from a bounce message) matched only if the pattern was an
2216 empty string. Non-empty patterns were not even tested. This was the wrong
2217 because it is perfectly reasonable to use an empty address as part of a
2218 database query. An empty address is now tested by patterns that are
2219 lookups. However, all the other forms of pattern expect the subject to
2220 contain a local part and a domain, and therefore, for them, an empty
2221 address still always fails if the pattern is not itself empty.
2223 30. Exim went into a mad DNS loop when attempting to do a callout where the
2224 host was specified on an smtp transport, and looking it up yielded more
2225 than one IP address.
2227 31. Re-factored the code for checking spool and log partition space into a
2228 function that finds that data and another that does the check. The former
2229 is then used to implement four new variables: $spool_space, $log_space,
2230 $spool_inodes, and $log_inodes.
2232 32. The RFC2047 encoding function was originally intended for short strings
2233 such as real names; it was not keeping to the 75-character limit for
2234 encoded words that the RFC imposes. It now respects the limit, and
2235 generates multiple encoded words if necessary. To be on the safe side, I
2236 have increased the buffer size for the ${rfc2047: expansion operator from
2239 33. It is now permitted to omit both strings after an "if" condition; if the
2240 condition is true, the result is "true". As before, when the second string
2241 is omitted, a false condition yields an empty string. This makes it less
2242 cumbersome to write custom ACL and router conditions.
2244 34. Failure to deliver a bounce message always caused it to be frozen, even if
2245 there was an errors_to setting on the router. The errors_to setting is now
2248 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2249 canonical form (fully expanded) before being placed in
2250 $sender_host_address.
2252 36. The table in the code that translates DNS record types into text (T_A to
2253 "A" for instance) was missing entries for NS and CNAME. It is just possible
2254 that this could have caused confusion if both these types were looked up
2255 for the same domain, because the text type is used as part of Exim's
2256 per-process caching. But the chance of anyone hitting this buglet seems
2259 37. The dnsdb lookup has been extended in a number of ways.
2261 (1) There is a new type, "zns", which walks up the domain tree until it
2262 finds some nameserver records. It should be used with care.
2264 (2) There is a new type, "mxh", which is like "mx" except that it returns
2265 just the host names, not the priorities.
2267 (3) It is now possible to give a list of domains (or IP addresses) to be
2268 looked up. The behaviour when one of the lookups defers can be
2269 controlled by a keyword.
2271 (4) It is now possible to specify the separator character for use when
2272 multiple records are returned.
2274 38. The dnslists ACL condition has been extended: it is now possible to supply
2275 a list of IP addresses and/or domains to be looked up in a particular DNS
2278 39. Added log_selector=+queue_time_overall.
2280 40. When running the queue in the test harness, wait just a tad after forking a
2281 delivery process, to get repeatability of debugging output.
2283 41. Include certificate and key file names in error message when GnuTLS fails
2284 to set them up, because the GnuTLS error message doesn't include the name
2285 of the failing file when there is a problem reading it.
2287 42. Allow both -bf and -bF in the same test run.
2289 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2291 44. The "Exiscan patch" is now merged into the mainline Exim source.
2293 45. Sometimes the final signoff response after QUIT could fail to get
2294 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2295 < 0 before doing a fflush(). This bug looks as though it goes back to the
2296 introduction of TLS in release 3.20, but "sometimes" must have been rare
2297 because the tests only now provoked it.
2299 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2300 (this can affect the format of dates).
2302 47. exim_tidydb, when checking for the continued existence of a message for
2303 which it has found a message-specific retry record, was not finding
2304 messages that were in split spool directories. Consequently, it was
2305 deleting retry records that should have stayed in existence.
2307 48. Steve fixed some bugs in eximstats.
2309 49. The SPA authentication driver was not abandoning authentication and moving
2310 on to the next authenticator when an expansion was forced to fail,
2311 contradicting the general specification for all authenticators. Instead it
2312 was generating a temporary error. It now behaves as specified.
2314 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2315 (the order specifies the preference for clients). The order is now AES256,
2316 AES128, 3DES, ARCFOUR128.
2318 51. Small patch to Sieve code - explicitly set From: when generating an
2321 52. Exim crashed if a remote delivery caused a very long error message to be
2322 recorded - for instance if somebody sent an entire SpamAssassin report back
2323 as a large number of 550 error lines. This bug was coincidentally fixed by
2324 increasing the size of one of Exim's internal buffers (big_buffer) that
2325 happened as part of the Exiscan merge. However, to be on the safe side, I
2326 have made the code more robust (and fixed the comments that describe what
2329 53. Now that there can be additional text after "Completed" in log lines (if
2330 the queue_time_overall log selector is set), a one-byte patch to exigrep
2331 was needed to allow it to recognize "Completed" as not the last thing in
2334 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2335 patch that reportedly fixes this has been added. I am not expert enough to
2336 create a test for it. This is what the patch creator wrote:
2338 "I found a little strange behaviour of ldap code when working with
2339 Windows 2003 AD Domain, where users was placed in more than one
2340 Organization Units. When I tried to give exim partial DN, the exit code
2341 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2342 But simultaneously result of request was absolutely normal ldap result,
2343 so I produce this patch..."
2345 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2346 REFERENCE, so I have modified the code to exclude the patch when that macro
2349 55. Some experimental protocols are using DNS PTR records for new purposes. The
2350 keys for these records are domain names, not reversed IP addresses. The
2351 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2352 leaves it alone. Component reversal etc. now happens only for IP addresses.
2355 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2357 57. Double the size of the debug message buffer (to 2048) so that more of very
2358 long debug lines gets shown.
2360 58. The exicyclog utility now does better if the number of log files to keep
2361 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2363 59. Two changes related to the smtp_active_hostname option:
2365 (1) $smtp_active_hostname is now available as a variable.
2366 (2) The default for smtp_banner uses $smtp_active_hostname instead
2367 of $primary_hostname.
2369 60. The host_aton() function is supposed to be passed a string that is known
2370 to be a valid IP address. However, in the case of IPv6 addresses, it was
2371 not checking this. This is a hostage to fortune. Exim now panics and dies
2372 if the condition is not met. A case was found where this could be provoked
2373 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2374 components; fortuitously, this particular loophole had already been fixed
2375 by change 4.50/55 above.
2377 If there are any other similar loopholes, the new check in host_aton()
2378 itself should stop them being exploited. The report I received stated that
2379 data on the command line could provoke the exploit when Exim was running as
2380 exim, but did not say which command line option was involved. All I could
2381 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2382 running as the user.
2385 61. There was a buffer overflow vulnerability in the SPA authentication code
2386 (which came originally from the Samba project). I have added a test to the
2387 spa_base64_to_bits() function which I hope fixes it.
2390 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2391 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2393 63. The daemon start-up calls getloadavg() while still root for those OS that
2394 need the first call to be done as root, but it missed one case: when
2395 deliver_queue_load_max is set with deliver_drop_privilege. This is
2396 necessary for the benefit of the queue runner, because there is no re-exec
2397 when deliver_drop_privilege is set.
2399 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2400 This has been fixed.
2402 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2403 was in use, was not putting the data itself into the right store pool;
2404 consequently, it could be overwritten for a subsequent message in the same
2405 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2408 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2410 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2411 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2412 the function were treating the return as a boolean value, which happened to
2413 work because 0=false and not-0=true, but is not correct code.
2415 68. The host_aton() function was not handling scoped IPv6 addresses (those
2416 with, for example, "%eth0" on the end) correctly.
2418 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2419 negated items (that is, ~something) in unsigned ints. Some compilers
2420 apparently mutter when there is no cast.
2422 70. If an address verification called from an ACL failed, and did not produce a
2423 user-specific message (i.e. there was only a "system" message), nothing was
2424 put in $acl_verify_message. In this situation, it now puts the system
2427 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2428 session; change 4.31/43 added the unwanted input to the log line - except
2429 that it did not do this in the start of session case. It now does.
2431 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2432 This is wrong and can cause the other end to generate a synchronization
2433 error if it is another Exim or anything else that does the synchronization
2434 check. A QUIT command is no longer sent after a timeout.
2436 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2437 during host lookups.
2439 74. The defer_ok option of callout verification was not working if it was used
2440 when verifying addresses in header lines, that is, for this case:
2442 verify = header_sender/callout=defer_ok
2444 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2445 those file descriptors could be used for SMTP connections. If anything
2446 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2447 could be sent to the SMTP client, causing chaos. The daemon now opens
2448 stdin, stdout, and stderr to /dev/null when it puts itself into the
2451 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2452 log by default. The user can override this with suitable Perl magic.
2454 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2455 the log message when discard triggers, was not working for the DATA ACL or
2456 for the non-SMTP ACL.
2458 78. Error message wording change in sieve.c.
2460 79. If smtp_accept_max_per_host was set, the number of connections could be
2461 restricted to fewer than expected, because the daemon was trying to set up
2462 a new connection before checking whether the processes handling previous
2463 connections had finished. The check for completed processes is now done
2464 earlier. On busy systems, this bug wouldn't be noticed because something
2465 else would have woken the daemon, and it would have reaped the completed
2468 80. If a message was submitted locally by a user whose login name contained one
2469 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2470 It caused a spool format error. I have fixed the spool reading code. A
2471 related problem was that the "from" clause in the Received: line became
2472 illegal because of the space(s). It is now covered by ${quote_local_part.
2474 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2477 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2479 83. Patch from Sieve maintainer to fix unterminated string problem in
2480 "vacation" handling.
2482 84. Some minor changes to the Linux configuration files to help with other
2483 OS variants using glibc.
2485 85. One more patch for Sieve to update vacation handling to latest spec.
2488 ----------------------------------------------------
2489 See the note above about the 4.44 and 4.50 releases.
2490 ----------------------------------------------------
2496 1. Change 4.43/35 introduced a bug that caused file counts to be
2497 incorrectly computed when quota_filecount was set in an appendfile
2500 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2501 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2504 3. Give more explanation in the error message when the command for a transport
2505 filter fails to execute.
2507 4. There are several places where Exim runs a non-Exim command in a
2508 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2509 was being done only for the command run by the queryprogram router. It is
2510 now done for all such subprocesses. The other cases are: ${run, transport
2511 filters, and the commands run by the lmtp and pipe transports.
2513 5. Some older OS have a limit of 256 on the maximum number of file
2514 descriptors. Exim was using setrlimit() to set 1000 as a large value
2515 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2516 systems. I've change it so that if it can't get 1000, it tries for 256.
2518 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2519 was an oversight, and furthermore, ever since the addition of extra
2520 controls (e.g. 4.43/32), the checks on when to allow different forms of
2521 "control" were broken. There should now be diagnostics for all cases when a
2522 control that does not make sense is encountered.
2524 7. $recipients is now available in the predata ACL (oversight).
2526 8. Tidy the search cache before the fork to do a delivery from a message
2527 received from the command line. Otherwise the child will trigger a lookup
2528 failure and thereby defer the delivery if it tries to use (for example) a
2529 cached ldap connection that the parent has called unbind on.
