1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.528 2007/08/29 13:58:57 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 PH/01 Another patch from the Sieve maintainer.
11 PH/02 When an IPv6 address is converted to a string for single-key lookup
12 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
13 dots are used instead of colons so that keys in lsearch files need not
14 contain colons. This was done some time before quoting was made available
15 in lsearch files. However, iplsearch files do require colons in IPv6 keys
16 (notated using the quote facility) so as to distinguish them from IPv4
17 keys. This meant that lookups for IP addresses in host lists did not work
18 for iplsearch lookups.
20 This has been fixed by arranging for IPv6 addresses to be expressed with
21 colons if the lookup type is iplsearch. This is not incompatible, because
22 previously such lookups could never work.
24 The situation is now rather anomolous, since one *can* have colons in
25 ordinary lsearch keys. However, making the change in all cases is
26 incompatible and would probably break a number of configurations.
28 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
31 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
32 conversion specification without a maximum field width, thereby enabling
33 a rogue spamd server to cause a buffer overflow. While nobody in their
34 right mind would setup Exim to query an untrusted spamd server, an
35 attacker that gains access to a server running spamd could potentially
36 exploit this vulnerability to run arbitrary code as the Exim user.
38 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
39 $primary_hostname instead of what libspf2 thinks the hosts name is.
41 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
42 a directory entry by the name of the lookup key. Previously, if a
43 symlink pointed to a non-existing file or a file in a directory that
44 Exim lacked permissions to read, a lookup for a key matching that
45 symlink would fail. Now it is enough that a matching directory entry
46 exists, symlink or not. (Bugzilla 503.)
48 PH/03 The body_linecount and body_zerocount variables are now exported in the
51 PH/04 Added the $dnslist_matched variable.
53 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
54 This means they are set thereafter only if the connection becomes
57 PH/06 Added the client_condition to authenticators so that some can be skipped
58 by clients under certain conditions.
60 PH/07 The error message for a badly-placed control=no_multiline_responses left
61 "_responses" off the end of the name.
63 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
65 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
66 (without spaces) instead of just copying the configuration text.
68 PH/10 Added the /noupdate option to the ratelimit ACL condition.
70 PH/11 Added $max_received_linelength.
72 PH/12 Added +ignore_defer and +include_defer to host lists.
74 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
75 way in which PCRE > 7.0 is built.
77 PH/14 Implemented queue_only_load_latch.
79 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
80 MAIL command. The effect was to mangle the value on 64-bit systems.
82 PH/16 Another patch from the Sieve maintainer.
84 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
86 PH/18 If a system quota error occurred while trying to create the file for
87 a maildir delivery, the message "Mailbox is full" was not appended to the
88 bounce if the delivery eventually timed out. Change 4.67/27 below applied
89 only to a quota excession during the actual writing of the file.
91 PH/19 It seems that peer DN values may contain newlines (and other non-printing
92 characters?) which causes problems in log lines. The DN values are now
93 passed through string_printing() before being added to log lines.
95 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
96 and InterBase are left for another time.)
98 PH/21 Added message_body_newlines option.
100 PH/22 Guard against possible overflow in moan_check_errorcopy().
106 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
107 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
110 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
111 issue a MAIL command.
113 PH/02 In an ACL statement such as
115 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
117 if a client was not listed at all, or was listed with a value other than
118 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
119 the condition was not true (as it should be), so access was not denied.
120 The bug was that the ! inversion was incorrectly passed on to the second
121 item. This has been fixed.
123 PH/03 Added additional dnslists conditions == and =& which are different from
124 = and & when the dns lookup returns more than one IP address.
126 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
127 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
129 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
130 FSYNC, which compiles an option called disable_fsync that allows for
131 bypassing fsync(). The documentation is heavily laced with warnings.
133 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
135 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
136 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
137 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
138 including adding "make clean"; (3) Added -fPIC when compiling the test
139 dynamically loaded module, to get rid of a warning.
141 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
142 message fails, move_frozen_messages = true and ignore_bounce_errors_after
143 = 0s. The bug is otherwise harmless.
145 PH/07 There was a bug in the dovecot authenticator such that the value of
146 $auth1 could be overwritten, and so not correctly preserved, after a
147 successful authentication. This usually meant that the value preserved by
148 the server_setid option was incorrect.
150 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
152 PH/09 Installed PCRE release 7.0.
154 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
155 run for batched SMTP input. It is now run at the start of every message
156 in the batch. While fixing this I discovered that the process information
157 (output by running exiwhat) was not always getting set for -bs and -bS
158 input. This is fixed, and it now also says "batched" for BSMTP.
160 PH/11 Added control=no_pipelining.
162 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
163 patch, slightly modified), and move the expansion of helo_data till after
164 the connection is made in the smtp transport (so it can use these
167 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
169 PH/14 Added log_selector = +pid.
171 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
173 PH/16 Add ${if forany and ${if forall.
175 PH/17 Added dsn_from option to vary the From: line in DSNs.
177 PH/18 Flush SMTP output before performing a callout, unless control =
178 no_callout_flush is set.
180 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
181 was true (the default) a successful delivery failed to delete the retry
182 item, thus causing premature timeout of the address. The bug is now
185 PH/20 Added hosts_avoid_pipelining to the smtp transport.
187 PH/21 Long custom messages for fakedefer and fakereject are now split up
188 into multiline reponses in the same way that messages for "deny" and
189 other ACL rejections are.
191 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
192 with slight modification.
194 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
195 draft, changing the syntax and factoring some duplicate code".
197 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
198 for deliveries of the second and subsequent messages over the same SMTP
201 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
202 ${reduce, with only minor "tidies".
204 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
206 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
207 expansion side effects.
209 PH/27 When a message times out after an over-quota error from an Exim-imposed
210 quota, the bounce message says "mailbox is full". This message was not
211 being given when it was a system quota that was exceeded. It now should
214 MH/03 Made $recipients available in local_scan(). local_scan() already has
215 better access to the recipient list through recipients_list[], but
216 $recipients can be useful in postmaster-provided expansion strings.
218 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
219 in the case of a MAIL command with additional options following the
220 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
221 were accidentally chopped off.
223 PH/29 SMTP synchronization checks are implemented when a command is read -
224 there is a check that no more input is waiting when there shouldn't be
225 any. However, for some commands, a delay in an ACL can mean that it is
226 some time before the response is written. In this time, more input might
227 arrive, invalidly. So now there are extra checks after an ACL has run for
228 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
229 pipelining has not been advertised.
231 PH/30 MH's patch to allow iscntrl() characters to be list separators.
233 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
234 returned in the SMTP response when smtp_return_error_details was false.
237 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
238 instead of the C I/O that was originally supplied, because problems were
241 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
242 Exim which did not show up earlier: it was assuming that a call to
243 SSL_CTX_set_info_callback() might give an error value. In fact, there is
244 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
245 was a macro that became an assignment, so it seemed to work. This has
246 changed to a proper function call with a void return, hence the compile
247 error. Exim's code has been fixed.
249 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
252 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
254 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
256 PH/37 If a message is not accepted after it has had an id assigned (e.g.
257 because it turns out to be too big or there is a timeout) there is no
258 "Completed" line in the log. When some messages of this type were
259 selected by exigrep, they were listed as "not completed". Others were
260 picked up by some special patterns. I have improved the selection
261 criteria to be more general.
263 PH/38 The host_find_failed option in the manualroute router can now be set
264 to "ignore", to completely ignore a host whose IP address cannot be
265 found. If all hosts are ignored, the behaviour is controlled by the new
266 host_all_ignored option.
268 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
269 homing or because of multiple MX records with /mx) generated more than
270 one IP address, and the following item turned out to be the local host,
271 all the secondary addresses of the first item were incorrectly removed
272 from the list, along with the local host and any following hosts (which
273 is what is supposed to happen).
275 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
276 whoever called Exim into the -H file. In the case of the daemon it was
277 behaving confusingly. When first started, it used values for whoever
278 started the daemon, but after a SIGHUP it used the Exim user (because it
279 calls itself on a restart). I have changed the code so that it now always
282 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
283 message are rejected with the same error (e.g. no authentication or bad
284 sender address), and a DATA command is nevertheless sent (as can happen
285 with PIPELINING or a stupid MUA), the error message that was given to the
286 RCPT commands is included in the rejection of the DATA command. This is
287 intended to be helpful for MUAs that show only the final error to their
290 PH/42 Another patch from the Sieve maintainer.
292 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
293 Eximstats - Fixed some broken HTML links and added missing column headers
295 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
296 columns for Rejects, Temp Rejects, Ham, and Spam rows.
298 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
300 PH/43 Yet another patch from the Sieve maintainer.
302 PH/44 I found a way to check for a TCP/IP connection going away before sending
303 the response to the final '.' that terminates a message, but only in the
304 case where the client has not sent further data following the '.'
305 (unfortunately, this is allowed). However, in many cases there won't be
306 any further data because there won't be any more messages to send. A call
307 to select() can be used: if it shows that the input is "ready", there is
308 either input waiting, or the socket has been closed. An attempt to read
309 the next input character can distinguish the two cases. Previously, Exim
310 would have sent an OK response which the client would never have see.
311 This could lead to message repetition. This fix should cure that, at
312 least in a lot of common cases.
314 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
315 advertised in response to EHLO.
321 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
322 fixed by 4.65/MH/01 (is this a record?) are fixed:
324 (i) An empty string was always treated as zero by the numeric comparison
325 operators. This behaviour has been restored.
327 (ii) It is documented that the numeric comparison operators always treat
328 their arguments as decimal numbers. This was broken in that numbers
329 starting with 0 were being interpreted as octal.
331 While fixing these problems I realized that there was another issue that
332 hadn't been noticed. Values of message_size_limit (both the global option
333 and the transport option) were treated as octal if they started with 0.
334 The documentation was vague. These values are now always treated as
335 decimal, and I will make that clear in the documentation.
341 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
342 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
345 MH/01 Don't check that the operands of numeric comparison operators are
346 integers when their expansion is in "skipping" mode (fixes bug
347 introduced by 4.64-PH/07).
349 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
350 child addresses, Exim now panics and dies. Previously, because the count
351 is held in a short int, deliveries were likely to be lost. As such a
352 large number of recipients for a single message is ridiculous
353 (performance will be very, very poor), I have chosen to impose a limit
354 rather than extend the field.
360 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
361 leftover -K file (the existence of which was triggered by #402).
362 While we were at it, introduced process PID as part of the -K
363 filename. This should rule out race conditions when creating
366 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
367 processing considerably. Previous code took too long for large mails,
368 triggering a timeout which in turn triggers #401.
370 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
371 in the DK code in transports.c. sendfile() is not really portable,
372 hence the _LINUX specificness.
374 TF/01 In the add_headers option to the mail command in an Exim filter,
375 there was a bug that Exim would claim a syntax error in any
376 header after the first one which had an odd number of characters
379 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
380 callout verification, Exim cached a "reject" for the entire domain. This
381 is correct for most verifications, but it is not correct for a recipient
382 verification with use_sender or use_postmaster set, because in that case
383 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
384 case of MAIL FROM:<> rejection from other early rejections (e.g.
385 rejection of HELO). When verifying a recipient using a non-null MAIL
386 address, the cache is ignored if it shows MAIL FROM:<> rejection.
387 Whatever the result of the callout, the value of the domain cache is
388 left unchanged (for any other kind of callout, getting as far as trying
389 RCPT means that the domain itself is ok).
391 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
394 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
395 manifest itself as EPIPE rather than ECONNECT. When tidying away a
396 session, the daemon ignores ECONNECT errors and logs others; it now
397 ignores EPIPE as well.
399 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
400 (quoted-printable decoding).
402 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
403 later the small subsequent patch to fix an introduced bug.
405 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
407 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
409 PH/08 An error is now given if message_size_limit is specified negative.
411 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
412 to be given (somewhat) arbitrary names.
414 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
417 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
418 miscellaneous code fixes
420 PH/10 Added the log_reject_target ACL modifier to specify where to log
423 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
424 hostname. This is wrong, because it relates to the incoming message (and
425 probably the interface on which it is arriving) and not to the outgoing
426 callout (which could be using a different interface). This has been
427 changed to use the value of the helo_data option from the smtp transport
428 instead - this is what is used when a message is actually being sent. If
429 there is no remote transport (possible with a router that sets up host
430 addresses), $smtp_active_hostname is used.
432 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
433 tweaks were necessary in order to get it to work (see also 21 below):
434 (a) The code assumed that strncpy() returns a negative number on buffer
435 overflow, which isn't the case. Replaced with Exim's string_format()
437 (b) There were several signed/unsigned issues. I just did the minimum
438 hacking in of casts. There is scope for a larger refactoring.
439 (c) The code used strcasecmp() which is not a standard C function.
440 Replaced with Exim's strcmpic() function.
441 (d) The code set only $1; it now sets $auth1 as well.
442 (e) A simple test gave the error "authentication client didn't specify
443 service in request". It would seem that Dovecot has changed its
444 interface. Fortunately there's a specification; I followed it and
445 changed what the client sends and it appears to be working now.
447 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
450 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
451 address A is aliased to B and C, where B exists and C does not. Without
452 -v the output is "A verified" because verification stops after a
453 successful redirection if more than one address is generated. However,
454 with -v the child addresses are also verified. Exim was outputting "A
455 failed to verify" and then showing the successful verification for C,
456 with its parentage. It now outputs "B failed to verify", showing B's
457 parentage before showing the successful verification of C.
459 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
460 look up a TXT record in a specific list after matching in a combined
463 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
464 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
465 they consult the DNS. I had assumed they would set it the way they
466 wanted; and indeed my experiments on Linux seem to show that in some
467 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
468 To be on the safe side, however, I have now made the interface to
469 host_find_byname() similar to host_find_bydns(), with an argument
470 containing the DNS resolver options. The host_find_byname() function now
471 sets these options at its start, just as host_find_bydns() does. The smtp
472 transport options dns_qualify_single and dns_search_parents are passed to
473 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
474 of host_find_byname() use the default settings of RES_DEFNAMES
475 (qualify_single) but not RES_DNSRCH (search_parents).
477 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
478 spool_read_header() do less string testing, by means of a preliminary
479 switch on the second character of optional "-foo" lines. (This is
480 overdue, caused by the large number of possibilities that now exist.
481 Originally there were few.) While I was there, I also converted the
482 str(n)cmp tests so they don't re-test the leading "-" and the first
483 character, in the hope this might squeeze out yet more improvement.
485 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
486 flag allowing group syntax was set by the header_syntax check but not
487 turned off, possible causing trouble later; (2) The flag was not being
488 set at all for the header_verify test, causing "group"-style headers to
489 be rejected. I have now set it in this case, and also caused header_
490 verify to ignore an empty address taken from a group. While doing this, I
491 came across some other cases where the code for allowing group syntax
492 while scanning a header line wasn't quite right (mostly, not resetting
493 the flag correctly in the right place). These bugs could have caused
494 trouble for malformed header lines. I hope it is now all correct.
496 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
497 with the "reply" argument non-NULL. The code, however (which originally
498 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
499 but it didn't always do it. This confused somebody who was copying the
500 code for some other use. I have removed all the tests.
502 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
503 feature that was used to support insecure browsers during the U.S. crypto
504 embargo. It requires special client support, and Exim is probably the
505 only MTA that supported it -- and would never use it because real RSA is
506 always available. This code has been removed, because it had the bad
507 effect of slowing Exim down by computing (never used) parameters for the
508 RSA_EXPORT functionality.
510 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
511 authenticator to fail if there's a tab character in the incoming data
512 (there should never be unless someone is messing about, as it's supposed
513 to be base64-encoded). Also added, on Timo's advice, the "secured" option
514 if the connection is using TLS or if the remote IP is the same as the
515 local IP, and the "valid-client-cert option" if a client certificate has
518 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
519 authenticators. This can be used for authorization after authentication
520 succeeds. (In the case of plaintext, it servers for both authentication
523 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
524 if any retry times were supplied.
526 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
527 connection, where there is no client IP address to check. In this
528 situation, the verify now always succeeds.
530 PH/25 Applied John Jetmore's -Mset patch.
532 PH/26 Added -bem to be like -Mset, but loading a message from a file.
534 PH/27 In a string expansion for a processed (not raw) header when multiple
535 headers of the same name were present, leading whitespace was being
536 removed from all of them, but trailing whitespace was being removed only
537 from the last one. Now trailing whitespace is removed from each header
538 before concatenation. Completely empty headers in a concatenation (as
541 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
542 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
544 PH/29 [Removed. This was a change that I later backed out, and forgot to
545 correct the ChangeLog entry (that I had efficiently created) before
546 committing the later change.]
548 PH/30 Exim was sometimes attempting to deliver messages that had suffered
549 address errors (4xx response to RCPT) over the same connection as other
550 messages routed to the same hosts. Such deliveries are always "forced",
551 so retry times are not inspected. This resulted in far too many retries
552 for the affected addresses. The effect occurred only when there were more
553 hosts than the hosts_max_try setting in the smtp transport when it had
554 the 4xx errors. Those hosts that it had tried were not added to the list
555 of hosts for which the message was waiting, so if all were tried, there
556 was no problem. Two fixes have been applied:
558 (i) If there are any address or message errors in an SMTP delivery, none
559 of the hosts (tried or untried) are now added to the list of hosts
560 for which the message is waiting, so the message should not be a
561 candidate for sending over the same connection that was used for a
562 successful delivery of some other message. This seems entirely
563 reasonable: after all the message is NOT "waiting for some host".
564 This is so "obvious" that I'm not sure why it wasn't done
565 previously. Hope I haven't missed anything, but it can't do any
566 harm, as the worst effect is to miss an optimization.
568 (ii) If, despite (i), such a delivery is accidentally attempted, the
569 routing retry time is respected, so at least it doesn't keep
570 hammering the server.
572 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
573 in ${readsocket because some servers need this prod.
575 PH/32 Added some extra debug output when updating a wait-xxx database.
577 PH/33 The hint "could be header name not terminated by colon", which has been
578 given for certain expansion errors for a long time, was not being given
579 for the ${if def:h_colon_omitted{... case.
581 PH/34 The spec says: "With one important exception, whenever a domain list is
582 being scanned, $domain contains the subject domain." There was at least
583 one case where this was not true.
585 PH/35 The error "getsockname() failed: connection reset by peer" was being
586 written to the panic log as well as the main log, but it isn't really
587 panic-worthy as it just means the connection died rather early on. I have
588 removed the panic log writing for the ECONNRESET error when getsockname()
591 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
592 runs only) independently of the message's sender address. This meant
593 that, if the 4xx error was in fact related to the sender, a different
594 message to the same recipient with a different sender could confuse
595 things. In particualar, this can happen when sending to a greylisting
596 server, but other circumstances could also provoke similar problems.
597 I have changed the default so that the retry time for these errors is now
598 based a combination of the sender and recipient addresses. This change
599 can be overridden by setting address_retry_include_sender=false in the
602 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
603 remote server are returned as part of bounce messages. This was not
604 happening for LMTP over a pipe (the lmtp transport), but now it is the
605 same for both kinds of LMTP.
607 PH/38 Despite being documented as not happening, Exim was rewriting addresses
608 in header lines that were in fact CNAMEs. This is no longer the case.
610 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
611 and queue runs started by the daemon processed all messages. This has
612 been fixed so that -R and -S can now usefully be given with -q<time>.
614 PH/40 Import PCRE release 6.7 (fixes some bugs).
616 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
618 PH/42 Give an error if -q is specified more than once.
620 PH/43 Renamed the variables $interface_address and $interface_port as
621 $received_ip_address and $received_port, to make it clear that these
622 values apply to message reception, and not to the outgoing interface when
623 a message is delivered. (The old names remain recognized, of course.)
625 PH/44 There was no timeout on the connect() call when using a Unix domain
626 socket in the ${readsocket expansion. There now is.
628 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
629 be meaningful with "accept".
631 SC/01 Eximstats V1.43
632 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
634 SC/02 Eximstats V1.44
635 Use a glob alias rather than an array ref in the generated
636 parser. This improves both readability and performance.
638 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
639 Collect SpamAssassin and rejection statistics.
640 Don't display local sender or destination tables unless
641 there is data to show.
642 Added average volumes into the top table text output.
644 SC/04 Eximstats V1.46
645 Collect data on the number of addresses (recipients)
646 as well as the number of messages.
648 SC/05 Eximstats V1.47
649 Added 'Message too big' to the list of mail rejection
650 reasons (thanks to Marco Gaiarin).
652 SC/06 Eximstats V1.48
653 Mainlog lines which have GMT offsets and are too short to
654 have a flag are now skipped.
656 SC/07 Eximstats V1.49 (Alain Williams)
657 Added the -emptyok flag.
659 SC/08 Eximstats V1.50
660 Fixes for obtaining the IP address from reject messages.
662 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
663 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
664 whitesspace changes from 4.64-PH/27
666 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
669 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
670 are found, allow negative numbers in numeric criteria)
672 JJ/06 exipick.20061117.2, added new $message_body_missing variable
674 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
675 to match changes made in 4.64-PH/43
677 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
679 PH/47 Put in an explicit test for a DNS lookup of an address record where the
680 "domain" is actually an IP address, and force a failure. This locks out
681 those revolvers/nameservers that support "A-for-A" lookups, in
682 contravention of the specifications.
684 PH/48 When a host name was looked up from an IP address, and the subsequent
685 forward lookup of the name timed out, the host name was left in
686 $sender_host_name, contrary to the specification.
688 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
689 restricted to single-key lookups, Exim was not diagnosing an error if
690 * or *@ was used with a query-style lookup.
692 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
694 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
695 long ago, but noone interested enough thought of it. Let's just say that
696 the "1.1" means that there are some new functions that weren't there at
697 some point in the past.
699 PH/51 Error processing for expansion failure of helo_data from an smtp
700 transport during callout processing was broken.
702 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
703 tested/used via the -bh/-bhc/-bs options.
705 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
706 bug, fixed in subsequent PCRE releases).
708 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
709 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
711 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
717 SC/01 Use a glob alias rather than an array ref in eximstats generated
718 parser. This improves both readability and performance.
720 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
721 Don't display local sender or destination tables in eximstats unless
722 there is data to show.
723 Added average volumes into the eximstats top table text output.
725 SC/03 Collect data on the number of addresses (recipients) as well
726 as the number of messages in eximstats.
728 TF/01 Correct an error in the documentation for the redirect router. Exim
729 does (usually) call initgroups() when daemonizing.
731 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
732 with consistent privilege compared to when running as a daemon.
734 TF/03 Note in the spec that $authenticated_id is not set for local
735 submissions from trusted users.
737 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
738 Thanks to Dean Brooks <dean@iglou.com> for the patch.
740 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
741 by adding some example configuration directives to the default
742 configuration file. A little bit of work is required to uncomment the
743 directives and define how usernames and passwords are checked, but
744 there is now a framework to start from.
746 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
747 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
748 without this. I don't know how relevant this is to other LDAP libraries.
750 PH/02 Add the verb name to the "unknown ACL verb" error.
752 PH/03 Magnus Holmgren's patch for filter_prepend_home.
754 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
756 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
757 directory not expanded when it should be if an expanded home directory
758 was set for the address (which is overridden by the transport).
760 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
763 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
764 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
765 because it is too late at that time, and has no effect.
767 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
768 security issue with \' (bugzilla #107). I could not use the
769 PQescapeStringConn() function, because it needs a PGconn value as one of
772 PH/08 When testing addresses using -bt, indicate those final addresses that
773 are duplicates that would not cause an additional delivery. At least one
774 person was confused, thinking that -bt output corresponded to deliveries.
775 (Suppressing duplicates isn't a good idea as you lose the information
776 about possibly different redirections that led to the duplicates.)
778 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
779 systems where poll() doesn't work, in particular OS X.
781 PH/10 Added more information to debugging output for retry time not reached.
783 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
784 operations in malware.c.
786 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
789 PH/13 If write_rejectlog was set false when logging was sent to syslog with
790 syslog_duplication set false, log lines that would normally be written
791 both the the main log and to the reject log were not written to syslog at
794 PH/14 In the default configuration, change the use of "message" in ACL warn
795 statements to "add_header".
797 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
798 not followed by a command (e.g. "seen endif").
800 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
801 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
804 PH/17 Added extra conditions to the default value of delay_warning_condition
808 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
809 { match{$h_precedence:}{(?i)bulk|list|junk} } \
810 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
813 The Auto-Submitted: and various List- headers are standardised, whereas I
814 don't think Precedence: ever was.
816 PH/18 Refactored debugging code in route_finduser() to show more information,
817 in particular, the error code if getpwnam() issues one.
819 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
820 This is apparently needed in addition to the PH/07 change above to avoid
821 any possible encoding problems.
823 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
824 but not after initializing Perl.
826 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
827 output them only if debugging. By default they are written stderr,
828 apparently, which is not desirable.
830 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
833 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
836 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
838 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
839 authenticated or an ident call has been made. Suppress the default
840 values for $authenticated_id and $authenticated_sender (but permit -oMai
841 and -oMas) when testing with -bh.
843 PH/24 Re-jigged the order of the tests in the default configuration so that the
844 tests for valid domains and recipients precede the DNS black list and CSA
845 tests, on the grounds that those ones are more expensive.
847 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
848 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
849 This bug exists in every version of Exim that I still have, right back to
852 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
853 However, an attempt to turn on case-sensitivity in a regex key by
854 including (?-i) didn't work because the subject string was already
855 lowercased, and the effects were non-intuitive. It turns out that a
856 one-line patch can be used to allow (?-i) to work as expected.
862 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
863 other effects) broke the use of negated acl sub-conditions.
865 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
868 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
869 "Deny" causes Exim to reject the incoming connection with a 554 error.
870 Unfortunately, if there is a major crisis, such as a disk failure,
871 tcp-wrappers gives "deny", whereas what one would like would be some
872 kind of temporary error. A kludge has been added to help with this.
873 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
874 554 error is used if errno is still zero or contains ENOENT (which occurs
875 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
878 PH/03 Add -lutil to the default FreeBSD LIBS setting.
880 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
881 errors. Otherwise a message that provokes a temporary error (when other
882 messages do not) can cause a whole host to time out.
884 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
885 addresses were routed directly to files or pipes from a redirect router.
886 File deliveries just didn't batch; pipe deliveries might have suffered
889 PH/06 A failure to get a lock for a hints database would erroneously always say
890 "Failed to get write lock", even when it was really a read lock.
892 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
893 of 0600. This has been changed to use the value of the lockfile_mode
894 option (which defaults to 0600).
896 PH/08 Applied small patch from the Sieve maintainer.
898 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
899 folder from quota calculations, a direct delivery into this folder messed
900 up the contents of the maildirsize file. This was because the regex was
901 used only to exclude .Trash (or whatever) when the size of the mailbox
902 was calculated. There was no check that a delivery was happening into an
903 excluded directory. This bug has been fixed by ignoring all quota
904 processing for deliveries into excluded directories.
906 PH/10 Added the maildirfolder_create_regex option to appendfile.
912 PH/01 The code for finding all the local interface addresses on a FreeBSD
913 system running IPv6 was broken. This may well have applied to all BSD
914 systems, as well as to others that have similar system calls. The broken
915 code found IPv4 interfaces correctly, but gave incorrect values for the
916 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
917 that it would not match correctly against @[] and not recognize the IPv6
920 PH/02 The ipliteral router was not recognizing addresses of the form user@
921 [ipv6:....] because it didn't know about the "ipv6:" prefix.
923 PH/03 Added disable_ipv6.
925 PH/04 Changed $reply_address to use the raw form of the headers instead of the
926 decoded form, because it is most often used to construct To: headers
927 lines in autoreplies, and the decoded form may well be syntactically
928 invalid. However, $reply_address has leading white space removed, and all
929 newlines turned into spaces so that the autoreply transport does not
932 PH/05 If group was specified without a user on a router, and no group or user
933 was specified on a transport, the group from the router was ignored.
935 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
936 for visible compile-time settings that can be used to change these
937 numbers, for those that want even more. Backwards compatibility with old
938 spool files has been maintained. However, going back to a previous Exim
939 release will lost any variables that are in spool files.
941 PH/07 Two small changes when running in the test harness: increase delay when
942 passing a TCP/IP connection to a new process, in case the original
943 process has to generate a bounce, and remove special handling of
944 127.0.0.2 (sic), which is no longer necessary.
946 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
947 be the same on different OS.
949 PH/09 Moved a debug statement in filter processing to avoid a race problem when
952 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
953 whether --show-vars was specified or not
955 JJ/02 exipick: Added support for new ACL variable spool format introduced
958 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
959 syntactically invalid From: or Reply-to: line, and a filter used this to
960 generate an autoreply, and therefore failed to obtain an address for the
961 autoreply, Exim could try to deliver to a non-existent relative file
962 name, causing unrelated and misleading errors. What now happens is that
963 it logs this as a hard delivery error, but does not attempt to create a
966 PH/11 The exinext utility has a -C option for testing purposes, but although
967 the given file was scanned by exinext itself; it wasn't being passed on
968 when Exim was called.
970 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
971 an end-of-file indication when reading a command response.
973 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
974 compiled. In many other places in Exim, IPv6 addresses are always
975 recognized, so I have changed this. It also means that IPv4 domain
976 literals of the form [IPV4:n.n.n.n] are now always recognized.
978 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
979 used if the router is not running as root, for example, when verifying at
980 ACL time, or when using -bh. The debugging output from this situation was
981 non-existent - all you got was a failure to exec. I have made two
984 (a) Failures to set uid/gid, the current directory, or a process leader
985 in a subprocess such as that created by queryprogram now generate
986 suitable debugging ouput when -d is set.
988 (b) The queryprogram router detects when it is not running as root,
989 outputs suitable debugging information if -d is set, and then runs
990 the subprocess without attempting to change uid/gid.