2531 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2532 of $address_data from the recipient verification was clobbered by the
2533 sender verification.
2535 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2536 was its contents. (It was OK if the option was not defined at all.)
2538 11. A "Completed" log line is now written for messages that are removed from
2539 the spool by the -Mrm option.
2541 12. $host_address is now set to the target address during the checking of
2542 ignore_target_hosts.
2544 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2545 being passed; this would have caused $sender_host_name to have been used if
2546 matching the list had actually called for a host name (not very likely,
2547 since this list is usually IP addresses). A host name is now passed as
2550 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2551 code that specifies a non-restarting handler (typically sigaction() in
2552 modern systems) in an attempt to fix a rare and obscure crash bug.
2554 15. Narrowed the window for a race in the daemon that could cause it to ignore
2555 SIGCHLD signals. This is not a major problem, because they are used only to
2556 wake it up if nothing else does.
2558 16. A malformed maildirsize file could cause Exim to calculate negative values
2559 for the mailbox size or file count. Odd effects could occur as a result.
2560 The maildirsize information is now recalculated if the size or filecount
2563 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2564 support for a long time. Removed HAVE_SYS_VFS_H.
2566 18. Updated exipick to current release from John Jetmore.
2568 19. Allow an empty sender to be matched against a lookup in an address list.
2569 Previously the only cases considered were a regular expression, or an
2572 20. Exim went into a mad DNS lookup loop when doing a callout where the
2573 host was specified on the transport, if the DNS lookup yielded more than
2576 21. The RFC2047 encoding function was originally intended for short strings
2577 such as real names; it was not keeping to the 75-character limit for
2578 encoded words that the RFC imposes. It now respects the limit, and
2579 generates multiple encoded words if necessary. To be on the safe side, I
2580 have increased the buffer size for the ${rfc2047: expansion operator from
2583 22. Failure to deliver a bounce message always caused it to be frozen, even if
2584 there was an errors_to setting on the router. The errors_to setting is now
2587 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2588 canonical form (fully expanded) before being placed in
2589 $sender_host_address.
2591 24. Updated eximstats to version 1.33
2593 25. Include certificate and key file names in error message when GnuTLS fails
2594 to set them up, because the GnuTLS error message doesn't include the name
2595 of the failing file when there is a problem reading it.
2597 26. Expand error message when OpenSSL has problems setting up cert/key files.
2600 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2601 (this can affect the format of dates).
2603 28. exim_tidydb, when checking for the continued existence of a message for
2604 which it has found a message-specific retry record, was not finding
2605 messages that were in split spool directories. Consequently, it was
2606 deleting retry records that should have stayed in existence.
2608 29. eximstats updated to version 1.35
2609 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2610 1.35 - bugfix such that pie charts by volume are generated correctly
2612 30. The SPA authentication driver was not abandoning authentication and moving
2613 on to the next authenticator when an expansion was forced to fail,
2614 contradicting the general specification for all authenticators. Instead it
2615 was generating a temporary error. It now behaves as specified.
2617 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2618 (the order specifies the preference for clients). The order is now AES256,
2619 AES128, 3DES, ARCFOUR128.
2621 31. Small patch to Sieve code - explicitly set From: when generating an
2624 32. Exim crashed if a remote delivery caused a very long error message to be
2625 recorded - for instance if somebody sent an entire SpamAssassin report back
2626 as a large number of 550 error lines. This bug was coincidentally fixed by
2627 increasing the size of one of Exim's internal buffers (big_buffer) that
2628 happened as part of the Exiscan merge. However, to be on the safe side, I
2629 have made the code more robust (and fixed the comments that describe what
2632 33. Some experimental protocols are using DNS PTR records for new purposes. The
2633 keys for these records are domain names, not reversed IP addresses. The
2634 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2635 leaves it alone. Component reversal etc. now happens only for IP addresses.
2638 34. The host_aton() function is supposed to be passed a string that is known
2639 to be a valid IP address. However, in the case of IPv6 addresses, it was
2640 not checking this. This is a hostage to fortune. Exim now panics and dies
2641 if the condition is not met. A case was found where this could be provoked
2642 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2643 components; fortuitously, this particular loophole had already been fixed
2644 by change 4.50/55 or 4.44/33 above.
2646 If there are any other similar loopholes, the new check in host_aton()
2647 itself should stop them being exploited. The report I received stated that
2648 data on the command line could provoke the exploit when Exim was running as
2649 exim, but did not say which command line option was involved. All I could
2650 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2651 running as the user.
2654 35. There was a buffer overflow vulnerability in the SPA authentication code
2655 (which came originally from the Samba project). I have added a test to the
2656 spa_base64_to_bits() function which I hope fixes it.
2659 36. The daemon start-up calls getloadavg() while still root for those OS that
2660 need the first call to be done as root, but it missed one case: when
2661 deliver_queue_load_max is set with deliver_drop_privilege. This is
2662 necessary for the benefit of the queue runner, because there is no re-exec
2663 when deliver_drop_privilege is set.
2665 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2666 was in use, was not putting the data itself into the right store pool;
2667 consequently, it could be overwritten for a subsequent message in the same
2668 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2671 38. Sometimes the final signoff response after QUIT could fail to get
2672 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2673 < 0 before doing a fflush(). This bug looks as though it goes back to the
2674 introduction of TLS in release 3.20, but "sometimes" must have been rare
2675 because the tests only now provoked it.
2681 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2682 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2683 Now it returns an integer. A number of calls were still expecting a T/F
2684 return. Fortuitously, in all cases, the tests worked in OK situations,
2685 which is the norm. However, things would have gone wrong on any write
2686 failures on the smtp file descriptor. This function is used when sending
2687 messages over SMTP and also when doing verify callouts.
2689 2. When Exim is called to do synchronous delivery of a locally submitted
2690 message (the -odf or -odi options), it no longer closes stderr before doing
2693 3. Implemented the mua_wrapper option.
2695 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2697 5. Implemented the functions header_remove(), header_testname(),
2698 header_add_at_position(), and receive_remove_recipient(), and exported them
2701 6. If an ACL "warn" statement specified the addition of headers, Exim already
2702 inserted X-ACL-Warn: at the start if there was no header name. However, it
2703 was not making this test for the second and subsequent header lines if
2704 there were newlines in the string. This meant that an invalid header could
2705 be inserted if Exim was badly configured.
2707 7. Allow an ACL "warn" statement to add header lines at the start or after all
2708 the Received: headers, as well as at the end.
2710 8. Added the rcpt_4xx retry error code.
2712 9. Added postmaster_mailfrom=xxx to callout verification option.
2714 10. Added mailfrom=xxxx to the callout verification option, for verify=
2717 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2718 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2720 12. Inserted some casts to stop certain compilers warning when using pointer
2721 differences as field lengths or precisions in printf-type calls (mostly
2722 affecting debugging statements).
2724 13. Added optional readline() support for -be (dynamically loaded).
2726 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2727 same clock tick as a message's arrival, so that its received time was the
2728 same as the "first fail" time on the retry record, and that message
2729 remained on the queue past the ultimate address timeout, every queue runner
2730 would try a delivery (because it was past the ultimate address timeout) but
2731 after another failure, the ultimate address timeout, which should have then
2732 bounced the address, did not kick in. This was a "< instead of <=" error;
2733 in most cases the first failure would have been in the next clock tick
2734 after the received time, and all would be well.
2736 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2737 being recognized when the domain list was tested by the match_domain
2738 condition in an expansion string.
2740 16. Added the ${str2b64: operator.
2742 17. Exim was always calling setrlimit() to set a large limit for the number of
2743 processes, without checking whether the existing limit was already
2744 adequate. (It did check for the limit on file descriptors.) Furthermore,
2745 errors from getrlimit() and setrlimit() were being ignored. Now they are
2746 logged to the main and panic logs, but Exim does carry on, to try to do its
2747 job under whatever limits there are.
2749 18. Imported PCRE 5.0.
2751 19. Trivial typo in log message " temporarily refused connection" (the leading
2754 20. If the log selector return_path_on_delivery was set and an address was
2755 redirected to /dev/null, the delivery process crashed because it assumed
2756 that a return path would always be set for a "successful" delivery. In this
2757 case, the whole delivery is bypassed as an optimization, and therefore no
2760 21. Internal re-arrangement: the function for sending a challenge and reading
2761 a response while authentication was assuming a zero-terminated challenge
2762 string. It's now changed to take a pointer and a length, to allow for
2763 binary data in such strings.
2765 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2767 23. Exim was not respecting finduser_retries when seeking the login of the
2768 uid under which it was called; it was always trying 10 times. (The default
2769 setting of finduser_retries is zero.) Also, it was sleeping after the final
2770 failure, which is pointless.
2772 24. Implemented tls_on_connect_ports.
2774 25. Implemented acl_smtp_predata.
2776 26. If the domain in control=submission is set empty, Exim assumes that the
2777 authenticated id is a complete email address when it generates From: or
2778 Sender: header lines.
2780 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2781 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2782 chown and chgrp in /bin and hostname in /usr/bin.
2784 28. Exim was keeping the "process log" file open after each use, just as it
2785 does for the main log. This opens the possibility of it remaining open for
2786 long periods when the USR1 signal hits a daemon. Occasional processlog
2787 errors were reported, that could have been caused by this. Anyway, it seems
2788 much more sensible not to leave this file open at all, so that is what now
2791 29. The long-running daemon process does not normally write to the log once it
2792 has entered its main loop, and it closes the log before doing so. This is
2793 so that log files can straightforwardly be renamed and moved. However,
2794 there are a couple of unusual error situations where the daemon does write
2795 log entries, and I had neglected to close the log afterwards.
2797 30. The text of an SMTP error response that was received during a remote
2798 delivery was being truncated at 512 bytes. This is too short for some of
2799 the long messages that one sometimes sees. I've increased the limit to
2802 31. It is now possible to make retry rules that apply only when a message has a
2803 specific sender, in particular, an empty sender.
2805 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2806 it possible to be selective about when SMTP synchronization is enforced.
2808 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2810 32. Implemented hosts_connection_nolog.
2812 33. Added an ACL for QUIT.
2814 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2817 35. Added mailbox_size and mailbox_filecount to appendfile.
2819 36. Added control = no_multiline_responses to ACLs.
2821 37. There was a bug in the logic of the code that waits for the clock to tick
2822 in the case where the clock went backwards by a substantial amount such
2823 that the microsecond fraction of "now" was more than the microsecond
2824 fraction of "then" (but the whole seconds number was less).
2826 38. Added support for the libradius Radius client library this is found on
2827 FreeBSD (previously only the radiusclient library was supported).
2833 1. When certain lookups returned multiple values in the form name=value, the
2834 quoting of the values was not always being done properly. Specifically:
2835 (a) If the value started with a double quote, but contained no whitespace,
2837 (b) If the value contained whitespace other than a space character (i.e.
2838 tabs or newlines or carriage returns) it was not quoted.
2839 This fix has been applied to the mysql and pgsql lookups by writing a
2840 separate quoting function and calling it from the lookup code. The fix
2841 should probably also be applied to nisplus, ibase and oracle lookups, but
2842 since I cannot test any of those, I have not disturbed their existing code.
2844 2. A hit in the callout cache for a specific address caused a log line with no
2845 reason for rejecting RCPT. Now it says "Previous (cached) callout
2846 verification failure".
2848 3. There was an off-by-one bug in the queryprogram router. An over-long
2849 return line was truncated at 256 instead of 255 characters, thereby
2850 overflowing its buffer with the terminating zero. As well as fixing this, I
2851 have increased the buffer size to 1024 (and made a note to document this).
2853 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2854 when Exim is waiting for an SMTP response from a remote server, Exim
2855 restarts its select() call on the socket, thereby resetting its timeout.
2856 This is not a problem when such interrupts are rare. Somebody set up a cron
2857 job to run exiwhat every 2 minutes, which is less than the normal select()
2858 timeout (5 or 10 minutes). This meant that the select() timeout never
2859 kicked in because it was always reset. I have fixed this by comparing the
2860 time when an interrupt arrives with the time at the start of the first call
2861 to select(). If more time than the timeout has elapsed, the interrupt is
2862 treated as a timeout.
2864 5. Some internal re-factoring in preparation for the addition of Sieve
2865 extensions (by MH). In particular, the "personal" test is moved to a
2866 separate function, and given an option for scanning Cc: and Bcc: (which is
2867 not set for Exim filters).
2869 6. When Exim created an email address using the login of the caller as the
2870 local part (e.g. when creating a From: or Sender: header line), it was not
2871 quoting the local part when it contained special characters such as @.
2873 7. Installed new OpenBSD configuration files.
2875 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2876 try to make them clearer.
2878 9. Callout options, other than the timeout value, were being ignored when
2879 verifying sender addresses in header lines. For example, when using
2881 verify = header_sender/callout=no_cache
2883 the cache was (incorrectly) being used.
2885 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2886 only the Cygwin environment.
2888 11. When return_path_on_delivery was set as a log selector, if different remote
2889 addresses in the same message used different return paths and parallel
2890 remote delivery occurred, the wrong values would sometimes be logged.
2891 (Whenever a remote delivery process finished, the return path value from
2892 the most recently started remote delivery process was logged.)
2894 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2895 lines when AUTH and/or TLS are in use. This is the "received protocol"
2896 field. Exim used to use "asmtp" for authenticated SMTP, without any
2897 indication (in the protocol name) for TLS use. Now it follows the RFC and
2898 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2899 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2900 names appear in log lines as well as in Received: header lines.
2902 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2903 extensions, and comparison tests, and to fix some bugs.
2905 14. Changes to the "personal" filter test:
2907 (1) The test was buggy in that it was just doing the equivalent of
2908 "contains" tests on header lines. For example, if a user's address was
2909 anne@some.where, the "personal" test would incorrectly be true for
2911 To: susanne@some.where
2913 This test is now done by extracting each address from the header in turn,
2914 and checking the entire address. Other tests that are part of "personal"
2915 are now done using regular expressions (for example, to check local parts
2916 of addresses in From: header lines).
2918 (2) The list of non-personal local parts in From: addresses has been
2919 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2920 taken from the Sieve specification recommendations.
2922 (3) If the message contains any header line starting with "List-" it is
2923 treated as non-personal.
2925 (4) The test for "circular" in the Subject: header line has been removed
2926 because it now seems ill-conceived.
2928 15. Minor typos in src/EDITME comments corrected.
2930 16. Installed latest exipick from John Jetmore.
2932 17. If headers_add on a router specified a text string that was too long for
2933 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2934 of string_sprintf() is now avoided.
2936 18. $message_body_size was not set (it was always zero) when running the DATA
2937 ACL and the local_scan() function.
2939 19. For the "mail" command in an Exim filter, no default was being set for
2940 the once_repeat time, causing a random time value to be used if "once" was
2941 specified. (If the value happened to be <= 0, no repeat happened.) The
2942 default is now 0s, meaning "never repeat". The "vacation" command was OK
2943 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2944 (I found it when inspecting the code).
2946 20. There is now an overall timeout for performing a callout verification. It
2947 defaults to 4 times the callout timeout, which applies to individual SMTP
2948 commands during the callout. The overall timeout applies when there is more
2949 than one host that can be tried. The timeout is checked before trying the
2950 next host. This prevents very long delays if there are a large number of
2951 hosts and all are timing out (e.g. when the network connections are timing
2952 out). The value of the overall timeout can be changed by specifying an
2953 additional sub-option for "callout", called "maxwait". For example:
2955 verify = sender/callout=5s,maxwait=20s
2957 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2958 to the end before writing, but this should make it even safer).
2960 22. Exim was forgetting that it had advertised PIPELINING for the second and
2961 subsequent messages on an SMTP connection. It was also not resetting its
2962 memory on STARTTLS and an internal HELO.
2964 23. When Exim logs an SMTP synchronization error within a session, it now
2965 records whether PIPELINING has been advertised or not.
2967 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2968 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2969 rather than long int.
2971 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2973 26. Added the never_mail option to autoreply.
2979 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2980 crash if the getsockname() call failed; this can happen if a connection is
2981 closed very soon after it is established. The problem was simply in the
2982 order in which certain operations were done, causing Exim to try to write
2983 to the SMTP stream before it had set up the file descriptor. The bug has
2984 been fixed by making things happen in the correct order.
2990 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2991 before the connection was closed, thus losing the rejection response.
2993 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2994 some early Solaris releases, but causes trouble in current releases where
2995 socklen_t is defined.
2997 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3000 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3003 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3004 that releasing the top of it at the end releases what was used for sub-
3005 expansions (unless the block got too big). However, discard this block if
3006 the first thing is a variable or header, so that we can use its block when
3007 it is dynamic (useful for very large $message_headers, for example).
3009 6. Lookups now cache *every* query, not just the most recent. A new, separate
3010 store pool is used for this. It can be recovered when all lookup caches are
3011 flushed. Lookups now release memory at the end of their result strings.
3012 This has involved some general refactoring of the lookup sources.
3014 7. Some code has been added to the store_xxx() functions to reduce the amount
3015 of flapping under certain conditions.
3017 8. log_incoming_interface used to affect only the <= reception log lines. Now
3018 it causes the local interface and port to be added to several more SMTP log
3019 lines, for example "SMTP connection from", and rejection lines.
3021 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3023 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3025 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3026 could be overwritten at the end of the current message (or the start of a
3027 new message if it was set in a HELO ACL). The value is now preserved for
3028 the duration of the SMTP connection.
3030 12. If a transport had a headers_rewrite setting, and a matching header line
3031 contained an unqualified address, that address was qualified, even if it
3032 did not match any rewriting rules. The underlying bug was that the values
3033 of the flags that permit the existence of unqualified sender and recipient
3034 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3035 non-local messages, and by -bnq for local messages) were not being
3036 preserved with the message after it was received.
3038 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3039 "next input=" as part of the text comprising the host identity instead of
3040 the correct text. The code was using the same buffer for two different
3041 strings. However, depending on which order the printing function evaluated
3042 its arguments, the bug did not always show up. Under Linux, for example, my
3043 test suite worked just fine.
3045 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3046 doesn't work with some older versions of Perl. It has been changed to "my",
3047 which in any case is probably the better facility to use.
3049 15. A really picky compiler found some instances of statements for creating
3050 error messages that either had too many or two few arguments for the format
3053 16. The size of the buffer for calls to the DNS resolver has been increased
3054 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3055 for addresses that have a lot of PTR records. This alleviates a problem; it
3056 does not fully solve it.
3058 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3059 buffer now truncates the list and logs the incident, which is the same
3060 action as happens when Exim is looking up a host name and its aliases.
3061 Previously in this situation something unpredictable would happen;
3062 sometimes it was "internal error: store_reset failed".
3064 18. If a server dropped the connection unexpectedly when an Exim client was
3065 using GnuTLS and trying to read a response, the client delivery process
3066 crashed while trying to generate an error log message.
3068 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3069 string, for example:
3071 warn message = H1: something\nH2: something
3073 the text was added as a single header line from Exim's point of view
3074 though it ended up OK in the delivered message. However, searching for the
3075 second and subsequent header lines using $h_h2: did not work. This has been
3076 fixed. Similarly, if a system filter added multiple headers in this way,
3077 the routers could not see them.
3079 20. Expanded the error message when iplsearch is called with an invalid key to
3080 suggest using net-iplsearch in a host list.
3082 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3083 ACLs are no longer actually imposed (and a message to that effect is
3086 22. If a "gecos" field in a passwd entry contained escaped characters, in
3087 particular, if it contained a \" sequence, Exim got it wrong when building
3088 a From: or a Sender: header from that name. A second bug also caused
3089 incorrect handling when an unquoted " was present following a character
3090 that needed quoting.
3092 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3093 was not being matched caselessly.
3095 24. Arranged for all hyphens in the exim.8 source to be escaped with
3098 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3099 properly. Recipient callout cache records were still being keyed to include
3100 the sender, even when use_sender was set false. This led to far more
3101 callouts that were necessary. The sender is no longer included in the key
3102 when use_sender is false.
3104 26. Added "control = submission" modifier to ACLs.
3106 27. Added the ${base62d: operator to decode base 62 numbers.
3108 28. dnsdb lookups can now access SRV records.
3110 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3111 the configuration file.
3113 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3114 (-v) mode. This makes the output for a verbose queue run more intelligible.
3116 31. Added a use_postmaster feature to recipient callouts.
3118 32. Added the $body_zerocount variable, containing the number of binary zero
3119 bytes in the message body.
3121 33. The time of last modification of the "new" subdirectory is now used as the
3122 "mailbox time last read" when there is a quota error for a maildir
3125 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3127 35. Added +ignore_unknown as a special item in host lists.
3129 36. Code for decoding IPv6 addresses in host lists is now included, even if
3130 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3131 address was recognized as an IP address, but was then not correctly decoded
3132 into binary, causing unexpected and incorrect effects when compared with
3139 1. Very minor rewording of debugging text in manualroute to say "list of
3140 hosts" instead of "hostlist".
3142 2. If verify=header_syntax was set, and a header line with an unqualified
3143 address (no domain) and a large number of spaces between the end of the
3144 name and the colon was received, the reception process suffered a buffer
3145 overflow, and (when I tested it) crashed. This was caused by some obsolete
3146 code that should have been removed. The fix is to remove it!
3148 3. When running in the test harness, delay a bit after writing a bounce
3149 message to get a bit more predictability in the log output.
3151 4. Added a call to search_tidyup() just before forking a reception process. In
3152 theory, someone could use a lookup in the expansion of smtp_accept_max_
3153 per_host which, without the tidyup, could leave open a database connection.