992 PH/15 Minor change to Makefile for building test_host (undocumented testing
995 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
996 additional section of a DNS packet that returns MX or SRV records.
997 Instead, it always explicitly searches for A/AAAA records. This avoids
998 major problems that occur when a DNS server includes only records of one
999 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1000 fixed another bug: if SRV records were looked up and the corresponding
1001 address records were *not* found in the additional section, the port
1002 values from the SRV records were lost.
1004 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1005 using the correct key (the original address) when searching the retry
1006 rules in order to find which one to use for generating the retry hint.
1008 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1009 adding the default one. Similarly, if it contains a Reply-To: header, the
1010 errors_reply_to option, if set, is not used.
1012 PH/19 When calculating a retry time, Exim used to measure the "time since
1013 failure" by looking at the "first failed" field in the retry record. Now
1014 it does not use this if it is later than than the arrival time of the
1015 message. Instead it uses the arrival time. This makes for better
1016 behaviour in cases where some deliveries succeed, thus re-setting the
1017 "first failed" field. An example is a quota failure for a huge message
1018 when small messages continue to be delivered. Without this change, the
1019 "time since failure" will always be short, possible causing more frequent
1020 delivery attempts for the huge message than are intended.
1021 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1023 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1024 $1, $2, $3) because the numerical variables can be reset during some
1025 expansion items (e.g. "match"), thereby losing the authentication data.
1027 PH/21 Make -bV show the size of off_t variables so that the test suite can
1028 decide whether to run tests for quotas > 2G.
1030 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1031 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1032 filecount value is greater than 2G or if a quota value is greater than 2G
1033 on a system where the size of off_t is not greater than 4, a panic error
1036 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1037 never match. The debug and -bh output now contains an explicit error
1038 message indicating a malformed IPv4 address or mask.
1040 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1041 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1042 PH/23 above applies.
1044 PH/25 Do not write to syslog when running in the test harness. The only
1045 occasion when this arises is a failure to open the main or panic logs
1046 (for which there is an explicit test).
1048 PH/26 Added the /no_tell option to "control=freeze".
1050 PH/27 If a host name lookup failed very early in a connection, for example, if
1051 the IP address matched host_lookup and the reverse lookup yielded a name
1052 that did not have a forward lookup, an error message of the form "no IP
1053 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1054 could be logged. Now it outputs the IP address instead of "NULL".
1056 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1057 allows the sender and the authenticated sender to be set when
1058 submitting a message from within Exim. Since child_open_exim() is
1059 documented for local_scan(), the new function should be too.
1061 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1062 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1063 results in an empty string is now treated as unset.
1065 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1067 PH/31 Added sender_verify_fail logging option.
1069 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1070 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1071 tidied the source and removed it altogether.
1073 PH/33 When a queue run was abandoned because the load average was too high, a
1074 log line was always written; now it is written only if the queue_run log
1075 selector is set. In addition, the log line for abandonment now contains
1076 information about the queue run such as the pid. This is always present
1077 in "start" and "stop" lines but was omitted from the "abandon" line.
1079 PH/34 Omit spaces between a header name and the colon in the error message that
1080 is given when verify = headers_syntax fails (if there are lots of them,
1081 the message gets confusing).
1083 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1084 names, as there are now some PTR records that contain slashes. This check
1085 is only to protect against broken name servers that fall over on strange
1086 characters, so the fact that it applies to all lookups doesn't matter.
1088 PH/36 Now that the new test suite is complete, we can remove some of the
1089 special code in Exim that was needed for the old test suite. For example,
1090 sorting DNS records because real resolvers return them in an arbitrary
1091 order. The new test suite's fake resolver always returns records in the
1094 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1095 bounces) except when queue_only is set, to avoid logging races between
1096 the different processes.
1098 PH/38 Panic-die if .include specifies a non-absolute path.
1100 PH/39 A tweak to the "H" retry rule from its user.
1102 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1103 a label. They prevented compilation on older perls.
1105 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1106 a warning to be raised on newish perls.
1108 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1109 on queue. Changes to match documented behaviour of showing count of
1110 messages matching specified criteria.
1112 PH/40 Changed the default ident timeout from 30s to 5s.
1114 PH/41 Added support for the use of login_cap features, on those BSD systems
1115 that have them, for controlling the resources used by pipe deliveries.
1117 PH/42 The content-scanning code uses fopen() to create files in which to put
1118 message data. Previously it was not paying any attention to the mode of
1119 the files. Exim runs with umask(0) because the rest of the code creates
1120 files with open(), and sets the required mode explicitly. Thus, these
1121 files were ending up world-writeable. This was not a big issue, because,
1122 being within the spool directory, they were not world-accessible. I have
1123 created a function called modefopen, which takes an additional mode
1124 argument. It sets umask(777), creates the file, chmods it to the required
1125 mode, then resets the umask. All the relevant calls to fopen() in the
1126 content scanning code have been changed to use this function.
1128 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1129 to 24 hours. This avoids potential overflow problems when processing G
1130 and H retry rules. I suspect nobody ever tinkers with this value.
1132 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1134 PH/45 When the plaintext authenticator is running as a client, the server's
1135 challenges are checked to ensure they are valid base64 strings. By
1136 default, the authentication attempt is cancelled if an invalid string is
1137 received. Setting client_ignore_invalid_base64 true ignores these errors.
1138 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1139 they are received. Thus, the responses can be made to depend on the
1140 challenges. If an invalid string is ignored, an empty string is placed in
1143 PH/46 Messages that are created by the autoreply transport now contains a
1144 References: header, in accordance with RFCs 2822 and 3834.
1146 PH/47 Added authenticated_sender_force to the smtp transport.
1148 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1150 PH/49 Installed latest patch from the Sieve maintainer.
1152 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1153 was also set, the appendfile transport was unnecessarily scanning a
1154 directory of message files (e.g. for maildir delivery) to find the count
1155 of files (along with the size), even though it did not need this
1156 information. It now does the scan only if it needs to find either the
1157 size of the count of files.
1159 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1161 PH/52 Two bugs concerned with error handling when the smtp transport is
1164 (i) Exim was not creating retry information for temporary errors given
1165 for individual recipients after the DATA command when the smtp transport
1166 was used in LMTP mode. This meant that they could be retried too
1167 frequently, and not timed out correctly.
1169 (ii) Exim was setting the flag that allows error details to be returned
1170 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1171 recipients that were returned after the DATA command.
1173 PH/53 This is related to PH/52, but is more general: for any failing address,
1174 when detailed error information was permitted to be returned to the
1175 sender, but the error was temporary, then after the final timeout, only
1176 "retry timeout exceeded" was returned. Now it returns the full error as
1177 well as "retry timeout exceeded".
1179 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1180 do this, and (what is worse) MTAs that accept it.
1182 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1183 will now be deprecated.
1185 PH/56 New os.c-cygwin from the Cygwin maintainer.
1187 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1188 formats (previously only available in exim formats via -bpr, -bpru,
1189 and -bpra. Now also available in native and exiqgrep formats)
1191 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1192 with very large, slow to parse queues
1194 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1196 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1198 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1199 responses to authentication challenges, though it was showing the
1200 challenges; (ii) I've removed the CR characters from the debug output for
1203 PH/58 Allow for the insertion of a newline as well as a space when a string
1204 is turned into more than one encoded-word during RFC 2047 encoding. The
1205 Sieve code now uses this.
1207 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1208 data_4xx, lost_connection, tls_required.
1210 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1211 message was being sent as an SMTP response.
1213 PH/61 Add -l and -k options to exicyclog.
1215 PH/62 When verifying, if an address was redirected to one new address, so that
1216 verification continued, and the new address failed or deferred after
1217 having set something in $address_data, the value of $address_data was not
1218 passed back to the ACL. This was different to the case when no
1219 redirection occurred. The value is now passed back in both cases.
1221 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1222 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1223 use login_cap.h, so on its own it isn't the distinguishing feature. The
1224 new name refers directly to the setclassresources() function.
1226 PH/65 Added configuration files for NetBSD3.
1228 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1230 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1231 is preferred over IPv4.
1233 PH/68 The bounce_return_message and bounce_return_body options were not being
1234 honoured for bounces generated during the reception of non-SMTP messages.
1235 In particular, this applied to messages rejected by the ACL. This bug has
1236 been fixed. However, if bounce_return_message is true and bounce_return_
1237 body is false, the headers that are returned for a non-SMTP message
1238 include only those that have been read before the error was detected.
1239 (In the case of an ACL rejection, they have all been read.)
1241 PH/69 The HTML version of the specification is now built in a directory called
1242 spec_html instead of spec.html, because the latter looks like a path with
1243 a MIME-type, and this confuses some software.
1245 PH/70 Catch two compiler warnings in sieve.c.
1247 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1248 function verify_get_ident() calls ip_connect() to connect a socket, but
1249 if the "connect()" function timed out, ip_connect() used to close the
1250 socket. However, verify_get_ident() also closes the socket later, and in
1251 between Exim writes to the log, which may get opened at this point. When
1252 the socket was closed in ip_connect(), the log could get the same file
1253 descriptor number as the socket. This naturally causes chaos. The fix is
1254 not to close the socket in ip_connect(); the socket should be closed by
1255 the function that creates it. There was only one place in the code where
1256 this was missing, in the iplookup router, which I don't think anybody now
1257 uses, but I've fixed it anyway.
1259 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1260 well as to direct DNS lookups. Otherwise the handling of names in host
1261 lists is inconsistent and therefore confusing.
1267 PH/01 Two changes to the default runtime configuration:
1269 (1) Move the checks for relay_from_hosts and authenticated clients from
1270 after to before the (commented out) DNS black list checks.
1272 (2) Add control=submission to the relay_from_hosts and authenticated
1273 clients checks, on the grounds that messages accepted by these
1274 statements are most likely to be submissions.
1276 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1278 (1) Generate an error if the third argument for the ${prvs expansion is
1281 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1284 (3) Reset the variables that are obtained from the first argument of
1285 ${prvscheck and used in the second argument before leaving the code,
1286 because their memory is reclaimed, so using them afterwards may do
1289 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1290 one (it's much easier than Tom thought :-).
1292 (5) Because of (4), we can now allow for the use of $prvscheck_result
1293 inside the third argument.
1295 PH/03 For some reason, the default setting of PATH when running a command from
1296 a pipe transport was just "/usr/bin". I have changed it to
1299 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1300 anything to be listed in the output from -bV.
1302 PH/05 When a filter generated an autoreply, the entire To: header line was
1303 quoted in the delivery log line, like this:
1305 => >A.N.Other <ano@some.domain> <original@ddress> ...
1307 This has been changed so that it extracts the operative address. There
1308 may be more than one such address. If so, they are comma-separated, like
1311 => >ano@some.domain,ona@other.domain <original@ddress> ...
1313 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1314 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1315 not being looked up in the rDNS to get a host name, Exim was showing the
1316 IP address twice in Received: lines, even though the IP addresses were
1317 identical. For example:
1319 Received: from [1.2.3.4] (helo=[1.2.3.4])
1321 However, if the real host name was known, it was omitting the HELO data
1322 if it matched the actual IP address. This has been tidied up so that it
1323 doesn't show the same IP address twice.
1325 PH/07 When both +timestamp and +memory debugging was on, the value given by
1326 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1327 called by the debug printing, thereby overwriting the timestamp buffer.
1328 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1330 PH/08 When the original message was included in an autoreply transport, it
1331 always said "this is a copy of the message, including all the headers",
1332 even if body_only or headers_only was set. It now gives an appropriate
1335 PH/09 Applied a patch from the Sieve maintainer which:
1337 o fixes some comments
1338 o adds the (disabled) notify extension core
1339 o adds some debug output for the result of if/elsif tests
1340 o points to the current vacation draft in the documentation
1341 and documents the missing references header update
1345 o fixes a bug in processing the envelope test (when testing
1346 multiple envelope elements, the last element determinted the
1349 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1350 Electronic Mail") by including:
1352 Auto-submitted: auto-generated
1354 in the messages that it generates (bounce messages and others, such as
1355 warnings). In the case of bounce messages for non-SMTP mesages, there was
1356 also a typo: it was using "Auto_submitted" (underscore instead of
1357 hyphen). Since every message generated by Exim is necessarily in response
1358 to another message, thes have all been changed to:
1360 Auto-Submitted: auto-replied
1362 in accordance with these statements in the RFC:
1364 The auto-replied keyword:
1366 - SHOULD be used on messages sent in direct response to another
1367 message by an automatic process,
1369 - MUST NOT be used on manually-generated messages,
1371 - MAY be used on Delivery Status Notifications (DSNs) and Message
1372 Disposition Notifications (MDNs),
1374 - MUST NOT be used on messages generated by automatic or periodic
1375 processes, except for messages which are automatic responses to
1378 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1379 to the default Received: header definition.
1381 PH/12 Added log selector acl_warn_skipped (default on).
1383 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1384 variables because (a) they are in the wrong storage pool and (b) even if
1385 they were copied, it wouldn't work properly because of the caching.
1387 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1388 checking when decoding. Apparently there are clients that generate
1389 overlong encoded strings. Why am I not surprised?
1391 PH/15 If the first argument of "${if match_address" was not empty, but did not
1392 contain an "@" character, Exim crashed. Now it writes a panic log message
1393 and treats the condition as false.
1395 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1397 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1398 extension "envelope-auth". The code is finished and in agreement with
1399 other implementations, but there is no documentation so far and in fact,
1400 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1401 not changing the active code.
1403 Print executed "if" and "elsif" statements when debugging is used. This
1404 helps a great deal to understand what a filter does.
1406 Document more things not specified clearly in RFC3028. I had all this
1407 sorted out, when out of a sudden new issues came to my mind. Oops."
1409 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1412 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1413 canonical form (as documented). However, after a host name lookup from
1414 the IP address, check_host() was doing a simple string comparison with
1415 addresses acquired from the DNS when checking that the found name did
1416 have the original IP as one of its addresses. Since any found IPv6
1417 addresses are likely to be in abbreviated form, the comparison could
1418 fail. Luckily, there already exists a function for doing the comparison
1419 by converting both addresses to binary, so now that is used instead of
1420 the text comparison.
1422 PH/20 There was another similar case to PH/19, when a complete host name was
1423 given in a host list; looking up its IP address could give an abbreviated
1424 form, whereas the current host's name might or might not be abbreviated.
1425 The same fix has been applied.
1431 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1432 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1435 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1436 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1438 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1440 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1441 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1442 Also, for "Auto-Submitted", treat anything other than "no" as
1443 non-personal, in accordance with RFC 3834. (Previously it treated
1444 anything starting "auto-" as non-personal.)
1446 TF/01 The control=submission/name=... option had a problem with syntax
1447 errors if the name included a slash character. The /name= option
1448 now slurps the rest of the string, so it can include any characters
1449 but it must come last in the list of options (after /sender_retain
1452 PH/05 Some modifications to the interface to the fake nameserver for the new
1460 TK/01 Added the "success_on_redirect" address verification option. See
1461 NewStuff for rationale and an example.
1463 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1465 PH/02 Patch to exigrep to allow it to work on syslog lines.