3155 5. Added the variables $recipient_data and $sender_data which get set from a
3156 lookup success in an ACL "recipients" or "senders" condition, or a router
3157 "senders" option, similar to $domain_data and $local_part_data.
3159 6. Moved the writing of debug_print from before to after the "senders" test
3162 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3163 problems for message scanning, either using a data ACL, or using
3164 local_scan() because the Received: header was not generated till after they
3165 were called (in order to set the time as the time of reception completion).
3166 I have revised the way this works. The header is now generated after the
3167 body is received, but before the ACL or local_scan() are called. After they
3168 are run, the timestamp in the header is updated.
3174 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3175 before starting a queue runner without re-exec. This happened only when
3176 deliver_drop_privilege was set or when the Exim user was set to root. The
3177 effect of the bug was that timeouts during subsequent deliveries caused
3178 crashes instead of being properly handled. The handler is now left at its
3179 default (and expected) setting.
3181 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3182 message, again when deliver_drop_privilege is set or Exim is run as root.
3183 The bug described in (1) was not present in this case, but the tidying up
3184 of the other signals was missing. I have made the two cases consistent.
3186 3. The ignore_target_hosts setting on a manualroute router was being ignored
3187 for hosts that were looked up using the /MX notation.
3189 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3192 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3193 operated on the sender address. After changing the $sender_address to <>
3194 for the sender address verify, Exim was re-instated it as the original
3195 (before rewriting) address, but remembering that it had rewritten it, so it
3196 wasn't rewriting it again. This bug also had the effect of breaking the
3197 sender address verification caching when the sender address was rewritten.
3199 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3200 This has been changed so that if the ip literal address matches
3201 ignore_target_hosts, the router declines.
3203 7. Added expansion conditions match_domain, match_address, and match_local_
3204 part (NOT match_host).
3206 8. The placeholder for the Received: header didn't have a length field set.
3208 9. Added code to Exim itself and to exim_lock to test for a specific race
3209 condition that could lead to file corruption when using MBX delivery. The
3210 issue is with the lockfile that is created in /tmp. If this file is removed
3211 after a process has opened it but before that process has acquired a lock,
3212 there is the potential for a second process to recreate the file and also
3213 acquire a lock. This could lead to two Exim processes writing to the file
3214 at the same time. The added code performs the same test as UW imapd; it
3215 checks after acquiring the lock that its file descriptor still refers to
3216 the same named file.
3218 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3219 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3222 11. Added the smtp_active_hostname option. If used, this will typically be made
3223 to depend on the incoming interface address. Because $interface_address is
3224 not set up until the daemon has forked a reception process, error responses
3225 that can happen earlier (such as "too many connections") no longer contain
3228 12. If an expansion in a condition on a "warn" statement fails because a lookup
3229 defers, the "warn" statement is abandoned, and the next ACL statement is
3230 processed. Previously this caused the whole ACL to be aborted.
3232 13. Added the iplsearch lookup type.
3234 14. Added ident_timeout as a log selector.
3236 15. Added tls_certificate_verified as a log selector.
3238 16. Added a global option tls_require_ciphers (compare the smtp transport
3239 option of the same name). This controls incoming TLS connections.
3241 17. I finally figured out how to make tls_require_ciphers do a similar thing
3242 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3243 before starting the TLS session.
3245 18. Tabs are now shown as \t in -bP output.
3247 19. If the log selector return_path_on_delivery was set, Exim crashed when
3248 bouncing a message because it had too many Received: header lines.
3250 20. If two routers both had headers_remove settings, and the first one included
3251 a superfluous trailing colon, the final name in the first list and the
3252 first name in the second list were incorrectly joined into one item (with a
3253 colon in the middle).
3259 1. Added -C and -D options to the exinext utility, mainly to make it easier
3260 to include in the automated testing, but these could be helpful when
3261 multiple configurations are in use.
3263 2. The exinext utility was not formatting the output nicely when there was
3264 an alternate port involved in the retry record key, nor when there was a
3265 message id as well (for retries that were specific to a specific message
3266 and a specific host). It was also confused by IPv6 addresses, because of
3267 the additional colons they contain. I have fixed the IPv4 problem, and
3268 patched it up to do a reasonable job for IPv6.
3270 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3271 delivery, the log line now contains "pipelined" if PIPELINING was used.
3273 4. An SMTP transport process used to panic and die if the bind() call to set
3274 an explicit outgoing interface failed. This has been changed; it is now
3275 treated in the same way as a connect() failure.
3277 5. A reference to $sender_host_name in the part of a conditional expansion
3278 that was being skipped was still causing a DNS lookup. This no longer
3281 6. The def: expansion condition was not recognizing references to header lines
3282 that used bh_ and bheader_.
3284 7. Added the _cache feature to named lists.
3286 8. The code for checking quota_filecount in the appendfile transport was
3287 allowing one more file than it should have been.
3289 9. For compatibility with Sendmail, the command line option
3297 and sets the incoming protocol and host name (for trusted callers). The
3298 host name and its colon can be omitted when only the protocol is to be set.
3299 Note the Exim already has two private options, -pd and -ps, that refer to
3300 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3301 "s", but I don't think that's a major issue.
3303 10. A number of refactoring changes to the code, none of which should affect
3306 (a) The number of logging options was getting close to filling up the
3307 32-bit word that was used as a bit map. I have split them into two classes:
3308 those that are passed in the argument to log_write(), and those that are
3309 only ever tested independently outside of that function. These are now in
3310 separate 32-bit words, so there is plenty of room for expansion again.
3311 There is no change in the user interface or the logging behaviour.
3313 (b) When building, for example, log lines, the code previously used a
3314 macro that called string_cat() twice, in order to add two strings. This is
3315 not really sufficiently general. Furthermore, there was one instance where
3316 it was actually wrong because one of the argument was used twice, and in
3317 one call a function was used. (As it happened, calling the function twice
3318 did not affect the overall behaviour.) The macro has been replaced by a
3319 function that can join an arbitrary number of extra strings onto a growing
3322 (c) The code for expansion conditions now uses a table and a binary chop
3323 instead of a serial search (which was left over from when there were very
3324 few conditions). Also, it now recognizes conditions like "pam" even when
3325 the relevant support is not compiled in: a suitably worded error message is
3326 given if an attempt is made to use such a condition.
3328 11. Added ${time_interval:xxxxx}.
3330 12. A bug was causing one of the ddress fields not to be passed back correctly
3331 from remote delivery subprocesses. The field in question was not being
3332 subsequently used, so this caused to problems in practice.
3334 13. Added new log selectors queue_time and deliver_time.
3336 14. Might have fixed a bug in maildirsizefile handling that threw up
3337 "unexpected character" debug warnings, and recalculated the data
3338 unnecessarily. In any case, I expanded the warning message to give more
3341 15. Added the message "Restricted characters in address" to the statements in
3342 the default ACL that block characters like @ and % in local parts.
3344 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3345 Three changes have been made:
3347 (a) There was a serious bug; a negative response to MAIL caused the whole
3348 recipient domain to be cached as invalid, thereby blocking all messages
3349 to all local parts at the same domain, from all senders. This bug has
3350 been fixed. The domain is no longer cached after a negative response to
3351 MAIL if the sender used is not empty.
3353 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3356 (c) A new callout option, "use_sender" has been added for people who want
3357 the modified behaviour.
3363 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3366 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3367 indeed breaks things for older releases.
3369 3. Added additional logging to the case where there is a problem reading data
3370 from a filter that is running in a subprocess using a pipe, in order to
3371 try to track down a specific problem.
3373 4. Testing facility fudge: when running in the test harness and attempting
3374 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3375 getting "No route to host". Convert this to a timeout.
3377 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3380 6. Some OS don't have socklen_t but use size_t instead. This affects the
3381 fifth argument of getsockopt() amongst other things. This is now
3382 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3383 can be set for individual OS. I have set it for SunOS5, OSF1, and
3384 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3385 some earlier ones do not.
3387 7. Change 4.30/15 was not doing the test caselessly.
3389 8. The standard form for an IPv6 address literal was being rejected by address
3390 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3391 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3392 this, as well as the form without the "IPv6" on the front (but only when
3393 address literals are enabled, of course).
3395 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3397 10. Exim crashed if a message with an empty sender address specified by -f
3398 encountered a router with an errors_to setting. This could be provoked only
3399 by a command such as
3403 where an empty string was supplied; "<>" did not hit this bug.
3405 11. Installed PCRE release 4.5.
3407 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3408 remained set. It is now erased.
3410 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3411 times from message ids (which are base 36 rather than the normal 62).
3413 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3414 were being counted as actual protocol errors, and logged if the log
3415 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3416 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3417 and DATA following a set of rejected RCPTs do not count as protocol errors.
3418 In other words, Exim assumes they were pipelined, though this may not
3419 actually be the case. Of course, in all cases the client gets an
3420 appropriate error code.
3422 15. If a lookup fails in an ACL condition, a message about the failure may
3423 be available; it is used if testing the ACL cannot continue, because most
3424 such messages specify what the cause of the deferral is. However, some
3425 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3426 that caused an old message to be retained and used if a later statement
3427 caused a defer, replacing the real cause of the deferral.
3429 16. If an IP address had so many PTR records that the DNS lookup buffer
3430 was not large enough to hold them, Exim could crash while trying to process
3431 the truncated data. It now detects and logs this case.
3433 17. Further to 4.21/58, another change has been made: if (and only if) the
3434 first line of a message (the first header line) ends with CRLF, a bare LF
3435 in a subsequent header line has a space inserted after it, so as not to
3436 terminate the header.
3438 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3439 unnecessarily, used atoi() instead of strtol(), and didn't check the
3440 termination when getting file sizes from file names by regex.
3442 19. Completely re-implemented the support for maildirsize files, in the light
3443 of a number of problems with the previous contributed implementation
3444 (4.30/29). In particular:
3446 . If the quota is zero, the maildirsize file is maintained, but no quota is
3449 . If the maildir directory does not exist, it is created before any attempt
3450 to write a maildirsize file.
3452 . The quota value in the file is just a cache; if the quota is changed in
3453 the transport, the new value overrides.
3455 . A regular expression is available for excluding directories from the
3458 20. The autoreply transport checks the characters in options that define the
3459 message's headers; it allows continued headers, but it was checking with
3460 isspace() after an embedded newline instead of explicitly looking for a
3463 21. If all the "regular" hosts to which an address was routed had passed their
3464 expiry times, and had not reached their retry times, the address was
3465 bounced, even if fallback hosts were defined. Now Exim should go on to try
3468 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3469 equivalent code in the SMTP transport. Some hosts send humungous responses
3470 to HELO/EHLO, more than 1024 it seems.
3472 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3473 gives compiler warnings in some environments. I've now done it "properly",
3476 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3477 (because of problems with the built-in one) was declared to return uschar *
3478 instead of char *, causing compiler failure.
3480 25. Fixed a file descriptor leak when processing alias/forward files.
3482 26. Fixed a minor format string issue in dbfn.c.
3484 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3486 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3487 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3490 29. When a sender address is verified, it is cached, to save repeating the test
3491 when there is more than one recipient in a message. However, when the
3492 verification involves a callout, it is possible for different callout
3493 options to be set for different recipients. It is too complicated to keep
3494 track of this in the cache, so now Exim always runs a verification when a
3495 callout is required, relying on the callout cache for the optimization.
3496 The overhead is duplication of the address routing, but this should not be
3499 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3500 to be verified with callout=postmaster, and the main callout worked but the
3501 postmaster check failed, the verification correctly failed. However, if a
3502 subsequent RCPT command asked for sender verification *without* the
3503 postmaster check, incorrect caching caused this verification also to fail,
3506 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3507 it was not caching the DNS options (qualify_single, search_parents) that
3508 were used when the lookup failed. A subsequent lookup with different
3509 options therefore always gave the same answer, though there were cases
3510 where it should not have. (Example: a "domains = !$mx_any" option on a
3511 dnslookup router: the "domains" option is always processed without any
3512 widening, but the router might have qualify_single set.) Now Exim uses the
3513 cached value only when the same options are set.
3515 32. Added John Jetmore's "exipick" utility to the distribution.
3517 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3518 than a timeout (e.g. a certificate is required, and is not provided), an
3519 Exim server now closes the connection immediately. Previously it waited for
3520 the client to close - but if the client is SSL, it seems that they each
3521 wait for each other, leading to a delay before one of them times out.
3523 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3524 maintained 0.8.x compatibility because I don't think many are using it, and
3525 it is clearly obsolete.
3527 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3530 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3531 client certificate was expired. A simple patch fixes this, though I don't
3532 understand the full logic of why the verify callback is called multiple
3535 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3536 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3537 which causes problems with some clients (such as the Certicom SSL Plus
3538 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3539 disables the coutermeasure allowing Eudora to connect."
3541 38. Exim was not checking that a write() to a log file succeeded. This could
3542 lead to Bad Things if a log got too big, in particular if it hit a file
3543 size limit. Exim now panics and dies if it cannot write to a log file, just
3544 as it does if it cannot open a log file.
3546 39. Modified OS/Makefile-Linux so that it now contains
3548 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3550 The two -D definitions ensure that Exim is compiled with large file
3551 support, which makes it possible to handle log files that are bigger than
3554 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3555 instance) a domain was checked against a named list that involved a lookup,
3556 causing $domain_data to be set, then another domain was checked against the
3557 same list, then the first domain was re-checked, the value of $domain_data
3558 after the final check could be wrong. In particular, if the second check
3559 failed, it could be set empty. This bug probably also applied to
3562 41. The strip_trailing_dot option was not being applied to the address given
3563 with the -f command-line option.
3565 42. The code for reading a message's header from the spool was incrementing
3566 $received_count, but never initializing it. This meant that the value was
3567 incorrect (doubled) while delivering a message in the same process in which
3568 it was received. In the most common configuration of Exim, this never
3569 happens - a fresh exec is done - but it can happen when
3570 deliver_drop_privilege is set.
3572 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3573 it now includes up to 150 characters of the unexpected data in the log
3576 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3577 and building data strings. The size of both of these buffers was 10 000
3578 bytes - far larger than anybody would *ever* want, thought I. Needless to
3579 say, somebody hit the limit. I have increased the maximum line length to
3580 20 000 and the maximum data length of concatenated lines to 100 000. I have
3581 also fixed two bugs, because there was no checking on these buffers. Tsk,
3582 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3583 buffer is too small.
3585 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3586 lsearch lookups. Now it does.
3588 46. When parsing a route_list item in a manualroute router, a fixed-length
3589 buffer was used for the list of hosts. I made this 1024 bytes long,
3590 thinking that nobody would ever have a list of hosts that long. Wrong.
3591 Somebody had a whole pile of complicated expansion conditions, and the
3592 string was silently truncated, leading to an expansion error. It turns out
3593 that it is easier to change to an unlimited length (owing to other changes
3594 that have happened since this code was originally written) than to build
3595 structure for giving a limitation error. The length of the item that
3596 expands into the list of hosts is now unlimited.
3598 47. The lsearch lookup could not handle data where the length of text line was
3599 more than 4095 characters. Such lines were truncated, leading to shortened
3600 data being returned. It should now handle lines of any length.
3602 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3603 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3606 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3607 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3608 no longer happens in comment lines. A list of replacements is now placed
3609 at the head of all of the source files, except those whose only change is
3610 to replace PERL_COMMAND in the very first #! line.
3612 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3613 messages on the queue, with a bottom-up merge sort, using code contributed
3614 by Michael Haardt. This should make operations like -bp somewhat faster on
3615 large queues. It won't affect queue runners, except when queue_run_in_order
3618 51. Installed eximstats 1.31 in the distribution.
3620 52. Added support for SRV lookups to the dnslookup router.
3622 53. If an ACL referred to $message_body or $message_body_end, the value was not
3623 reset for any messages that followed in the same SMTP session.
3625 54. The store-handling optimization for building very long strings was not
3626 differentiating between the different store pools. I don't think this
3627 actually made any difference in practice, but I've tidied it.
3629 55. While running the routers to verify a sender address, $sender_address
3630 was still set to the sender address. This is wrong, because when routing to
3631 send a bounce to the sender, it would be empty. Therefore, I have changed
3632 it so that, while verifying a sender address, $sender_address is set to <>.
3633 (There is no change to what happens when verifying a recipient address.)
3635 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3636 target A or AAAA records (if not already returned) without resetting the
3637 qualify_single or search_parents options of the DNS resolver. These are
3638 inappropriate in this case because the targets of MX and SRV records must
3639 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3640 target that, when qualified, matched something in the local domain. These
3641 two options are now turned off when doing these lookups.
3643 57. It seems that at least some releases of Reiserfs (which does not have the
3644 concept of a fixed number of inodes) returns zero and not -1 for the
3645 number of available inodes. This interacted badly with check_spool_inodes,
3646 which assumed that -1 was the "no such thing" setting. What I have done is
3647 to check that the total number of inodes is greater than zero before doing
3648 the test of how many are available.
3650 58. When a "warn" ACL statement has a log_message modifier, the message is
3651 remembered, and not repeated. This is to avoid a lot of repetition when a
3652 message has many recipients that cause the same warning to be written.
3653 Howewer, Exim was preserving the list of already written lines for an
3654 entire SMTP session, which doesn't seem right. The memory is now reset if a
3655 new message is started.
3657 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3658 debugging output unless log_rewrite was also set.
3660 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3661 of (int)(handle) when we know that handle contains (void *)(-1).
3663 61. The Exim daemon panic-logs an error return when it closes the incoming
3664 connection. However "connection reset by peer" seems to be common, and
3665 isn't really an error worthy of noting specially, so that particular error
3668 62. When Exim is trying to find all the local interfaces, it used to panic and
3669 die if the ioctl to get the interface flags failed. However, it seems that
3670 on at least one OS (Solaris 9) it is possible to have an interface that is
3671 included in the list of interfaces, but for which you get a failure error
3672 for this call. This happens when the interface is not "plumbed" into a
3673 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3674 failure of the "get flags" call assumes that the interface is down.
3676 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3677 makes life easier for people who are doing arithmetic on fields extracted
3678 from dates, where you often get leading zeros that should not be
3679 interpreted as octal.
3681 64. Added qualify_domain to the redirect router, to override the global
3684 65. If a pathologically long header line contained very many addresses (the
3685 report of this problem mentioned 10 000) and each of them was rewritten,
3686 Exim could use up a very large amount of memory. (It kept on making new
3687 copies of the header line as it rewrote, and never released the old ones.)
3688 At the expense of a bit more processing, the header rewriting function has
3689 been changed so that it no longer eats memory in this way.
3691 66. The generation of the Received: header has been moved from the time that a
3692 message starts to be received, to the time that it finishes. The timestamp
3693 in the Received: header should now be very close to that of the <= log
3694 line. There are two side-effects of this change:
3696 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3697 logged header lines no longer include the local Received: line, because
3698 it has not yet been created. The same applies to a copy of the message
3699 that is returned to a non-SMTP sender when a message is rejected.
3701 (b) When a filter file is tested using -bf, no additional Received: header
3702 is added to the test message. After some thought, I decided that this
3705 This change does not affect the value of $received_for. It is still set
3706 after address rewriting, but before local_scan() is called.
3708 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3710 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3711 gave an unhelpful panic error message, and a defer error. I have managed to
3712 change this behaviour so that it now rejects any supplied certificate,
3713 which seems right, as the list of acceptable certificates is empty.
3715 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3716 gave an unhelpful defer error. I have not managed to make this reject any
3717 supplied certificates, but the error message it gives is "no certificate
3718 supplied", which is not helpful.
3720 70. exigrep's output now also includes lines that are not associated with any
3721 message, but which match the given pattern. Implemented by a patch from
3722 Martin Sluka, which also tidied up the Perl a bit.
3724 71. Recipient callout verification, like sender verification, was using <> in
3725 the MAIL FROM command. This isn't really the right thing, since the actual
3726 sender may affect whether the remote host accepts the recipient or not. I
3727 have changed it to use the actual sender in the callout; this means that
3728 the cache record is now keyed on a recipient/sender pair, not just the
3729 recipient address. There doesn't seem to be a real danger of callout loops,
3730 since a callout by the remote host to check the sender would use <>.
3731 [SEE ABOVE: changed after hitting problems.]
3733 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3734 temporary errors. However, in the case of such a code being given after
3735 the end of a data transmission (i.e. after ".") Exim was failing to write
3736 a retry record for the message. (Yes, there was some broken host that was
3737 actually sending 8xx at this point.)
3739 73. An unknown lookup type in a host list could cause Exim to panic-die when
3740 the list was checked. (An example that provoked this was putting <; in the
3741 middle of a list instead of at the start.) If this happened during a DATA
3742 ACL check, a -D file could be left lying around. This kind of configuration
3743 error no longer causes Exim to die; instead it causes a defer errror. The
3744 incident is still logged to the main and panic logs.
3746 74. Buglet left over from Exim 3 conversion. The message "too many messages
3747 in one connection" was written to the rejectlog but not the mainlog, except
3748 when address rewriting (yes!) was being logged.
3750 75. Added write_rejectlog option.
3752 76. When a system filter was run not as root (that is, when system_filter_user
3753 was set), the values of the $n variables were not being returned to the
3754 main process; thus, they were not subsequently available in the $sn
3757 77. Added +return_path_on_delivery log selector.
3759 78. A connection timeout was being treated differently from recipients deferred
3760 when testing hosts_max_try with a message that was older than the host's
3761 retry timeout. (The host should not be counted, thus allowing all hosts to
3762 be tried at least once before bouncing.) This may have been the cause of an
3763 occasionally reported bug whereby a message would remain on the queue
3764 longer than the retry timeout, but would be bounced if a delivery was
3765 forced. I say "may" because I never totally pinned down the problem;
3766 setting up timeout/retry tests is difficult. See also the next item.