1467 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1468 fread() to skip over the body file's header line, because in Cygwin the
1469 header line is locked and is inaccessible.
1471 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1472 co-exist for some time) to make it clear that it is the Exim ID that is
1473 referenced, not the Message-ID: header line.
1475 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1476 string_format() function, because snprintf() does not exist on all
1479 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1480 ${stat: expansion item.
1482 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1483 protocol synchronization error", to keep the pedants happy.
1485 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1486 well as for IRIX systems, when gcc is being used. See the host.c source
1489 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1491 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1494 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1495 transport and to the smtp transport in LMTP mode.
1497 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1499 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1500 run a filter in a subprocess. This could lead to confusion in subsequent
1501 lookups in the parent process. There should also be a search_tidyup() at
1502 the end of the subprocess.
1504 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1505 only if the host matched helo_try_verify_hosts, which caused the
1506 verification to occur when the EHLO/HELO command was issued. The ACL just
1507 tested the remembered result. Now, if a previous verification attempt has
1508 not happened, "verify = helo" does it there and then.
1510 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1512 TK/03 Fix log output including CR from clamd.
1514 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1515 exist provoked a memory error which could cause a segfault.
1517 PH/15 Installed PCRE 6.2
1519 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1521 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1522 of the problem. Specifically, suggested +O2 rather than +O1 for the
1525 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1527 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1528 record, the port was not correctly logged when the outgoing_port log
1529 selector was set (it logged the transort's default port).
1531 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1532 fallback_hosts, and "hosts" in the smtp transport.
1534 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1535 host errors such as "Connection refused".
1537 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1538 authentication with radiusclient 0.4.9:
1540 - Error returned from rc_read_config was caught wrongly
1541 - Username/password not passed on to radius server due to wrong length.
1543 The presumption is that some radiusclient API changes for 4.51/PH/17
1544 were not taken care of correctly. The code is still untested by me (my
1545 Linux distribution still has 0.3.2 of radiusclient), but it was
1546 contributed by a Radius user.
1548 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1549 expanding the "port" option of the smtp transport.
1551 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1552 while reading a MIME header. Thanks to Tom Hughes for a patch.
1554 PH/24 Include config.h inside local_scan.h so that configuration settings are
1557 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1558 that in an ACL for RCPT (for example), you can examine exactly what was
1561 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1562 commands, but it was not correctly comparing the address with the actual
1563 client host address. Thus, it would show the EHLO address in Received:
1564 header lines when this was not necessary.
1566 PH/27 Added the % operator to ${eval:}.
1568 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1569 it should be ignoring failures (because with -C, for example, it has lost
1570 privilege). It wasn't ignoring creation failures other than "already
1573 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1576 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1577 because an input line was too long, either on its own, or by virtue of
1578 too many continuations, the temporary file was not being removed, and the
1579 return code was incorrect.
1581 PH/31 Missing "BOOL" in function definition in filtertest.c.
1583 PH/32 Applied Sieve patches from the maintainer.
1585 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1587 PH/33 Added "verify = not_blind".
1589 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1590 Local/Makefile (with some defaults set). These are used in built scripts
1591 such as exicyclog, but they have never been used in the exim_install
1592 script (though there are many overriding facilities there). I have
1593 arranged that the exim_install script now takes note of these two
1596 PH/35 Installed configuration files for Dragonfly.
1598 PH/36 When a locally submitted message by a trusted user did not contain a
1599 From: header, and the sender address was obtained from -f or from an SMTP
1600 MAIL command, and the trusted user did not use -F to supply a sender
1601 name, $originator_name was incorrectly used when constructing a From:
1602 header. Furthermore, $originator_name was used for submission mode
1603 messages from external hosts without From: headers in a similar way,
1604 which is clearly wrong.
1606 PH/37 Added control=suppress_local_fixups.
1608 PH/38 When log_selector = +received_sender was set, and the addition of the
1609 sender made the log line's construction buffer exactly full, or one byte
1610 less than full, an overflow happened when the terminating "\n" was
1613 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1614 when the result of a list match is failure because a DNS lookup failed.
1616 PH/40 RM_COMMAND is now used in the building process.
1618 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1619 the "build-* directories that it finds.
1621 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1622 address in a domain literal was a prefix of an interface address.
1624 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1625 when verifying a sender address, unless rewrite_headers is false.
1627 PH/44 Wrote a long comment about why errors_to addresses are verified as
1628 recipients, not senders.
1630 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1631 the ratelimit ACL was added.
1633 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1635 PH/46 Added extra information about PostgreSQL errors to the error string.
1637 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1638 avoiding the need to install special zones in a real server. This is
1639 backwards compatible; if it can't find the fake resolver, it drops back.
1640 Thus, both old and new test suites can be run.
1642 TF/02 Added util/ratelimit.pl
1644 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1645 clock is set back in time.
1647 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1648 Candler <B.Candler@pobox.com>.
1650 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1651 OK for addresses that are the result of redirections.
1653 PH/48 A number of further additions for the benefit of the new test suite,
1654 including a fake gethostbyname() that interfaces to the fake DNS resolver
1657 TF/06 The fix for widen_domains has also been applied to qualify_single and
1658 search_parents which are the other dnslookup options that can cause
1661 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1664 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1666 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1667 service" if it encountered an MX record with an empty target hostname.
1668 The message is now "an MX or SRV record indicated no SMTP service".
1670 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1671 if the DNS of the sending site is misconfigured. This is quite a
1672 common situation. This change restores the behaviour of treating a
1673 helo verification defer as a failure.
1675 PH/51 If self=fail was set on a router, the bounce message did not include the
1676 actual error message.
1682 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1684 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1685 command itself ended in error, the underlying message about the transport
1686 filter timeout was being overwritten with the pipe command error. Now the
1687 underlying error message should be appended to the second error message.
1689 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1691 PH/02 Reduce the amount of output that "make" produces by default. Full output
1692 can still be requested.
1694 PH/03 The warning log line about a condition test deferring for a "warn" verb
1695 was being output only once per connection, rather than after each
1696 occurrence (because it was using the same function as for successful
1697 "warn" verbs). This seems wrong, so I have changed it.
1699 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1700 it should not have, which might have caused a crash in the right
1701 circumstances, but probably never did.
1703 PH/04 Installed a modified version of Tony Finch's patch to make submission
1704 mode fix the return path as well as the Sender: header line, and to
1705 add a /name= option so that you can make the user's friendly name appear
1708 TF/03 Added the control = fakedefer ACL modifier.
1710 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1711 Mark Lowes for thorough testing.
1713 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1715 TK/03 Merged latest SRS patch from Miles Wilton.
1717 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1718 with the definition in sysexits.h (which is #included earlier).
1719 Fortunately, Exim does not actually use EX_OK. The code used to try to
1720 preserve the sysexits.h value, by assumimg that macro definitions were
1721 scanned for macro replacements. I have been disabused of this notion,
1722 so now the code just undefines EX_OK before #including unistd.h.
1724 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1725 in the smtp transport. When a block could not be written in a single
1726 write() function, the timeout was being re-applied to each part-write.
1727 This seems wrong - if the receiver was accepting one byte at a time it
1728 would take for ever. The timeout is now adjusted when this happens. It
1729 doesn't have to be particularly precise.
1731 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1732 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1734 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1735 without a domain if the check to <postmaster@domain> fails.
1737 SC/01 Eximstats: added -xls and the ability to specify output files
1738 (patch written by Frank Heydlauf).
1740 SC/02 Eximstats: use FileHandles for outputing results.
1742 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1744 SC/04 Eximstats: fixed display of large numbers with -nvr option
1746 SC/05 Eximstats: fixed merging of reports with empty tables.
1748 SC/06 Eximstats: added the -include_original_destination flag
1750 SC/07 Eximstats: removed tabs and trailing whitespace.
1752 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1754 TK/06 MBOX spool code: Add real "From " MBOX separator line
1755 so the .eml file is really in mbox format (even though
1756 most programs do not really care). Patch from Alex Miller.
1758 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1759 The latter is generated from $received_to and is only set if the
1760 message has one envelope recipient. SA can use these headers,
1761 obviously out-of-the-box. Patch from Alex Miller.
1763 PH/08 The ${def test on a variable was returning false if the variable's
1764 value was "0", contrary to what the specification has always said!
1765 The result should be true unless the variable is empty.
1767 PH/09 The syntax error of a character other than { following "${if
1768 def:variable_name" (after optional whitespace) was not being diagnosed.
1769 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1770 accidental colon was present, for example, could give incorrect results.
1772 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1773 result is used (not including appendfile, where other changes are about
1776 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1777 This involved changing a lot of size variables from int to off_t. It
1778 should work with maildirs and everything.
1780 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1781 spamd dying while we are connected to it.
1783 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1786 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1787 function for BDB 4.3.
1789 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1791 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1792 the sources for PCRE have been re-arranged and more files are now
1795 PH/15 The code I had for printing potentially long long variables in PH/11
1796 above was not the best (it lost precision). The length of off_t variables
1797 is now inspected at build time, and an appropriate printing format (%ld
1798 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1799 to be "long long int" or "long int". This is needed for the internal
1800 formatting function string_vformat().
1802 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1803 the configuration file to be ":syslog", then the script "guesses" where
1804 the logs files are, rather than using the compiled in default. In our
1805 case the guess is not the same as the compiled default, so the script
1806 suddenly stopped working when I started to use syslog. The patch checks
1807 to see if log_file_path is "". If so, it attempts to read it from exim
1808 with no configuration file to get the compiled in version, before it
1809 falls back to the previous guessing code."
1811 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1812 implementing BATV in an Exim configuration. See NewStuff for the gory
1815 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1816 Makefile that are specific to HP-UX.
1818 PH/18 If the "use_postmaster" option was set for a recipient callout together
1819 with the "random" option, the postmaster address was used as the MAIL
1820 FROM address for the random test, but not for the subsequent recipient
1821 test. It is now used for both.
1823 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1824 patch removes a few documentation additions to RFC 3028, because the
1825 latest draft now contains them. It adds the new en;ascii-case comparator
1826 and a new error check for 8bit text in MIME parts. Comparator and
1827 require names are now matched exactly. I enabled the subaddress
1828 extension, but it is not well tested yet (read: it works for me)."
1830 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1831 rework some of the code of TK/09 above to avoid the hardwired use of
1832 "%lld" and "long long". Replaced the call to snprintf() with a call to
1835 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1836 records point to non-existent hosts", "retry timeout exceeded", and
1837 "retry time not reached for any host after a long failure period".
1839 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1840 experimental DomainKeys support:
1842 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1843 (2) On an error such as an illegally used "control", the wrong name for
1844 the control was given.
1846 These problems did NOT occur unless DomainKeys support was compiled.
1848 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1850 PH/24 Added ${if match_ip condition.
1852 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1853 databases so that it will be absolutely obvious if a crash occurs in the
1854 DB library. This is a regular occurrence (often caused by mis-matched
1857 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1858 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1859 picked up on a user's system that detects such things. There doesn't seem
1860 to be a gcc warning option for this - only an attribute that has to be
1861 put on the function's prototype. It seems that in Fedora Core 4 they have
1862 set this on a number of new functions. No doubt there will be more in due
1865 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1866 specify a transport. However, if an address that was verified by such a
1867 router was the subject of a callout, Exim crashed because it tried to
1868 read the rcpt_include_affixes from the non-existent transport. Now it
1869 just assumes that the setting of that option is false. This bug was
1870 introduced by 4.51/PH/31.
1872 PH/28 Changed -d+all to exclude +memory, because that information is very
1873 rarely of interest, but it makes the output a lot bigger. People tend to
1874 do -d+all out of habit.
1876 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1877 code in os-type was giving problems when libc.so lives in lib64, like on
1880 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1881 aren't the modern standard, and it seems that some systems' include files
1882 don't always have them. Exim was already checking for some of the newer
1883 ones like T_AAAA, and defining it itself. I've added checks for all the
1884 record types that Exim uses.
1886 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1887 not automatically generating a new one, as it is supposed to. This
1888 prevented TLS from working. If the file did exist, but contained invalid
1889 data, a new version was generated, as expected. It was only the case of a
1890 non-existent file that was broken.
1892 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1893 with a change in libdomainkeys > 0.64.
1895 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1896 from DNS. If the selector record carries the flag, it now has
1897 precedence over the domain-wide flag.
1899 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1901 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1902 the use of a port name that isn't defined in /etc/services) occurred, the
1903 message was deferred as in a normal delivery, and thus remained on the
1904 spool, instead of being failed because of the mua_wrapper setting. This
1905 is now fixed, and I tidied up some of the mua_wrapper messages at the
1908 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1909 the messages in a hash of arrays rather than using individual hashes.
1910 This is a bit cleaner and results in dramatic memory savings, albeit
1911 at a slight CPU cost.
1913 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1914 as requested by Marc Sherman.
1916 SC/10 Eximstats: added histograms for user specified patterns as requested
1919 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1921 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1922 fopen() in the content-scanning modules that did not already have it.
1928 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1929 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1931 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1933 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1935 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1936 to test Sieve filters that use "vacation".
1938 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1939 that changes the way the GnuTLS parameters are stored in the cache file.
1940 The new format can be generated externally. For backward compatibility,
1941 if the data in the cache doesn't make sense, Exim assumes it has read an
1942 old-format file, and it generates new data and writes a new file. This
1943 means that you can't go back to an older release without removing the
1946 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1947 work if there are any delivery delays because "one_time" forces the
1948 parent to be marked "delivered", so its unseen clone is never tried
1949 again. For this reason, Exim now forbids the simultaneous setting of
1952 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1953 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1954 entry if you want to know the details. The fix, however, neglected to
1955 consider the case when local delivery batching is involved. The test for
1956 "previously delivered" was not happening when checking to see if an
1957 address could be batched with a previous (undelivered) one; under
1958 certain circumstances this could lead to multiple deliveries to the same
1961 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1962 in its include files, and this causes problems building Exim.
1964 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1965 header_syntax) but Exim was just ignoring anything given after a slash.
1966 In particular, this caused confusion with an attempt to use "verify =
1967 reverse_host_lookup/defer_ok". An error is now given when options are
1968 supplied for verify items that do not have them. (Maybe reverse_host_
1969 lookup should have a defer_ok option, but that's a different point.)
1971 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1972 defined by RFC 821) to 2048, because there were problems with some AUTH
1973 commands, and RFC 1869 says the size should be increased for extended
1974 SMTP commands that take arguments.
1976 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1979 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1980 "unknown" error; now it says that the functionality isn't in the binary.
1982 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1983 an address' error message when a string expansion fails (syntax or
1984 whatever). Otherwise the password may appear in the log. Following change
1985 PH/42 below, there is no longer a chance of it appearing in a bounce
1988 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1990 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1991 of its addresses was ever tried. (Bugzilla bug #2.)
1993 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1994 the result incorrectly in the debug output. (It correctly added a newline
1995 to what was transported.)
1997 TF/01 Added $received_time.
1999 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2000 commented out examples of how to interface to a virus scanner and to
2001 SpamAssassin. Also added commented examples of av_scanner and
2002 spamd_address settings.
2004 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2005 and controls are allowed in which ACLs. There were a couple of minor
2006 errors. Some of the entries in the conditions table (which is a table of
2007 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2008 negation of where the condition IS allowed.
2010 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2012 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2013 header file does not have a version number, so I've had to invent a new
2014 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2015 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2016 radiusclient), but it was contributed by a Radius user.
2018 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2019 files or directories, for OpenSSL.
2021 PH/19 When an Exim process that is running as root has to create an Exim log
2022 file, it does so in a subprocess that runs as exim:exim so as to get the
2023 ownership right at creation (otherwise, other Exim processes might see
2024 the file with the wrong ownership). There was no test for failure of this
2025 fork() call, which would lead to the process getting stuck as it waited
2026 for a non-existent subprocess. Forks do occasionally fail when resources
2027 run out. I reviewed all the other calls to fork(); they all seem to check
2030 PH/20 When checking for unexpected SMTP input at connect time (before writing
2031 the banner), Exim was not dealing correctly with a non-positive return
2032 from the read() function. If the client had disconnected by this time,
2033 the result was a log entry for a synchronization error with an empty
2034 string after "input=" when read() returned zero. If read() returned -1
2035 (an event I could not check), uninitialized data bytes were printed.
2036 There were reports of junk text (parts of files, etc) appearing after
2039 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2041 PH/22 Added support for macro redefinition, and (re)definition in between
2042 driver and ACL definitions.
2044 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2045 forgetting to use the resulting value; it was using the unexpanded value.
2047 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2048 hadn't been configured. The fix is from Juergen Kreileder, who
2049 understands it better than I do:
2051 "Here's what I see happening with three configured cyrus_sasl
2052 authenticators configured (plain, login, cram-md5):
2054 On startup auth_cyrus_sasl_init() gets called for each of these.
2055 This means three calls to sasl_listmech() without a specified mech_list.
2056 => SASL tests which mechs of all available mechs actually work
2057 => three warnings about OTP not working
2058 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2060 With the patch, sasl_listmech() also gets called three times. But now
2061 SASL's mech_list option is set to the server_mech specified in the the
2062 authenticator. Or in other words, the answer from sasl_listmech()
2063 gets limited to just the mech you're testing for (which is different
2065 => the return list contains just 'plain' or 'login', 'cram-md5' or
2066 nothing depending on the value of ob->server_mech.
2068 I've just tested the patch: Authentication still works fine,
2069 unavailable mechs specified in the exim configuration are still
2070 caught, and the auth.log warnings about OTP are gone."
2072 PH/25 When debugging is enabled, the contents of the command line are added
2073 to the debugging output, even when log_selector=+arguments is not
2076 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2077 answer is "GNU", and only if the return is "GNU/something" is the answer
2080 PH/27 $acl_verify_message is now set immediately after the failure of a
2081 verification in an ACL, and so is available in subsequent modifiers. In
2082 particular, the message can be preserved by coding like this:
2084 warn !verify = sender
2085 set acl_m0 = $acl_verify_message
2087 Previously, $acl_verify_message was set only while expanding "message"
2088 and "log_message" when a very denied access.
2090 PH/28 Modified OS/os.c-Linux with
2092 -#ifndef OS_LOAD_AVERAGE
2093 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2095 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2096 nomenclature these days.)
2098 PH/29 Installed patch from the Sieve maintainer that adds the options
2099 sieve_useraddress and sieve_subaddress to the redirect router.
2101 PH/30 In these circumstances:
2102 . Two addresses routed to the same list of hosts;
2103 . First host does not offer TLS;
2104 . First host accepts first address;
2105 . First host gives temporary error to second address;
2106 . Second host offers TLS and a TLS session is established;
2107 . Second host accepts second address.
2108 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2109 and peerdn, if requested) that were in fact used only for the second
2112 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2113 attention to any local part prefix or suffix that was matched by the
2114 router that accepted the address. It now behaves in the same way as it
2115 does for delivery: the affixes are removed from the local part unless
2116 rcpt_include_affixes is set on the transport.
2118 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2119 timeout during the DATA phase of an incoming message.
2121 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2122 applied a patch sanctioned by the Sieve maintainer.
2124 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2125 the uid or gid is negative. A case of a negative gid caused this to be
2126 noticed. The fix allows for either to be negative.
2128 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2129 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2132 PH/36 The change PH/12 above was broken. Fixed it.
2134 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2135 the grounds that routing the same address twice would always produce the
2136 same answer. This might have been true once, but it is certainly no
2137 longer true now. Routing a child address may depend on the previous
2138 routing that produced that child. Some complicated redirection strategies
2139 went wrong when messages had multiple recipients, and made Exim's
2140 behaviour dependent on the order in which the addresses were given.
2142 I have moved the duplicate checking until after the routing is complete.
2143 Exim scans the addresses that are assigned to local and remote
2144 transports, and removes any duplicates. This means that more work will be
2145 done, as duplicates will always all be routed, but duplicates are
2146 presumably rare, so I don't expect this is of any significance.
2148 For deliveries to pipes, files, and autoreplies, the duplicate checking
2149 still happens during the routing process, since they are not going to be
2152 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2153 It corrects a timeout issue with spamd. This is Ian's comment: "The
2154 background is that sometimes spamd either never reads data from a
2155 connection it has accepted, or it never writes response data. The exiscan
2156 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2157 blindly assumes that writes won't block so it may never time out."
2159 PH/39 Allow G after quota size as well as K and M.
2161 PH/40 The value set for $authenticated_id in an authenticator may not contain
2162 binary zeroes or newlines because the value is written to log lines and
2163 to spool files. There was no check on this. Now the value is run through
2164 the string_printing() function so that such characters are converted to
2165 printable escape sequences.
2167 PH/41 $message_linecount is a new variable that contains the total number of
2168 lines in the message. Compare $body_linecount, which is the count for the
2171 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2172 bounce and delay warning messages, except in certain special cases, which
2175 (a) An SMTP error message from a remote host;
2176 (b) A message specified in a :fail: redirection;
2177 (c) A message specified in a "fail" command in a system filter;
2178 (d) A message specified in a FAIL return from the queryprogram router;
2179 (e) A message specified by the cannot_route_message router option.
2181 In these cases only, Exim does include the error details in bounce and
2182 warning messages. There are also a few cases where bland messages such
2183 as "unrouteable address" or "local delivery error" are given.
2185 PH/43 $value is now also set for the "else" part of a ${run expansion.
2187 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2188 being worked on, but at least Exim now implements the latest version to
2191 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2192 process to complete was treated as a delivery failure, a timeout while
2193 writing the message to the pipe was logged, but erroneously treated as a
2194 successful delivery. Such timeouts include transport filter timeouts. For
2195 consistency with the overall process timeout, these timeouts are now
2196 treated as errors, giving rise to delivery failures by default. However,
2197 there is now a new Boolean option for the pipe transport called
2198 timeout_defer, which, if set TRUE, converts the failures into defers for
2199 both kinds of timeout. A transport filter timeout is now identified in
2202 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2203 systems where "make" and "gmake" are different, calling "gmake" at top
2204 level broke things. I've arranged for the value of $(MAKE) to be passed
2205 from the Makefile to this script so that it can call the same version of
2209 A note about Exim versions 4.44 and 4.50
2210 ----------------------------------------
2212 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2213 changes of various kinds. As a consequence, a big documentation update was
2214 needed. This delayed the release for rather longer than seemed good, especially
2215 in the light of a couple of (minor) security issues. Therefore, the changes
2216 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2217 release. So 4.44 and 4.50 are in effect two different branches that both start
2220 I have left the 4.50 change log unchanged; it contains all the changes since
2221 4.43. The change log for 4.44 is below; many of its items are identical to
2222 those for 4.50. This seems to be the most sensible way to preserve the
2223 historical information.
2229 1. Minor wording change to the doc/README.SIEVE file.
2231 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2232 computation of the current number of files was incorrect.
2234 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2235 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2238 4. Give more explanation in the error message when the command for a transport
2239 filter fails to execute.
2241 5. There are several places where Exim runs a non-Exim command in a
2242 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2243 was being done only for the command run by the queryprogram router. It is
2244 now done for all such subprocesses. The other cases are: ${run, transport
2245 filters, and the commands run by the lmtp and pipe transports.
2247 6. Added CONFIGURE_GROUP build-time option.
2249 7. Some older OS have a limit of 256 on the maximum number of file
2250 descriptors. Exim was using setrlimit() to set 1000 as a large value
2251 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2252 systems. I've change it so that if it can't get 1000, it tries for 256.
2254 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2255 was an oversight, and furthermore, ever since the addition of extra
2256 controls (e.g. 4.43/32), the checks on when to allow different forms of
2257 "control" were broken. There should now be diagnostics for all cases when a
2258 control that does not make sense is encountered.
2260 9. Added the /retain_sender option to "control=submission".
2262 10. $recipients is now available in the predata ACL (oversight).
2264 11. Tidy the search cache before the fork to do a delivery from a message
2265 received from the command line. Otherwise the child will trigger a lookup
2266 failure and thereby defer the delivery if it tries to use (for example) a
2267 cached ldap connection that the parent has called unbind on.
2269 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2270 of $address_data from the recipient verification was clobbered by the
2271 sender verification.
2273 13. The value of address_data from a sender verification is now available in
2274 $sender_address_data in subsequent conditions in the ACL statement.
2276 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2278 15. Added a new option "connect=<time>" to callout options, to set a different
2281 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2282 was its contents. (It was OK if the option was not defined at all.)
2284 17. A "Completed" log line is now written for messages that are removed from
2285 the spool by the -Mrm option.
2287 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2288 information about exactly what failed.
2290 19. Added -dd to debug only the daemon process.
2292 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2293 handles timeouts, both on the server side and network timeouts. Renamed the
2294 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2296 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2297 It is now set to "smtps".
2299 22. $host_address is now set to the target address during the checking of
2300 ignore_target_hosts.
2302 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2303 being passed; this would have caused $sender_host_name to have been used if
2304 matching the list had actually called for a host name (not very likely,
2305 since this list is usually IP addresses). A host name is now passed as
2308 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2309 code that specifies a non-restarting handler (typically sigaction() in
2310 modern systems) in an attempt to fix a rare and obscure crash bug.
2312 25. Narrowed the window for a race in the daemon that could cause it to ignore
2313 SIGCHLD signals. This is not a major problem, because they are used only to
2314 wake it up if nothing else does.
2316 26. A malformed maildirsize file could cause Exim to calculate negative values
2317 for the mailbox size or file count. Odd effects could occur as a result.
2318 The maildirsize information is now recalculated if the size or filecount
2321 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2322 support for a long time. Removed HAVE_SYS_VFS_H.
2324 28. Installed the latest version of exipick from John Jetmore.
2326 29. In an address list, if the pattern was not a regular expression, an empty
2327 subject address (from a bounce message) matched only if the pattern was an
2328 empty string. Non-empty patterns were not even tested. This was the wrong
2329 because it is perfectly reasonable to use an empty address as part of a
2330 database query. An empty address is now tested by patterns that are
2331 lookups. However, all the other forms of pattern expect the subject to
2332 contain a local part and a domain, and therefore, for them, an empty
2333 address still always fails if the pattern is not itself empty.
2335 30. Exim went into a mad DNS loop when attempting to do a callout where the
2336 host was specified on an smtp transport, and looking it up yielded more
2337 than one IP address.
2339 31. Re-factored the code for checking spool and log partition space into a
2340 function that finds that data and another that does the check. The former
2341 is then used to implement four new variables: $spool_space, $log_space,
2342 $spool_inodes, and $log_inodes.
2344 32. The RFC2047 encoding function was originally intended for short strings
2345 such as real names; it was not keeping to the 75-character limit for
2346 encoded words that the RFC imposes. It now respects the limit, and
2347 generates multiple encoded words if necessary. To be on the safe side, I
2348 have increased the buffer size for the ${rfc2047: expansion operator from
2351 33. It is now permitted to omit both strings after an "if" condition; if the
2352 condition is true, the result is "true". As before, when the second string
2353 is omitted, a false condition yields an empty string. This makes it less
2354 cumbersome to write custom ACL and router conditions.
2356 34. Failure to deliver a bounce message always caused it to be frozen, even if
2357 there was an errors_to setting on the router. The errors_to setting is now
2360 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2361 canonical form (fully expanded) before being placed in
2362 $sender_host_address.
2364 36. The table in the code that translates DNS record types into text (T_A to
2365 "A" for instance) was missing entries for NS and CNAME. It is just possible
2366 that this could have caused confusion if both these types were looked up
2367 for the same domain, because the text type is used as part of Exim's
2368 per-process caching. But the chance of anyone hitting this buglet seems
2371 37. The dnsdb lookup has been extended in a number of ways.
2373 (1) There is a new type, "zns", which walks up the domain tree until it
2374 finds some nameserver records. It should be used with care.
2376 (2) There is a new type, "mxh", which is like "mx" except that it returns
2377 just the host names, not the priorities.
2379 (3) It is now possible to give a list of domains (or IP addresses) to be
2380 looked up. The behaviour when one of the lookups defers can be
2381 controlled by a keyword.
2383 (4) It is now possible to specify the separator character for use when
2384 multiple records are returned.
2386 38. The dnslists ACL condition has been extended: it is now possible to supply
2387 a list of IP addresses and/or domains to be looked up in a particular DNS
2390 39. Added log_selector=+queue_time_overall.
2392 40. When running the queue in the test harness, wait just a tad after forking a
2393 delivery process, to get repeatability of debugging output.
2395 41. Include certificate and key file names in error message when GnuTLS fails
2396 to set them up, because the GnuTLS error message doesn't include the name
2397 of the failing file when there is a problem reading it.
2399 42. Allow both -bf and -bF in the same test run.
2401 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2403 44. The "Exiscan patch" is now merged into the mainline Exim source.
2405 45. Sometimes the final signoff response after QUIT could fail to get
2406 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2407 < 0 before doing a fflush(). This bug looks as though it goes back to the
2408 introduction of TLS in release 3.20, but "sometimes" must have been rare
2409 because the tests only now provoked it.
2411 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2412 (this can affect the format of dates).
2414 47. exim_tidydb, when checking for the continued existence of a message for
2415 which it has found a message-specific retry record, was not finding
2416 messages that were in split spool directories. Consequently, it was
2417 deleting retry records that should have stayed in existence.
2419 48. Steve fixed some bugs in eximstats.
2421 49. The SPA authentication driver was not abandoning authentication and moving
2422 on to the next authenticator when an expansion was forced to fail,
2423 contradicting the general specification for all authenticators. Instead it
2424 was generating a temporary error. It now behaves as specified.