3768 79. The ultimate address timeout was not being applied to errors that involved
3769 a combination of host plus message (for example, a timeout on a MAIL
3770 command). When an address resolved to a number of possible hosts, and they
3771 were not all tried for each delivery (e.g. because of hosts_max_try), a
3772 message could remain on the queue longer than the retry timeout.
3774 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3777 81. Fixed an obscure SMTP outgoing bug which required at least the following
3778 conditions: (a) there was another message waiting for the same server;
3779 (b) the server returned 5xx to all RCPT commands in the first message so
3780 that the message was not completed; (c) the server dropped the connection
3781 or gave a negative response to the RSET that Exim sends to abort the
3782 transaction. The observed case was a dropped connection after DATA that had
3783 been sent in pipelining mode. That is, the server had advertised PIPELINING
3784 but was not implementing it correctly. The effect of the bug was incorrect
3785 behaviour, such as trying another host, and this could lead to a crash.
3791 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3792 and daemon.c were passed as pointers to ints; they should have been
3793 pointers to socklen_t variables (which are typically unsigned ints).
3795 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3798 3. Fixed a really odd bug that affected only the testing scheme; patching a
3799 certain fixed string in the binary changed the value of another string that
3800 happened to be identical to the end of the original first string.
3802 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3803 name", it returns that address as the IP address. On some operating
3804 systems (e.g. Solaris), it also passes back the IP address string as the
3805 "host name". However, on others (e.g. Linux), it passes back an empty
3806 string. Exim wasn't checking for this, and was changing the host name to an
3807 empty string, assuming it had been canonicized.
3809 5. Although rare, it is permitted to have more than one PTR record for a given
3810 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3811 all the names associated with an address, because they do in Solaris.
3812 However, it seems that they do not in Linux for data that comes from the
3813 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3814 I found this out when I moved to a new Linux workstation and tried to run
3815 the Exim test suite.
3817 To get round this problem I have changed the code so that it now does its
3818 own call to the DNS to look up PTR records when searching for a host name.
3819 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3820 addresses that are only in /etc/hosts are still found.
3822 This behaviour is, however, controlled by an option called host_lookup_
3823 order, which defaults to "bydns:byaddr". If people want to use the other
3824 order, or indeed, just use one or the other means of lookup, they can
3825 specify it in this variable.
3827 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3828 some operating systems, this comes back from gethostbyaddr() as an empty
3829 string, and this is what Exim used to test for. However, it seems that in
3830 other systems, "." is yielded. Exim now tests for this case too.
3832 7. The values of check_spool_space and check_log_space are now held internally
3833 as a number of kilobytes instead of an absolute number of bytes. If a
3834 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3835 kilobyte. This means that much larger values can be stored.
3837 8. Exim monitor: an attempt to get the action menu when not actually pointing
3838 at a message produces an empty menu entitled "No message selected". This
3839 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3840 no entries in it ("Shell widget menu has zero width and/or height"). So I
3841 have added a single, blank menu entry in this case.
3843 9. Added ${quote_local_part.
3845 10. MIME decoding is now applied to the contents of Subject: header lines when
3848 11. Now that a reference to $sender_host_address automatically causes a reverse
3849 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3850 host lookup before query-style lookups in lists that might use this
3851 variable. This has therefore been abolished, and the "net-" prefix is no
3852 longer necessary for query-style lookups.
3854 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3855 have been a typo for LFLAGS, so it has been changed.
3857 13. The install script calls Exim with "-C /dev/null" in order to find the
3858 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3859 to be output. Howeve, since Exim outputs its version number before the
3860 error, it didn't break the script. It just looked ugly. I fixed this by
3861 always allowing "-C /dev/null" if the caller is root.
3863 14. Ignore overlarge ACL variable number when reading spool file - insurance
3864 against a later release with more variables having written the file.
3866 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3867 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3868 this, as well as the form without the "IPv6" on the front.
3870 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3871 OS/Makefile-Darwin file.
3873 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3874 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3876 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3877 variable. However, some OS use ints for inodes. I've added cast to long int
3878 to get rid of the compiler warning.
3880 19. I had forgotten to lock out "/../" in configuration file names when
3881 ALT_CONFIG_PREFIX was set.
3883 20. Routers used for verification do not need to specify transports. However,
3884 if such a router generated a host list, and callout was configured, Exim
3885 crashed, because it could not find a port number from the (non-existent)
3886 transport. It now assumes port 25 in this circumstance.
3888 21. Added the -t option to exigrep.
3890 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3891 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3892 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3893 them is set, LOOKUP_LSEARCH is forced.
3895 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3897 24. Added sender and host information to the "rejected by local_scan()" log
3898 line; previously there was no indication of these.
3900 25. Added .include_if_exists.
3902 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3903 Linux. It turns out that not all file systems support this. Apparently some
3904 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3905 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3906 not supported on the file descriptor, is now ignored when Exim is trying to
3907 sync a directory. This applies only to Linux.
3909 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3911 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3912 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3915 29. Added support for maildirsize files from supplied patch (modified a bit).
3917 30. The use of :fail: followed by an empty string could lead Exim to respond to
3918 sender verification failures with (e.g.):
3920 550 Verification failed for <xxx>
3921 550 Sender verify failed
3923 where the first response line was missing the '-' that indicates it is not
3924 the final line of the response.
3926 31. The loop for finding the name of the user that called Exim had a hardwired
3927 limit of 10; it now uses the value of finduser_retries, which is used for
3928 all other user lookups.
3930 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3933 33. Exim was neglecting to zero errno before one call of strtol() when
3934 expanding a string and expecting an integer value. On some systems this
3935 resulted in spurious "integer overflow" errors. Also, it was casting the
3936 result into an int without checking.
3938 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3939 did not work. The code looks as if it has *never* worked, though it appears
3940 to have been documented since at least releast 1.62. I have made it work.
3942 35. The "timeout_DNS" error in retry rules, also documented since at least
3943 1.62, also never worked. As it isn't clear exactly what this means, and
3944 clearly it isn't a major issue, I have abolished the feature by treating it
3945 as "timeout", and writing a warning to the main and panic logs.
3947 36. The display of retry rules for -brt wasn't always showing the error code
3950 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3951 timeout_connect_A, timeout_connect_MX.
3953 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3954 to the empty sender.
3956 39. The daemon was not analysing the content of -oX till after it had closed
3957 stderr and disconnected from the controlling terminal. This meant that any
3958 syntax errors were only noted on the panic log, and the return code from
3959 the command was 0. By re-arranging the code a little, I've made the
3960 decoding happen first, so such errors now appear on stderr, and the return
3961 code is 1. However, the actual setting up of the sockets still happens in
3962 the disconnected process, so errors there are still only recorded on the
3965 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3966 connections (as happens on some IP stacks) was logged at start up time as
3967 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3968 it from "IPv6 and IPv4", which means that two separate sockets are being
3971 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3972 says whether AF_INET or AF_INET6 was passed as an argument.
3974 42. Exiwhat output was messed up when time zones were included in log
3977 43. Exiwhat now gives more information about the daemon's listening ports,
3978 and whether -tls-on-connect was used.
3980 44. The "port" option of the smtp transport is now expanded.
3982 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3983 silently ignored. Now an error message is written to the main and panic
3986 46. There's a new ACL modifier called "logwrite" which writes to a log file
3987 as soon as it is encountered.
3989 47. Added $local_user_uid and $local_user_gid at routing time.
3991 48. Exim crashed when trying to verify a sender address that was being
3994 49. Exim was recognizing only a space character after ".include". It now also
3995 recognizes a tab character.
3997 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3998 extracting the relevant information from the specification. The man page no
3999 longer contains scrambled data for the -d option, and I've added a section
4000 at the front about calling Exim under different names.
4002 51. Added "extra_headers" argument to the "mail" command in filter files.
4004 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4007 53. Installed eximstats 1.29.
4009 54. Added transport_filter_timeout as a generic transport option.
4011 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4012 Cc: header lines. This was required by RFC 822, but it not required by RFC
4015 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4016 incoming messages that did not have them. Now it does so only if the
4017 message originates locally, that is, if there is no associated remote host
4018 address. When Resent- header lines are present, this applies to the Resent-
4019 lines rather than the non-Resent- lines.
4021 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4022 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4024 58. Messages for configuration errors now include the name of the main
4025 configuration file - useful now that there may be more than one file in a
4026 list (.included file names were always shown).
4028 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4029 for those rare installations that do not start the daemon as root or run it
4030 setuid root. I've cut out the call to initgroups() if the daemon is not
4033 60. The Exim user and group can now be bound into the binary as text strings
4034 that are looked up at the start of Exim's processing.
4036 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4038 62. Added $mailstore_basename variable.
4040 63. Installed patch to sieve.c from Michael Haardt.
4042 64. When Exim failed to open the panic log after failing to open the main log,
4043 the original message it was trying to log was written to stderr and debug
4044 output, but if they were not available (the usual case in production), it
4045 was lost. Now it is written to syslog before the two lines that record the
4046 failures to open the logs.
4048 65. Users' Exim filters run in subprocesses under the user's uid. It is
4049 possible for a "deliver" command or an alias in a "personal" command to
4050 provoke an address rewrite. If logging of address rewriting is configured,
4051 this fails because the process is not running as root or exim. There may be
4052 a better way of dealing with this, but for the moment (because 4.30 needs
4053 to be released), I have disabled address rewrite logging when running a
4054 filter in a non-root, non-exim process.
4060 1. The buildconfig auxiliary program wasn't quoting the value set for
4061 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4062 not defined. This bug was masked in 4.22 by the effect that was fixed in
4065 2. Some messages that were rejected after a message id was allocated were
4066 shown as "incomplete" by exigrep. It no longer does this for messages that
4067 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4069 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4070 have allow_domain_literals set, the ID did not get logged in the <= line.
4071 Domain literals are now always recognized in Message-ID: header lines.
4073 4. The first argument for a ${extract expansion item is the key name or field
4074 number. Leading and trailing spaces in this item were not being ignored,
4075 causing some misleading effects.
4077 5. When deliver_drop_privilege was set, single queue runner processes started
4078 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4079 same command in the process it spins off) were not dropping privilege.
4081 6. When the daemon running as "exim" started a queue runner, it always
4082 re-executed Exim in the spun-off process. This is a waste of effort when
4083 deliver_drop_privilege is set. The new process now just calls the
4084 queue-runner function directly.
4090 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4093 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4094 ignored. Though the use of -oP was forcing the writing of a pid file, it
4095 was always written to the default place.
4097 3. If the message "no IP address found for host xxxx" is generated during
4098 incoming verification, it is now followed by identification of the incoming
4099 connection (so you can more easily find what provoked it).
4101 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4103 5. Added some features to "harden" Exim a bit more against certain attacks:
4105 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4106 be put in Local/Makefile. This is like the never_users runtime option,
4107 but it cannot be overridden. The default setting is "root".