2426 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2427 (the order specifies the preference for clients). The order is now AES256,
2428 AES128, 3DES, ARCFOUR128.
2430 51. Small patch to Sieve code - explicitly set From: when generating an
2433 52. Exim crashed if a remote delivery caused a very long error message to be
2434 recorded - for instance if somebody sent an entire SpamAssassin report back
2435 as a large number of 550 error lines. This bug was coincidentally fixed by
2436 increasing the size of one of Exim's internal buffers (big_buffer) that
2437 happened as part of the Exiscan merge. However, to be on the safe side, I
2438 have made the code more robust (and fixed the comments that describe what
2441 53. Now that there can be additional text after "Completed" in log lines (if
2442 the queue_time_overall log selector is set), a one-byte patch to exigrep
2443 was needed to allow it to recognize "Completed" as not the last thing in
2446 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2447 patch that reportedly fixes this has been added. I am not expert enough to
2448 create a test for it. This is what the patch creator wrote:
2450 "I found a little strange behaviour of ldap code when working with
2451 Windows 2003 AD Domain, where users was placed in more than one
2452 Organization Units. When I tried to give exim partial DN, the exit code
2453 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2454 But simultaneously result of request was absolutely normal ldap result,
2455 so I produce this patch..."
2457 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2458 REFERENCE, so I have modified the code to exclude the patch when that macro
2461 55. Some experimental protocols are using DNS PTR records for new purposes. The
2462 keys for these records are domain names, not reversed IP addresses. The
2463 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2464 leaves it alone. Component reversal etc. now happens only for IP addresses.
2467 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2469 57. Double the size of the debug message buffer (to 2048) so that more of very
2470 long debug lines gets shown.
2472 58. The exicyclog utility now does better if the number of log files to keep
2473 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2475 59. Two changes related to the smtp_active_hostname option:
2477 (1) $smtp_active_hostname is now available as a variable.
2478 (2) The default for smtp_banner uses $smtp_active_hostname instead
2479 of $primary_hostname.
2481 60. The host_aton() function is supposed to be passed a string that is known
2482 to be a valid IP address. However, in the case of IPv6 addresses, it was
2483 not checking this. This is a hostage to fortune. Exim now panics and dies
2484 if the condition is not met. A case was found where this could be provoked
2485 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2486 components; fortuitously, this particular loophole had already been fixed
2487 by change 4.50/55 above.
2489 If there are any other similar loopholes, the new check in host_aton()
2490 itself should stop them being exploited. The report I received stated that
2491 data on the command line could provoke the exploit when Exim was running as
2492 exim, but did not say which command line option was involved. All I could
2493 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2494 running as the user.
2497 61. There was a buffer overflow vulnerability in the SPA authentication code
2498 (which came originally from the Samba project). I have added a test to the
2499 spa_base64_to_bits() function which I hope fixes it.
2502 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2503 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2505 63. The daemon start-up calls getloadavg() while still root for those OS that
2506 need the first call to be done as root, but it missed one case: when
2507 deliver_queue_load_max is set with deliver_drop_privilege. This is
2508 necessary for the benefit of the queue runner, because there is no re-exec
2509 when deliver_drop_privilege is set.
2511 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2512 This has been fixed.
2514 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2515 was in use, was not putting the data itself into the right store pool;
2516 consequently, it could be overwritten for a subsequent message in the same
2517 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2520 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2522 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2523 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2524 the function were treating the return as a boolean value, which happened to
2525 work because 0=false and not-0=true, but is not correct code.
2527 68. The host_aton() function was not handling scoped IPv6 addresses (those
2528 with, for example, "%eth0" on the end) correctly.
2530 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2531 negated items (that is, ~something) in unsigned ints. Some compilers
2532 apparently mutter when there is no cast.
2534 70. If an address verification called from an ACL failed, and did not produce a
2535 user-specific message (i.e. there was only a "system" message), nothing was
2536 put in $acl_verify_message. In this situation, it now puts the system
2539 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2540 session; change 4.31/43 added the unwanted input to the log line - except
2541 that it did not do this in the start of session case. It now does.
2543 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2544 This is wrong and can cause the other end to generate a synchronization
2545 error if it is another Exim or anything else that does the synchronization
2546 check. A QUIT command is no longer sent after a timeout.
2548 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2549 during host lookups.
2551 74. The defer_ok option of callout verification was not working if it was used
2552 when verifying addresses in header lines, that is, for this case:
2554 verify = header_sender/callout=defer_ok
2556 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2557 those file descriptors could be used for SMTP connections. If anything
2558 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2559 could be sent to the SMTP client, causing chaos. The daemon now opens
2560 stdin, stdout, and stderr to /dev/null when it puts itself into the
2563 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2564 log by default. The user can override this with suitable Perl magic.
2566 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2567 the log message when discard triggers, was not working for the DATA ACL or
2568 for the non-SMTP ACL.
2570 78. Error message wording change in sieve.c.
2572 79. If smtp_accept_max_per_host was set, the number of connections could be
2573 restricted to fewer than expected, because the daemon was trying to set up
2574 a new connection before checking whether the processes handling previous
2575 connections had finished. The check for completed processes is now done
2576 earlier. On busy systems, this bug wouldn't be noticed because something
2577 else would have woken the daemon, and it would have reaped the completed
2580 80. If a message was submitted locally by a user whose login name contained one
2581 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2582 It caused a spool format error. I have fixed the spool reading code. A
2583 related problem was that the "from" clause in the Received: line became
2584 illegal because of the space(s). It is now covered by ${quote_local_part.
2586 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2589 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2591 83. Patch from Sieve maintainer to fix unterminated string problem in
2592 "vacation" handling.
2594 84. Some minor changes to the Linux configuration files to help with other
2595 OS variants using glibc.
2597 85. One more patch for Sieve to update vacation handling to latest spec.
2600 ----------------------------------------------------
2601 See the note above about the 4.44 and 4.50 releases.
2602 ----------------------------------------------------
2608 1. Change 4.43/35 introduced a bug that caused file counts to be
2609 incorrectly computed when quota_filecount was set in an appendfile
2612 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2613 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2616 3. Give more explanation in the error message when the command for a transport
2617 filter fails to execute.
2619 4. There are several places where Exim runs a non-Exim command in a
2620 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2621 was being done only for the command run by the queryprogram router. It is
2622 now done for all such subprocesses. The other cases are: ${run, transport
2623 filters, and the commands run by the lmtp and pipe transports.
2625 5. Some older OS have a limit of 256 on the maximum number of file
2626 descriptors. Exim was using setrlimit() to set 1000 as a large value
2627 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2628 systems. I've change it so that if it can't get 1000, it tries for 256.
2630 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2631 was an oversight, and furthermore, ever since the addition of extra
2632 controls (e.g. 4.43/32), the checks on when to allow different forms of
2633 "control" were broken. There should now be diagnostics for all cases when a
2634 control that does not make sense is encountered.
2636 7. $recipients is now available in the predata ACL (oversight).
2638 8. Tidy the search cache before the fork to do a delivery from a message
2639 received from the command line. Otherwise the child will trigger a lookup
2640 failure and thereby defer the delivery if it tries to use (for example) a
2641 cached ldap connection that the parent has called unbind on.
2643 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2644 of $address_data from the recipient verification was clobbered by the
2645 sender verification.
2647 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2648 was its contents. (It was OK if the option was not defined at all.)
2650 11. A "Completed" log line is now written for messages that are removed from
2651 the spool by the -Mrm option.
2653 12. $host_address is now set to the target address during the checking of
2654 ignore_target_hosts.
2656 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2657 being passed; this would have caused $sender_host_name to have been used if
2658 matching the list had actually called for a host name (not very likely,
2659 since this list is usually IP addresses). A host name is now passed as
2662 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2663 code that specifies a non-restarting handler (typically sigaction() in
2664 modern systems) in an attempt to fix a rare and obscure crash bug.
2666 15. Narrowed the window for a race in the daemon that could cause it to ignore
2667 SIGCHLD signals. This is not a major problem, because they are used only to
2668 wake it up if nothing else does.
2670 16. A malformed maildirsize file could cause Exim to calculate negative values
2671 for the mailbox size or file count. Odd effects could occur as a result.
2672 The maildirsize information is now recalculated if the size or filecount
2675 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2676 support for a long time. Removed HAVE_SYS_VFS_H.
2678 18. Updated exipick to current release from John Jetmore.
2680 19. Allow an empty sender to be matched against a lookup in an address list.
2681 Previously the only cases considered were a regular expression, or an
2684 20. Exim went into a mad DNS lookup loop when doing a callout where the
2685 host was specified on the transport, if the DNS lookup yielded more than
2688 21. The RFC2047 encoding function was originally intended for short strings
2689 such as real names; it was not keeping to the 75-character limit for
2690 encoded words that the RFC imposes. It now respects the limit, and
2691 generates multiple encoded words if necessary. To be on the safe side, I
2692 have increased the buffer size for the ${rfc2047: expansion operator from
2695 22. Failure to deliver a bounce message always caused it to be frozen, even if
2696 there was an errors_to setting on the router. The errors_to setting is now
2699 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2700 canonical form (fully expanded) before being placed in
2701 $sender_host_address.
2703 24. Updated eximstats to version 1.33
2705 25. Include certificate and key file names in error message when GnuTLS fails
2706 to set them up, because the GnuTLS error message doesn't include the name
2707 of the failing file when there is a problem reading it.
2709 26. Expand error message when OpenSSL has problems setting up cert/key files.
2712 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2713 (this can affect the format of dates).
2715 28. exim_tidydb, when checking for the continued existence of a message for
2716 which it has found a message-specific retry record, was not finding
2717 messages that were in split spool directories. Consequently, it was
2718 deleting retry records that should have stayed in existence.
2720 29. eximstats updated to version 1.35
2721 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2722 1.35 - bugfix such that pie charts by volume are generated correctly
2724 30. The SPA authentication driver was not abandoning authentication and moving
2725 on to the next authenticator when an expansion was forced to fail,
2726 contradicting the general specification for all authenticators. Instead it
2727 was generating a temporary error. It now behaves as specified.
2729 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2730 (the order specifies the preference for clients). The order is now AES256,
2731 AES128, 3DES, ARCFOUR128.
2733 31. Small patch to Sieve code - explicitly set From: when generating an
2736 32. Exim crashed if a remote delivery caused a very long error message to be
2737 recorded - for instance if somebody sent an entire SpamAssassin report back
2738 as a large number of 550 error lines. This bug was coincidentally fixed by
2739 increasing the size of one of Exim's internal buffers (big_buffer) that
2740 happened as part of the Exiscan merge. However, to be on the safe side, I
2741 have made the code more robust (and fixed the comments that describe what
2744 33. Some experimental protocols are using DNS PTR records for new purposes. The
2745 keys for these records are domain names, not reversed IP addresses. The
2746 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2747 leaves it alone. Component reversal etc. now happens only for IP addresses.
2750 34. The host_aton() function is supposed to be passed a string that is known
2751 to be a valid IP address. However, in the case of IPv6 addresses, it was
2752 not checking this. This is a hostage to fortune. Exim now panics and dies
2753 if the condition is not met. A case was found where this could be provoked
2754 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2755 components; fortuitously, this particular loophole had already been fixed
2756 by change 4.50/55 or 4.44/33 above.
2758 If there are any other similar loopholes, the new check in host_aton()
2759 itself should stop them being exploited. The report I received stated that
2760 data on the command line could provoke the exploit when Exim was running as
2761 exim, but did not say which command line option was involved. All I could
2762 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2763 running as the user.
2766 35. There was a buffer overflow vulnerability in the SPA authentication code
2767 (which came originally from the Samba project). I have added a test to the
2768 spa_base64_to_bits() function which I hope fixes it.
2771 36. The daemon start-up calls getloadavg() while still root for those OS that
2772 need the first call to be done as root, but it missed one case: when
2773 deliver_queue_load_max is set with deliver_drop_privilege. This is
2774 necessary for the benefit of the queue runner, because there is no re-exec
2775 when deliver_drop_privilege is set.
2777 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2778 was in use, was not putting the data itself into the right store pool;
2779 consequently, it could be overwritten for a subsequent message in the same
2780 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2783 38. Sometimes the final signoff response after QUIT could fail to get
2784 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2785 < 0 before doing a fflush(). This bug looks as though it goes back to the
2786 introduction of TLS in release 3.20, but "sometimes" must have been rare
2787 because the tests only now provoked it.
2793 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2794 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2795 Now it returns an integer. A number of calls were still expecting a T/F
2796 return. Fortuitously, in all cases, the tests worked in OK situations,
2797 which is the norm. However, things would have gone wrong on any write
2798 failures on the smtp file descriptor. This function is used when sending
2799 messages over SMTP and also when doing verify callouts.
2801 2. When Exim is called to do synchronous delivery of a locally submitted
2802 message (the -odf or -odi options), it no longer closes stderr before doing
2805 3. Implemented the mua_wrapper option.
2807 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2809 5. Implemented the functions header_remove(), header_testname(),
2810 header_add_at_position(), and receive_remove_recipient(), and exported them
2813 6. If an ACL "warn" statement specified the addition of headers, Exim already
2814 inserted X-ACL-Warn: at the start if there was no header name. However, it
2815 was not making this test for the second and subsequent header lines if
2816 there were newlines in the string. This meant that an invalid header could
2817 be inserted if Exim was badly configured.
2819 7. Allow an ACL "warn" statement to add header lines at the start or after all
2820 the Received: headers, as well as at the end.
2822 8. Added the rcpt_4xx retry error code.
2824 9. Added postmaster_mailfrom=xxx to callout verification option.
2826 10. Added mailfrom=xxxx to the callout verification option, for verify=
2829 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2830 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2832 12. Inserted some casts to stop certain compilers warning when using pointer
2833 differences as field lengths or precisions in printf-type calls (mostly
2834 affecting debugging statements).
2836 13. Added optional readline() support for -be (dynamically loaded).
2838 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2839 same clock tick as a message's arrival, so that its received time was the
2840 same as the "first fail" time on the retry record, and that message
2841 remained on the queue past the ultimate address timeout, every queue runner
2842 would try a delivery (because it was past the ultimate address timeout) but
2843 after another failure, the ultimate address timeout, which should have then
2844 bounced the address, did not kick in. This was a "< instead of <=" error;
2845 in most cases the first failure would have been in the next clock tick
2846 after the received time, and all would be well.
2848 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2849 being recognized when the domain list was tested by the match_domain
2850 condition in an expansion string.
2852 16. Added the ${str2b64: operator.
2854 17. Exim was always calling setrlimit() to set a large limit for the number of
2855 processes, without checking whether the existing limit was already
2856 adequate. (It did check for the limit on file descriptors.) Furthermore,
2857 errors from getrlimit() and setrlimit() were being ignored. Now they are
2858 logged to the main and panic logs, but Exim does carry on, to try to do its
2859 job under whatever limits there are.
2861 18. Imported PCRE 5.0.
2863 19. Trivial typo in log message " temporarily refused connection" (the leading
2866 20. If the log selector return_path_on_delivery was set and an address was
2867 redirected to /dev/null, the delivery process crashed because it assumed
2868 that a return path would always be set for a "successful" delivery. In this
2869 case, the whole delivery is bypassed as an optimization, and therefore no
2872 21. Internal re-arrangement: the function for sending a challenge and reading
2873 a response while authentication was assuming a zero-terminated challenge
2874 string. It's now changed to take a pointer and a length, to allow for
2875 binary data in such strings.
2877 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2879 23. Exim was not respecting finduser_retries when seeking the login of the
2880 uid under which it was called; it was always trying 10 times. (The default
2881 setting of finduser_retries is zero.) Also, it was sleeping after the final
2882 failure, which is pointless.
2884 24. Implemented tls_on_connect_ports.
2886 25. Implemented acl_smtp_predata.
2888 26. If the domain in control=submission is set empty, Exim assumes that the
2889 authenticated id is a complete email address when it generates From: or
2890 Sender: header lines.
2892 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2893 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2894 chown and chgrp in /bin and hostname in /usr/bin.
2896 28. Exim was keeping the "process log" file open after each use, just as it
2897 does for the main log. This opens the possibility of it remaining open for
2898 long periods when the USR1 signal hits a daemon. Occasional processlog
2899 errors were reported, that could have been caused by this. Anyway, it seems
2900 much more sensible not to leave this file open at all, so that is what now
2903 29. The long-running daemon process does not normally write to the log once it
2904 has entered its main loop, and it closes the log before doing so. This is
2905 so that log files can straightforwardly be renamed and moved. However,
2906 there are a couple of unusual error situations where the daemon does write
2907 log entries, and I had neglected to close the log afterwards.
2909 30. The text of an SMTP error response that was received during a remote
2910 delivery was being truncated at 512 bytes. This is too short for some of
2911 the long messages that one sometimes sees. I've increased the limit to
2914 31. It is now possible to make retry rules that apply only when a message has a
2915 specific sender, in particular, an empty sender.
2917 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2918 it possible to be selective about when SMTP synchronization is enforced.
2920 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2922 32. Implemented hosts_connection_nolog.
2924 33. Added an ACL for QUIT.
2926 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2929 35. Added mailbox_size and mailbox_filecount to appendfile.
2931 36. Added control = no_multiline_responses to ACLs.
2933 37. There was a bug in the logic of the code that waits for the clock to tick
2934 in the case where the clock went backwards by a substantial amount such
2935 that the microsecond fraction of "now" was more than the microsecond
2936 fraction of "then" (but the whole seconds number was less).
2938 38. Added support for the libradius Radius client library this is found on
2939 FreeBSD (previously only the radiusclient library was supported).
2945 1. When certain lookups returned multiple values in the form name=value, the
2946 quoting of the values was not always being done properly. Specifically:
2947 (a) If the value started with a double quote, but contained no whitespace,
2949 (b) If the value contained whitespace other than a space character (i.e.
2950 tabs or newlines or carriage returns) it was not quoted.
2951 This fix has been applied to the mysql and pgsql lookups by writing a
2952 separate quoting function and calling it from the lookup code. The fix
2953 should probably also be applied to nisplus, ibase and oracle lookups, but
2954 since I cannot test any of those, I have not disturbed their existing code.
2956 2. A hit in the callout cache for a specific address caused a log line with no
2957 reason for rejecting RCPT. Now it says "Previous (cached) callout
2958 verification failure".
2960 3. There was an off-by-one bug in the queryprogram router. An over-long
2961 return line was truncated at 256 instead of 255 characters, thereby
2962 overflowing its buffer with the terminating zero. As well as fixing this, I
2963 have increased the buffer size to 1024 (and made a note to document this).
2965 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2966 when Exim is waiting for an SMTP response from a remote server, Exim
2967 restarts its select() call on the socket, thereby resetting its timeout.
2968 This is not a problem when such interrupts are rare. Somebody set up a cron
2969 job to run exiwhat every 2 minutes, which is less than the normal select()
2970 timeout (5 or 10 minutes). This meant that the select() timeout never
2971 kicked in because it was always reset. I have fixed this by comparing the
2972 time when an interrupt arrives with the time at the start of the first call
2973 to select(). If more time than the timeout has elapsed, the interrupt is
2974 treated as a timeout.
2976 5. Some internal re-factoring in preparation for the addition of Sieve
2977 extensions (by MH). In particular, the "personal" test is moved to a
2978 separate function, and given an option for scanning Cc: and Bcc: (which is
2979 not set for Exim filters).
2981 6. When Exim created an email address using the login of the caller as the
2982 local part (e.g. when creating a From: or Sender: header line), it was not
2983 quoting the local part when it contained special characters such as @.
2985 7. Installed new OpenBSD configuration files.
2987 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2988 try to make them clearer.
2990 9. Callout options, other than the timeout value, were being ignored when
2991 verifying sender addresses in header lines. For example, when using
2993 verify = header_sender/callout=no_cache
2995 the cache was (incorrectly) being used.
2997 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2998 only the Cygwin environment.
3000 11. When return_path_on_delivery was set as a log selector, if different remote
3001 addresses in the same message used different return paths and parallel
3002 remote delivery occurred, the wrong values would sometimes be logged.
3003 (Whenever a remote delivery process finished, the return path value from
3004 the most recently started remote delivery process was logged.)
3006 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3007 lines when AUTH and/or TLS are in use. This is the "received protocol"
3008 field. Exim used to use "asmtp" for authenticated SMTP, without any
3009 indication (in the protocol name) for TLS use. Now it follows the RFC and
3010 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3011 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3012 names appear in log lines as well as in Received: header lines.
3014 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3015 extensions, and comparison tests, and to fix some bugs.
3017 14. Changes to the "personal" filter test:
3019 (1) The test was buggy in that it was just doing the equivalent of
3020 "contains" tests on header lines. For example, if a user's address was
3021 anne@some.where, the "personal" test would incorrectly be true for
3023 To: susanne@some.where
3025 This test is now done by extracting each address from the header in turn,
3026 and checking the entire address. Other tests that are part of "personal"
3027 are now done using regular expressions (for example, to check local parts
3028 of addresses in From: header lines).
3030 (2) The list of non-personal local parts in From: addresses has been
3031 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3032 taken from the Sieve specification recommendations.
3034 (3) If the message contains any header line starting with "List-" it is
3035 treated as non-personal.
3037 (4) The test for "circular" in the Subject: header line has been removed
3038 because it now seems ill-conceived.
3040 15. Minor typos in src/EDITME comments corrected.
3042 16. Installed latest exipick from John Jetmore.
3044 17. If headers_add on a router specified a text string that was too long for
3045 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3046 of string_sprintf() is now avoided.
3048 18. $message_body_size was not set (it was always zero) when running the DATA
3049 ACL and the local_scan() function.
3051 19. For the "mail" command in an Exim filter, no default was being set for
3052 the once_repeat time, causing a random time value to be used if "once" was
3053 specified. (If the value happened to be <= 0, no repeat happened.) The
3054 default is now 0s, meaning "never repeat". The "vacation" command was OK
3055 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3056 (I found it when inspecting the code).
3058 20. There is now an overall timeout for performing a callout verification. It
3059 defaults to 4 times the callout timeout, which applies to individual SMTP
3060 commands during the callout. The overall timeout applies when there is more
3061 than one host that can be tried. The timeout is checked before trying the
3062 next host. This prevents very long delays if there are a large number of
3063 hosts and all are timing out (e.g. when the network connections are timing
3064 out). The value of the overall timeout can be changed by specifying an
3065 additional sub-option for "callout", called "maxwait". For example:
3067 verify = sender/callout=5s,maxwait=20s
3069 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3070 to the end before writing, but this should make it even safer).
3072 22. Exim was forgetting that it had advertised PIPELINING for the second and
3073 subsequent messages on an SMTP connection. It was also not resetting its
3074 memory on STARTTLS and an internal HELO.
3076 23. When Exim logs an SMTP synchronization error within a session, it now
3077 records whether PIPELINING has been advertised or not.
3079 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3080 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3081 rather than long int.
3083 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3085 26. Added the never_mail option to autoreply.
3091 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3092 crash if the getsockname() call failed; this can happen if a connection is
3093 closed very soon after it is established. The problem was simply in the
3094 order in which certain operations were done, causing Exim to try to write
3095 to the SMTP stream before it had set up the file descriptor. The bug has
3096 been fixed by making things happen in the correct order.
3102 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3103 before the connection was closed, thus losing the rejection response.
3105 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3106 some early Solaris releases, but causes trouble in current releases where
3107 socklen_t is defined.
3109 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3112 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3115 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3116 that releasing the top of it at the end releases what was used for sub-
3117 expansions (unless the block got too big). However, discard this block if
3118 the first thing is a variable or header, so that we can use its block when
3119 it is dynamic (useful for very large $message_headers, for example).
3121 6. Lookups now cache *every* query, not just the most recent. A new, separate
3122 store pool is used for this. It can be recovered when all lookup caches are
3123 flushed. Lookups now release memory at the end of their result strings.
3124 This has involved some general refactoring of the lookup sources.
3126 7. Some code has been added to the store_xxx() functions to reduce the amount
3127 of flapping under certain conditions.
3129 8. log_incoming_interface used to affect only the <= reception log lines. Now
3130 it causes the local interface and port to be added to several more SMTP log
3131 lines, for example "SMTP connection from", and rejection lines.
3133 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3135 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3137 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3138 could be overwritten at the end of the current message (or the start of a
3139 new message if it was set in a HELO ACL). The value is now preserved for
3140 the duration of the SMTP connection.
3142 12. If a transport had a headers_rewrite setting, and a matching header line
3143 contained an unqualified address, that address was qualified, even if it
3144 did not match any rewriting rules. The underlying bug was that the values
3145 of the flags that permit the existence of unqualified sender and recipient
3146 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3147 non-local messages, and by -bnq for local messages) were not being
3148 preserved with the message after it was received.
3150 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3151 "next input=" as part of the text comprising the host identity instead of
3152 the correct text. The code was using the same buffer for two different
3153 strings. However, depending on which order the printing function evaluated
3154 its arguments, the bug did not always show up. Under Linux, for example, my
3155 test suite worked just fine.
3157 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3158 doesn't work with some older versions of Perl. It has been changed to "my",
3159 which in any case is probably the better facility to use.
3161 15. A really picky compiler found some instances of statements for creating
3162 error messages that either had too many or two few arguments for the format
3165 16. The size of the buffer for calls to the DNS resolver has been increased
3166 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3167 for addresses that have a lot of PTR records. This alleviates a problem; it
3168 does not fully solve it.
3170 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3171 buffer now truncates the list and logs the incident, which is the same
3172 action as happens when Exim is looking up a host name and its aliases.
3173 Previously in this situation something unpredictable would happen;
3174 sometimes it was "internal error: store_reset failed".
3176 18. If a server dropped the connection unexpectedly when an Exim client was
3177 using GnuTLS and trying to read a response, the client delivery process
3178 crashed while trying to generate an error log message.
3180 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3181 string, for example:
3183 warn message = H1: something\nH2: something
3185 the text was added as a single header line from Exim's point of view
3186 though it ended up OK in the delivered message. However, searching for the
3187 second and subsequent header lines using $h_h2: did not work. This has been
3188 fixed. Similarly, if a system filter added multiple headers in this way,
3189 the routers could not see them.
3191 20. Expanded the error message when iplsearch is called with an invalid key to
3192 suggest using net-iplsearch in a host list.
3194 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3195 ACLs are no longer actually imposed (and a message to that effect is
3198 22. If a "gecos" field in a passwd entry contained escaped characters, in
3199 particular, if it contained a \" sequence, Exim got it wrong when building
3200 a From: or a Sender: header from that name. A second bug also caused
3201 incorrect handling when an unquoted " was present following a character
3202 that needed quoting.
3204 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3205 was not being matched caselessly.
3207 24. Arranged for all hyphens in the exim.8 source to be escaped with
3210 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3211 properly. Recipient callout cache records were still being keyed to include
3212 the sender, even when use_sender was set false. This led to far more
3213 callouts that were necessary. The sender is no longer included in the key
3214 when use_sender is false.
3216 26. Added "control = submission" modifier to ACLs.
3218 27. Added the ${base62d: operator to decode base 62 numbers.
3220 28. dnsdb lookups can now access SRV records.
3222 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3223 the configuration file.
3225 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3226 (-v) mode. This makes the output for a verbose queue run more intelligible.
3228 31. Added a use_postmaster feature to recipient callouts.
3230 32. Added the $body_zerocount variable, containing the number of binary zero
3231 bytes in the message body.
3233 33. The time of last modification of the "new" subdirectory is now used as the
3234 "mailbox time last read" when there is a quota error for a maildir
3237 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3239 35. Added +ignore_unknown as a special item in host lists.
3241 36. Code for decoding IPv6 addresses in host lists is now included, even if
3242 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3243 address was recognized as an IP address, but was then not correctly decoded
3244 into binary, causing unexpected and incorrect effects when compared with
3251 1. Very minor rewording of debugging text in manualroute to say "list of
3252 hosts" instead of "hostlist".
3254 2. If verify=header_syntax was set, and a header line with an unqualified
3255 address (no domain) and a large number of spaces between the end of the
3256 name and the colon was received, the reception process suffered a buffer
3257 overflow, and (when I tested it) crashed. This was caused by some obsolete
3258 code that should have been removed. The fix is to remove it!
3260 3. When running in the test harness, delay a bit after writing a bounce
3261 message to get a bit more predictability in the log output.
3263 4. Added a call to search_tidyup() just before forking a reception process. In
3264 theory, someone could use a lookup in the expansion of smtp_accept_max_
3265 per_host which, without the tidyup, could leave open a database connection.
3267 5. Added the variables $recipient_data and $sender_data which get set from a
3268 lookup success in an ACL "recipients" or "senders" condition, or a router
3269 "senders" option, similar to $domain_data and $local_part_data.
3271 6. Moved the writing of debug_print from before to after the "senders" test
3274 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3275 problems for message scanning, either using a data ACL, or using
3276 local_scan() because the Received: header was not generated till after they
3277 were called (in order to set the time as the time of reception completion).
3278 I have revised the way this works. The header is now generated after the
3279 body is received, but before the ACL or local_scan() are called. After they
3280 are run, the timestamp in the header is updated.
3286 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3287 before starting a queue runner without re-exec. This happened only when
3288 deliver_drop_privilege was set or when the Exim user was set to root. The
3289 effect of the bug was that timeouts during subsequent deliveries caused
3290 crashes instead of being properly handled. The handler is now left at its
3291 default (and expected) setting.
3293 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3294 message, again when deliver_drop_privilege is set or Exim is run as root.
3295 The bug described in (1) was not present in this case, but the tidying up
3296 of the other signals was missing. I have made the two cases consistent.
3298 3. The ignore_target_hosts setting on a manualroute router was being ignored
3299 for hosts that were looked up using the /MX notation.
3301 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3304 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3305 operated on the sender address. After changing the $sender_address to <>
3306 for the sender address verify, Exim was re-instated it as the original
3307 (before rewriting) address, but remembering that it had rewritten it, so it
3308 wasn't rewriting it again. This bug also had the effect of breaking the
3309 sender address verification caching when the sender address was rewritten.