4109 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4110 prefix string with which any file named in a -C command line option
4113 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4114 is retained for -C and -D only if the caller of Exim is root. Without
4115 it, the exim user may also use -C and -D and retain privilege.
4117 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4118 command line option is disabled.
4120 6. Macro names set by the -D option must start with an upper case letter, just
4121 like macro names defined in the configuration file.
4123 7. Added "dereference=" facility to LDAP.
4125 8. Two instances of the typo "uknown" in the source files are fixed.
4127 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4128 the Configure-Makefile script screwed up while processing it.
4130 10. Incorporated PCRE 4.4.
4132 11. The SMTP synchronization check was not operating right at the start of an
4133 SMTP session. For example, it could not catch a HELO sent before the client
4134 waited for the greeting. There is now a check for outstanding input at the
4135 point when the greeting is written. Because of the duplex, asynchronous
4136 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4137 way, but not yet received, when the check is performed.
4139 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4140 on TCP/IP sockets, because this apparently causes some broken clients to
4143 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4144 unchanged) from the Cygwin maintainer.
4146 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4147 was supported instead of testing for mbx. Effectively a typo.
4149 15. The spa authenticator server code was not checking that the input it
4150 received was valid base64.
4152 16. The debug output line for the "set" modifier in ACLs was not showing the
4153 name of the variable that was being set.
4155 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4157 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4158 lookup on its own. Something else was needed to trigger the lookup. For
4159 example, a match in host_lookup or the need for a host name in a host list.
4160 Now, if $sender_host_name is referenced and the host name has not yet been
4161 looked up, a lookup is performed. If the lookup fails, the variable remains
4162 empty, and $host_lookup_failed is set to "1".
4164 19. Added "eqi" as a case-independent comparison operator.
4166 20. The saslauthd authentication condition could segfault if neither service
4167 nor realm was specified.
4169 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4170 error message that was logged was misleading, and incoming SMTP
4171 connections were dropped. The message is now more accurate, and temporary
4172 errors are given to SMTP connections.
4174 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4175 (except RSET) with a 421 error, until QUIT is received. However, it was
4176 failing to send a response to QUIT.
4178 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4179 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4180 ACL is now run after the helo_try_verify_hosts code.
4182 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4183 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4184 case-independent, so other case variants are also recognized). Apparently
4185 some systems use these upper case variants.
4187 25. If more than two messages were waiting for the same host, and a transport
4188 filter was specified for the transport, Exim sent two messages over the
4189 same TCP/IP connection, and then failed with "socket operation on non-
4190 socket" when it tried to send the third.
4192 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4194 27. The extern definition of crypt16() in expand.c was not being excluded when
4195 the OS had its own crypt16() function.
4197 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4198 as a preferred synonym for return_size_limit, both as an option and as an
4201 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4203 30. Changed the default configuration ACL to relax the local part checking rule
4204 for addresses that are not in any local domains. For these addresses,
4205 slashes and pipe symbols are allowed within local parts, but the sequence
4206 /../ is explicitly forbidden.
4208 31. SPA server authentication was not clearing the challenge buffer before
4211 32. log_message in a "warn" ACL statement was writing to the reject log as
4212 well as to the main log, which contradicts the documentation and doesn't
4213 seem right (because no rejection is happening). So I have stopped it.
4215 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4216 However, I am unable to do any testing of this.
4218 34. Fixed an infelicity in the appendfile transport. When checking directories
4219 for a mailbox, to see if any needed to be created, it was accidentally
4220 using path names with one or more superfluous leading slashes; tracing
4221 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4223 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4224 contents are added to the log line that is written for every discarded
4225 recipient. (Previously a log_message setting was ignored.)
4227 36. The ${quote: operator now quotes the string if it is empty.
4229 37. The install script runs exim in order to find its version number. If for
4230 some reason other than non-existence or emptiness, which it checks, it
4231 could not run './exim', it was installing it with an empty version number,
4232 i.e. as "exim-". This error state is now caught, and the installation is
4235 38. An argument was missing from the function that creates an error message
4236 when Exim fails to connect to the socket for saslauthd authentication.
4237 This could cause Exim to crash, or give a corrupted message.
4239 39. Added isip, isip4, and isip6 to ${if conditions.
4241 40. The ACL variables $acl_xx are now saved with the message, and can be
4242 accessed later in routers, transports, and filters.
4244 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4245 strings in the file are not string-expanded.
4247 42. If a MAIL command specified a SIZE value that was too large to fit into an
4248 int variable, the check against message_size_limit failed. Such values are
4249 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4250 day this will have to be increased, but I don't think I want to be around
4251 when emails are that large.
4258 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4259 iconv() is not standard in FreeBSD.
4261 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4262 IPv6 enabled. The observed symptom was a segmentation fault on return from
4263 the function os_common_find_running_interfaces() in src/os.c.
4265 3. In the check_special_case() function in daemon.c I had used "errno" as an
4266 argument name, which causes warnings on some systems. This was basically a
4267 typo, since it was named "eno" in the comments!
4269 4. The code that waits for the clock to tick (at a resolution of some fraction
4270 of a second) so as to ensure message-id uniqueness was always waiting for
4271 at least one whole tick, when it could have waited for less. [This is
4272 almost certainly not relevant at current processor speeds, where it is
4273 unlikely to ever wait at all. But we try to future-proof.]
4275 5. The function that sleeps for a time interval that includes fractions of a
4276 second contained a race. It did not block SIGALRM between setting the
4277 timer, and suspending (a couple of lines later). If the interval was short
4278 and the sigsuspend() was delayed until after it had expired, the suspension
4279 never ended. On busy systems this could lead to processes getting stuck for
4282 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4283 process, before it forks any delivery processes. The open lookup caching
4284 mechanism meant that the open file or database connection was passed into
4285 the delivery process. The problem was that delivery processes always tidy
4286 up cached lookup data. This could cause a problem for the next delivery
4287 process started by the queue runner, because the external queue runner
4288 process does not know about the closure. So the next delivery process
4289 still has data in the lookup cache. In the case of a file lookup, there was
4290 no problem because closing a file descriptor in a subprocess doesn't affect
4291 the parent. However, if the lookup was caching a connection to a database,
4292 the connection was closed, and the second delivery process was likely to
4293 see errors such as "PGSQL: query failed: server closed the connection
4294 unexpectedly". The problem has been fixed by closing all cached lookups
4295 in a queue runner before running a delivery process.
4297 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4298 seem to have the "const" qualifier which it has on other OS. I've
4301 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4302 *of the same type* (client or server) with the same public name that an
4303 error should be diagnosed.
4305 9. When Exim looked up a host name for an IP address, but failed to find the
4306 original IP address when looking up the host name (a safety check), it
4307 output the message "<ip address> does not match any IP for NULL", which was
4308 confusing, to say the least. The bug was that the host name should have
4309 appeared instead of "NULL".
4311 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4312 user that is built into the binary, and the -C or -D options is used, root
4313 privilege is dropped before the configuration file is read. In addition,
4314 logging is switched to stderr instead of the normal log files. If the
4315 configuration then re-defines the Exim user, the unprivileged environment
4316 is probably not what is expected, so Exim logs a panic warning message (but
4319 However, if deliver_drop_privilege is set, the unprivileged state may well
4320 be exactly what is intended, so the warning has been cut out in that case,
4321 and Exim is allowed to try to write to its normal log files.
4327 1. smtp_return_error_details was not giving details for temporary sender
4328 or receiver verification errors.
4330 2. Diagnose a configuration error if two authenticators have the same public
4333 3. Exim used not to create the message log file for a message until the first
4334 delivery attempt. This could be confusing when incoming messages were held
4335 for policy or load reasons. The message log file is now created at the time
4336 the message is received, and an initial "Received" line is written to it.
4338 4. The automatically generated man page for command line options had a minor
4339 bug that caused no ill effects; however, a more serious problem was that
4340 the procedure for building the man page automatically didn't always
4341 operate. Consequently, release 4.20 contains an out-of-date version. This
4342 shouldn't happen again.
4344 5. When building Exim with embedded Perl support, the script that builds the
4345 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4346 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4348 6. The freeze_tell option was not being used for messages that were frozen on
4349 arrival, either by an ACL or by local_scan().
4351 7. Added the smtp_incomplete_transaction log selector.
4353 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4354 was accepting AUTH without a new EHLO.
4356 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4357 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4360 10. Logging of TCP/IP connections (when configured) now happens in the main
4361 daemon process instead of the child process, so that the TCP/IP connection
4362 count is more accurate (but it can never be perfect).
4364 11. The use of "drop" in a nested ACL was not being handled correctly in the
4365 outer ACL. Now, if condition failure induced by the nested "drop" causes
4366 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4367 or "require"), the connection is dropped.
4369 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4370 that yield "discard" can now be used with an "accept" or a "discard" verb,
4371 but an error is generated for any others (because I can't see a useful way
4372 to define what should happen).
4374 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4375 are now processed in the same way as lines in the Exim configuration file.
4376 In particular, continuation lines are supported.
4378 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4380 15. Added -ti meaning -t -i.
4382 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4383 domains, and warn by logging if others are found.
4385 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4386 returned from GIFCONF when Exim is trying to find the list of interfaces on
4387 a host. The code in os.c has been modified to copy each ifreq to an aligned
4388 structure in all cases.
4390 Also, in some cases, the returned ifreq's were being copied to a 'struct
4391 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4392 means the last couple of bytes of an IPv6 address could be chopped if the
4393 ifreq contained only a normal sockaddr (14 bytes storage).
4395 18. Named domain lists were not supported in the hosts_treat_as_local option.
4396 An entry such as +xxxx was not recognized, and was treated as a literal
4399 19. Ensure that header lines added by a DATA ACL are included in the reject log
4400 if the ACL subsequently rejects the message.
4402 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4403 MD5 (which is deprecated).
4405 21. When testing a filter file using -bf, Exim was writing a message when it
4406 took the sender from a "From " line in the message, but it was not doing so
4407 when it took $return_path from a Return-Path: header line. It now does.
4409 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4410 with a valid header line field name (a series of printing characters
4411 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4413 23. Changed "disc" in the source to "disk" to conform to the documentation and
4414 the book and for uniformity.
4416 24. Ignore Sendmail's -Ooption=value command line item.
4418 25. When execve() failed while trying to run a command in a pipe transport,
4419 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4420 could be confused with a return value of 69 from the command itself. This
4421 has been changed to 127, the value the shell returns if it is asked to run
4422 a non-existent command. The wording for the related log line suggests a
4423 non-existent command as the problem.
4425 26. If received_header_text expands to an empty string, do not add a Received:
4426 header line to the message. (Well, it adds a token one on the spool, but
4427 marks it "old" so that it doesn't get used or transmitted.)
4429 27. Installed eximstats 1.28 (addition of -nt option).