3311 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3312 This has been changed so that if the ip literal address matches
3313 ignore_target_hosts, the router declines.
3315 7. Added expansion conditions match_domain, match_address, and match_local_
3316 part (NOT match_host).
3318 8. The placeholder for the Received: header didn't have a length field set.
3320 9. Added code to Exim itself and to exim_lock to test for a specific race
3321 condition that could lead to file corruption when using MBX delivery. The
3322 issue is with the lockfile that is created in /tmp. If this file is removed
3323 after a process has opened it but before that process has acquired a lock,
3324 there is the potential for a second process to recreate the file and also
3325 acquire a lock. This could lead to two Exim processes writing to the file
3326 at the same time. The added code performs the same test as UW imapd; it
3327 checks after acquiring the lock that its file descriptor still refers to
3328 the same named file.
3330 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3331 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3334 11. Added the smtp_active_hostname option. If used, this will typically be made
3335 to depend on the incoming interface address. Because $interface_address is
3336 not set up until the daemon has forked a reception process, error responses
3337 that can happen earlier (such as "too many connections") no longer contain
3340 12. If an expansion in a condition on a "warn" statement fails because a lookup
3341 defers, the "warn" statement is abandoned, and the next ACL statement is
3342 processed. Previously this caused the whole ACL to be aborted.
3344 13. Added the iplsearch lookup type.
3346 14. Added ident_timeout as a log selector.
3348 15. Added tls_certificate_verified as a log selector.
3350 16. Added a global option tls_require_ciphers (compare the smtp transport
3351 option of the same name). This controls incoming TLS connections.
3353 17. I finally figured out how to make tls_require_ciphers do a similar thing
3354 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3355 before starting the TLS session.
3357 18. Tabs are now shown as \t in -bP output.
3359 19. If the log selector return_path_on_delivery was set, Exim crashed when
3360 bouncing a message because it had too many Received: header lines.
3362 20. If two routers both had headers_remove settings, and the first one included
3363 a superfluous trailing colon, the final name in the first list and the
3364 first name in the second list were incorrectly joined into one item (with a
3365 colon in the middle).
3371 1. Added -C and -D options to the exinext utility, mainly to make it easier
3372 to include in the automated testing, but these could be helpful when
3373 multiple configurations are in use.
3375 2. The exinext utility was not formatting the output nicely when there was
3376 an alternate port involved in the retry record key, nor when there was a
3377 message id as well (for retries that were specific to a specific message
3378 and a specific host). It was also confused by IPv6 addresses, because of
3379 the additional colons they contain. I have fixed the IPv4 problem, and
3380 patched it up to do a reasonable job for IPv6.
3382 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3383 delivery, the log line now contains "pipelined" if PIPELINING was used.
3385 4. An SMTP transport process used to panic and die if the bind() call to set
3386 an explicit outgoing interface failed. This has been changed; it is now
3387 treated in the same way as a connect() failure.
3389 5. A reference to $sender_host_name in the part of a conditional expansion
3390 that was being skipped was still causing a DNS lookup. This no longer
3393 6. The def: expansion condition was not recognizing references to header lines
3394 that used bh_ and bheader_.
3396 7. Added the _cache feature to named lists.
3398 8. The code for checking quota_filecount in the appendfile transport was
3399 allowing one more file than it should have been.
3401 9. For compatibility with Sendmail, the command line option
3409 and sets the incoming protocol and host name (for trusted callers). The
3410 host name and its colon can be omitted when only the protocol is to be set.
3411 Note the Exim already has two private options, -pd and -ps, that refer to
3412 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3413 "s", but I don't think that's a major issue.
3415 10. A number of refactoring changes to the code, none of which should affect
3418 (a) The number of logging options was getting close to filling up the
3419 32-bit word that was used as a bit map. I have split them into two classes:
3420 those that are passed in the argument to log_write(), and those that are
3421 only ever tested independently outside of that function. These are now in
3422 separate 32-bit words, so there is plenty of room for expansion again.
3423 There is no change in the user interface or the logging behaviour.
3425 (b) When building, for example, log lines, the code previously used a
3426 macro that called string_cat() twice, in order to add two strings. This is
3427 not really sufficiently general. Furthermore, there was one instance where
3428 it was actually wrong because one of the argument was used twice, and in
3429 one call a function was used. (As it happened, calling the function twice
3430 did not affect the overall behaviour.) The macro has been replaced by a
3431 function that can join an arbitrary number of extra strings onto a growing
3434 (c) The code for expansion conditions now uses a table and a binary chop
3435 instead of a serial search (which was left over from when there were very
3436 few conditions). Also, it now recognizes conditions like "pam" even when
3437 the relevant support is not compiled in: a suitably worded error message is
3438 given if an attempt is made to use such a condition.
3440 11. Added ${time_interval:xxxxx}.
3442 12. A bug was causing one of the ddress fields not to be passed back correctly
3443 from remote delivery subprocesses. The field in question was not being
3444 subsequently used, so this caused to problems in practice.
3446 13. Added new log selectors queue_time and deliver_time.
3448 14. Might have fixed a bug in maildirsizefile handling that threw up
3449 "unexpected character" debug warnings, and recalculated the data
3450 unnecessarily. In any case, I expanded the warning message to give more
3453 15. Added the message "Restricted characters in address" to the statements in
3454 the default ACL that block characters like @ and % in local parts.
3456 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3457 Three changes have been made:
3459 (a) There was a serious bug; a negative response to MAIL caused the whole
3460 recipient domain to be cached as invalid, thereby blocking all messages
3461 to all local parts at the same domain, from all senders. This bug has
3462 been fixed. The domain is no longer cached after a negative response to
3463 MAIL if the sender used is not empty.
3465 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3468 (c) A new callout option, "use_sender" has been added for people who want
3469 the modified behaviour.
3475 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3478 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3479 indeed breaks things for older releases.
3481 3. Added additional logging to the case where there is a problem reading data
3482 from a filter that is running in a subprocess using a pipe, in order to
3483 try to track down a specific problem.
3485 4. Testing facility fudge: when running in the test harness and attempting
3486 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3487 getting "No route to host". Convert this to a timeout.
3489 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3492 6. Some OS don't have socklen_t but use size_t instead. This affects the
3493 fifth argument of getsockopt() amongst other things. This is now
3494 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3495 can be set for individual OS. I have set it for SunOS5, OSF1, and
3496 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3497 some earlier ones do not.
3499 7. Change 4.30/15 was not doing the test caselessly.
3501 8. The standard form for an IPv6 address literal was being rejected by address
3502 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3503 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3504 this, as well as the form without the "IPv6" on the front (but only when
3505 address literals are enabled, of course).
3507 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3509 10. Exim crashed if a message with an empty sender address specified by -f
3510 encountered a router with an errors_to setting. This could be provoked only
3511 by a command such as
3515 where an empty string was supplied; "<>" did not hit this bug.
3517 11. Installed PCRE release 4.5.
3519 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3520 remained set. It is now erased.
3522 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3523 times from message ids (which are base 36 rather than the normal 62).
3525 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3526 were being counted as actual protocol errors, and logged if the log
3527 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3528 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3529 and DATA following a set of rejected RCPTs do not count as protocol errors.
3530 In other words, Exim assumes they were pipelined, though this may not
3531 actually be the case. Of course, in all cases the client gets an
3532 appropriate error code.
3534 15. If a lookup fails in an ACL condition, a message about the failure may
3535 be available; it is used if testing the ACL cannot continue, because most
3536 such messages specify what the cause of the deferral is. However, some
3537 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3538 that caused an old message to be retained and used if a later statement
3539 caused a defer, replacing the real cause of the deferral.
3541 16. If an IP address had so many PTR records that the DNS lookup buffer
3542 was not large enough to hold them, Exim could crash while trying to process
3543 the truncated data. It now detects and logs this case.
3545 17. Further to 4.21/58, another change has been made: if (and only if) the
3546 first line of a message (the first header line) ends with CRLF, a bare LF
3547 in a subsequent header line has a space inserted after it, so as not to
3548 terminate the header.
3550 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3551 unnecessarily, used atoi() instead of strtol(), and didn't check the
3552 termination when getting file sizes from file names by regex.
3554 19. Completely re-implemented the support for maildirsize files, in the light
3555 of a number of problems with the previous contributed implementation
3556 (4.30/29). In particular:
3558 . If the quota is zero, the maildirsize file is maintained, but no quota is
3561 . If the maildir directory does not exist, it is created before any attempt
3562 to write a maildirsize file.
3564 . The quota value in the file is just a cache; if the quota is changed in
3565 the transport, the new value overrides.
3567 . A regular expression is available for excluding directories from the
3570 20. The autoreply transport checks the characters in options that define the
3571 message's headers; it allows continued headers, but it was checking with
3572 isspace() after an embedded newline instead of explicitly looking for a
3575 21. If all the "regular" hosts to which an address was routed had passed their
3576 expiry times, and had not reached their retry times, the address was
3577 bounced, even if fallback hosts were defined. Now Exim should go on to try
3580 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3581 equivalent code in the SMTP transport. Some hosts send humungous responses
3582 to HELO/EHLO, more than 1024 it seems.
3584 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3585 gives compiler warnings in some environments. I've now done it "properly",
3588 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3589 (because of problems with the built-in one) was declared to return uschar *
3590 instead of char *, causing compiler failure.
3592 25. Fixed a file descriptor leak when processing alias/forward files.
3594 26. Fixed a minor format string issue in dbfn.c.
3596 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3598 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3599 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3602 29. When a sender address is verified, it is cached, to save repeating the test
3603 when there is more than one recipient in a message. However, when the
3604 verification involves a callout, it is possible for different callout
3605 options to be set for different recipients. It is too complicated to keep
3606 track of this in the cache, so now Exim always runs a verification when a
3607 callout is required, relying on the callout cache for the optimization.
3608 The overhead is duplication of the address routing, but this should not be
3611 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3612 to be verified with callout=postmaster, and the main callout worked but the
3613 postmaster check failed, the verification correctly failed. However, if a
3614 subsequent RCPT command asked for sender verification *without* the
3615 postmaster check, incorrect caching caused this verification also to fail,
3618 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3619 it was not caching the DNS options (qualify_single, search_parents) that
3620 were used when the lookup failed. A subsequent lookup with different
3621 options therefore always gave the same answer, though there were cases
3622 where it should not have. (Example: a "domains = !$mx_any" option on a
3623 dnslookup router: the "domains" option is always processed without any
3624 widening, but the router might have qualify_single set.) Now Exim uses the
3625 cached value only when the same options are set.
3627 32. Added John Jetmore's "exipick" utility to the distribution.
3629 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3630 than a timeout (e.g. a certificate is required, and is not provided), an
3631 Exim server now closes the connection immediately. Previously it waited for
3632 the client to close - but if the client is SSL, it seems that they each
3633 wait for each other, leading to a delay before one of them times out.
3635 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3636 maintained 0.8.x compatibility because I don't think many are using it, and
3637 it is clearly obsolete.
3639 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3642 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3643 client certificate was expired. A simple patch fixes this, though I don't
3644 understand the full logic of why the verify callback is called multiple
3647 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3648 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3649 which causes problems with some clients (such as the Certicom SSL Plus
3650 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3651 disables the coutermeasure allowing Eudora to connect."
3653 38. Exim was not checking that a write() to a log file succeeded. This could
3654 lead to Bad Things if a log got too big, in particular if it hit a file
3655 size limit. Exim now panics and dies if it cannot write to a log file, just
3656 as it does if it cannot open a log file.
3658 39. Modified OS/Makefile-Linux so that it now contains
3660 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3662 The two -D definitions ensure that Exim is compiled with large file
3663 support, which makes it possible to handle log files that are bigger than
3666 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3667 instance) a domain was checked against a named list that involved a lookup,
3668 causing $domain_data to be set, then another domain was checked against the
3669 same list, then the first domain was re-checked, the value of $domain_data
3670 after the final check could be wrong. In particular, if the second check
3671 failed, it could be set empty. This bug probably also applied to
3674 41. The strip_trailing_dot option was not being applied to the address given
3675 with the -f command-line option.
3677 42. The code for reading a message's header from the spool was incrementing
3678 $received_count, but never initializing it. This meant that the value was
3679 incorrect (doubled) while delivering a message in the same process in which
3680 it was received. In the most common configuration of Exim, this never
3681 happens - a fresh exec is done - but it can happen when
3682 deliver_drop_privilege is set.
3684 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3685 it now includes up to 150 characters of the unexpected data in the log
3688 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3689 and building data strings. The size of both of these buffers was 10 000
3690 bytes - far larger than anybody would *ever* want, thought I. Needless to
3691 say, somebody hit the limit. I have increased the maximum line length to
3692 20 000 and the maximum data length of concatenated lines to 100 000. I have
3693 also fixed two bugs, because there was no checking on these buffers. Tsk,
3694 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3695 buffer is too small.
3697 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3698 lsearch lookups. Now it does.
3700 46. When parsing a route_list item in a manualroute router, a fixed-length
3701 buffer was used for the list of hosts. I made this 1024 bytes long,
3702 thinking that nobody would ever have a list of hosts that long. Wrong.
3703 Somebody had a whole pile of complicated expansion conditions, and the
3704 string was silently truncated, leading to an expansion error. It turns out
3705 that it is easier to change to an unlimited length (owing to other changes
3706 that have happened since this code was originally written) than to build
3707 structure for giving a limitation error. The length of the item that
3708 expands into the list of hosts is now unlimited.
3710 47. The lsearch lookup could not handle data where the length of text line was
3711 more than 4095 characters. Such lines were truncated, leading to shortened
3712 data being returned. It should now handle lines of any length.
3714 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3715 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3718 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3719 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3720 no longer happens in comment lines. A list of replacements is now placed
3721 at the head of all of the source files, except those whose only change is
3722 to replace PERL_COMMAND in the very first #! line.
3724 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3725 messages on the queue, with a bottom-up merge sort, using code contributed
3726 by Michael Haardt. This should make operations like -bp somewhat faster on
3727 large queues. It won't affect queue runners, except when queue_run_in_order
3730 51. Installed eximstats 1.31 in the distribution.
3732 52. Added support for SRV lookups to the dnslookup router.
3734 53. If an ACL referred to $message_body or $message_body_end, the value was not
3735 reset for any messages that followed in the same SMTP session.
3737 54. The store-handling optimization for building very long strings was not
3738 differentiating between the different store pools. I don't think this
3739 actually made any difference in practice, but I've tidied it.
3741 55. While running the routers to verify a sender address, $sender_address
3742 was still set to the sender address. This is wrong, because when routing to
3743 send a bounce to the sender, it would be empty. Therefore, I have changed
3744 it so that, while verifying a sender address, $sender_address is set to <>.
3745 (There is no change to what happens when verifying a recipient address.)
3747 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3748 target A or AAAA records (if not already returned) without resetting the
3749 qualify_single or search_parents options of the DNS resolver. These are
3750 inappropriate in this case because the targets of MX and SRV records must
3751 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3752 target that, when qualified, matched something in the local domain. These
3753 two options are now turned off when doing these lookups.
3755 