4431 28. There was no check for failure on the call to getsockname() in the daemon
4432 code. This can fail if there is a shortage of resources on the system, with
4433 ENOMEM, for example. A temporary error is now given on failure.
4435 29. Contrary to the C standard, it seems that in some environments, the
4436 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4437 program. Exim now does this explicitly; it affects the formatting of
4438 timestamps using strftime().
4440 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4441 complaints. I've now initialized all the variables, to avoid this.
4443 32. Header lines added by a system filter were not being "seen" during
4444 transport-time rewrites.
4446 33. The info_callback() function passed to OpenSSL is set up with type void
4447 (*)(SSL *, int, int), as described somewhere. However, when calling the
4448 function (actually a macro) that sets it up, the type void(*)() is
4449 expected. I've put in a cast to prevent warnings from picky compilers.
4451 34. If a DNS black list lookup found a CNAME record, but there were no A
4452 records associated with the domain it pointed at, Exim crashed.
4454 35. If a DNS black list lookup returned more than one A record, Exim ignored
4455 all but the first. It now scans all returned addresses if a particular IP
4456 value is being sought. In this situation, the contents of the
4457 $dnslist_value variable are a list of all the addresses, separated by a
4460 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4461 to accept a host name and all its aliases if the forward lookup for any of
4462 them yielded the IP address of the incoming connection. Now it accepts only
4463 those names whose forward lookup yields the correct IP address. Any other
4464 names are discarded. This closes a loophole whereby a rogue DNS
4465 administrator could create reverse DNS records to break through a
4466 wildcarded host restriction in an ACL.
4468 37. If a user filter or a system filter that ran in a subprocess used any of
4469 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4470 the wrong values were passed to the pipe command ($thisaddress had the
4471 value of $0, $0 had the value of $1, etc). This bug was introduced by
4472 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4474 38. Improved the line breaking for long SMTP error messages from ACLs.
4475 Previously, if there was no break point between 40 and 75 characters, Exim
4476 left the rest of the message alone. Two changes have been made: (a) I've
4477 reduced the minimum length to 35 characters; (b) if it can't find a break
4478 point between 35 and 75 characters, it looks ahead and uses the first one
4479 that it finds. This may give the occasional overlong line, but at least the
4480 remaining text gets split now.
4482 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4483 file descriptors might be low, and that setting 1000 would always raise it.
4484 It turns out that in some environments, the limit is already over 1000 and
4485 that lowering it causes trouble. So now Exim takes care not to decrease it.
4487 40. When delivering a message, the value of $return_path is set to $sender_
4488 address at the start of routing (routers may change the value). By an
4489 oversight, this default was not being set up when an address was tested by
4490 -bt or -bv, which affected the outcome if any router or filter referred to
4493 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4494 log only when "message" or "log_message" are set. However, if one of the
4495 conditions was an address verification, or a call to a nested ACL, the
4496 messages generated by the underlying test were being passed through. This
4497 no longer happens. The underlying message is available in $acl_verify_
4498 message for both "message" and "log_message" expansions, so it can be
4499 passed through if needed.
4501 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4502 new expansion $bh_ to give the encoded byte string without charset
4503 translation. Translation happens only if iconv() is available; HAVE_ICONV
4504 indicates this at build time. HEADERS_CHARSET gives the charset to
4505 translate to; headers_charset can change it in the configuration, and
4506 "headers charset" can change it in an individual filter file.
4508 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4509 that creates RFC 2047 encoded "words" labels them as that charset instead
4510 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4511 expansion operator; (ii) when Exim creates a From: line for a local
4512 message; (iii) when a header line is rewritten to include a "phrase" part.
4514 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4515 buggy, causing it to skip the first lines of messages whose message ID
4516 ended in 'D'. This would not have bitten before Exim release 4.14, because
4517 message IDs were unlikely to end in 'D' before then. The effect was to have
4518 incorrect size information for certain domains.
4520 45. #include "config.h" was missing at the start of the crypt16.c module. This
4521 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4524 46. If there was a timeout during a "random" callout check, Exim treated it as
4525 a failure of the random address, and carried on sending RSET and the real
4526 address. If the delay was just some slowness somewhere, the response to the
4527 original RCPT would be taken as a response to RSET and so on, causing
4528 mayhem of various kinds.
4530 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4531 when I implemented it. It didn't allow for the fact that some option values
4532 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4533 the right test for positive values.
4535 48. Domain names in DNS records are case-independent. Exim always looks them up
4536 in lower case. Some resolvers return domain names in exactly the case they
4537 appear in the zone file, that is, they may contain uppercase letters. Not
4538 all resolvers do this - some return always lower case. Exim was treating a
4539 change of case by a resolver as a change of domain, similar to a widening
4540 of a domain abbreviation. This triggered its re-routing code and so it was
4541 trying to route what was effectively the same domain again. This normally
4542 caused routing to fail (because the router wouldn't handle the domain
4543 twice). Now Exim checks for this case specially, and just changes the
4544 casing of the domain that it ultimately uses when it transmits the message
4547 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4550 50. If a filter generated a file delivery with a non-absolute name (possible if
4551 no home directory exists for the router), the forbid_file option was not
4554 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4555 the existing equality matching.
4557 52. Exim was using ints instead of ino_t variables in some places where it was
4558 dealing with inode numbers.
4560 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4561 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4562 called TMPDIR, and if it finds it is different, it changes its value.
4564 54. The smtp_printf() function is now made available to local_scan() so
4565 additional output lines can be written before returning. There is also an
4566 smtp_fflush() function to enable the detection of a dropped connection.
4567 The variables smtp_input and smtp_batched_input are exported to
4570 55. Changed the default runtime configuration: the message "Unknown user"
4571 has been removed from the ACL, and instead placed on the localuser router,
4572 using the cannot_route_message feature. This means that any verification
4573 failures that generate their own messages won't get overridden. Similarly,
4574 the "Unrouteable address" message that was in the ACL for unverifiable
4575 relay addresses has also been removed.
4577 56. Added hosts_avoid_esmtp to the smtp transport.
4579 57. The exicyclog script was not checking for the esoteric option
4580 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4581 will work only if exicyclog is run under the appropriate euid.
4583 58. Following a discussion on the list, the rules by which Exim recognises line
4584 endings on incoming messages have been changed. The -dropcr and drop_cr
4585 options are now no-ops, retained only for backwards compatibility. The
4586 following line terminators are recognized: LF CRLF CR. However, special
4587 processing applies to CR:
4589 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4590 nor a local message in the state where . is a terminator.
4592 (ii) If a bare CR is encountered in a header line, an extra space is added
4593 after the line terminator so as not to end the header. The reasoning
4594 behind this is that bare CRs in header lines are most likely either
4595 to be mistakes, or people trying to play silly games.
4597 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4598 was being incorrectly given as 18 bytes larger than it should have been.
4599 This is a VOB (very old bug).
4601 60. This may never have affected anything current, but just in case it has:
4602 When the local host is found other than at the start of a list of hosts,
4603 the local host, those with the same MX, and any that follow, are discarded.
4604 When the list in question was part of a longer list of hosts, the following
4605 hosts (not currently being processed) were also being discarded. This no
4606 longer happens. I'm not sure if this situation could ever has previously
4609 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4612 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4613 header. This is something that is recommended in a new Internet Draft, and
4614 is something that is documented as being done by Sendmail. There are two
4615 possible values. For messages generated by the autoreply transport, Exim
4618 Auto-Submitted: auto-replied
4620 whereas for all other generated messages (e.g. bounces) it adds
4622 Auto-Submitted: auto-generated
4624 63. The "personal" condition in filters now includes a test for the
4625 Auto-Submitted: header. If it contains the string "auto-" the message it
4626 not considered personal.
4628 64. Added rcpt_include_affixes as a generic transport option.
4630 65. Added queue_only_override (default true).
4632 66. Added the syslog_duplication option.
4634 67. If what should have been the first header line of a message consisted of
4635 a space followed by a colon, Exim was mis-interpreting it as a header line.
4636 It isn't of course - it is syntactically invalid and should therefore be
4637 treated as the start of the message body. The misbehaviour could have
4638 caused a number of strange effects, including loss of data in subsequent
4639 header lines, and spool format errors.
4641 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4642 client host had authenticated. This control can now be exercised by an ACL
4643 for more flexibility.
4645 69. By default, callouts do not happen when testing with -bh. There is now a
4646 variant, -bhc, which does actually run the callout code, including
4647 consulting and updating the callout cache.
4649 70. Added support for saslauthd authentication, courtesy of Alexander
4652 71. If statvfs() failed on the spool or log directories while checking their
4653 size for availability, Exim confusingly gave the error "space shortage".
4654 Furthermore, in debugging mode it crashed with a floating point exception.
4655 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4656 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4657 serious problem, Exim now writes to the main and panic logs when this
4658 happens, with details of the failure. It then refuses to accept the
4659 incoming message, giving the message "spool directory problem" or "log
4660 directory problem" with a 421 code for SMTP messages.
4662 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4663 0, 1, and 2 exist, because some OS complain for execs without them (see
4664 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4665 descriptors. However, the code omitted to check that the open succeeded,
4666 causing mysterious errors if for some reason the permissions on /dev/null
4667 got screwed. Now Exim writes a message to the main and panic logs, and
4668 bombs out if it can't open /dev/null.
4670 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4671 interact so that it is all more flexible. It is supposed to remain
4672 backwards compatible. Also added extra_local_interfaces.
4674 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4675 to bomb out with an assertion failure - to the client this appears as a
4676 connection drop. This problem occurs in the part of the code that was taken
4677 from the Samba project. Fortunately, the assertion is in a very simple
4678 function, so I have fixed this by reproducing the function inline in the
4679 one place where it is called, and arranging for authentication to fail
4680 instead of killing the process with assert().
4682 75. The SPA client code was not working when the server requested OEM rather
4683 than Unicode encoding.
4685 76. Added code to make require_files with a specific uid setting more usable in
4686 the case where statting the file as root fails - usually a non-root-mounted
4687 NFS file system. When this happens and the failure is EACCES, Exim now
4688 forks a subprocess and does the per-uid checking as the relevant uid.
4690 77. Added process_log_path.
4692 78. If log_file_path was not explicitly set, a setting of check_log_space or
4693 check_log_inodes was ignored.
4695 79. If a space check for the spool or log partitions fails, the incident is now
4696 logged. Of course, in the latter case the data may get lost...
4698 80. Added the %p formatting code to string_format() so that it can be used to
4699 print addresses in debug_print(). Adjusted all the address printing in the
4700 debugging in store.c to use %p rather than %d.
4702 81. There was a concern that a line of code in smtp_in.c could overflow a
4703 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4704 initially expressed, the concern was not well-founded, because trailing
4705 spaces are removed early. However, if the trailing spaces were followed by
4706 a NULL, they did not get removed, so the overflow was possible. Two fixes
4709 (a) I re-wrote the offending code in a cleaner fashion.
4710 (b) If an incoming SMTP command contains a NULL character, it is rejected
4713 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4714 runs initgroups(), so that if the Exim user is in any additional groups,
4715 they will be used during message reception.
4721 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob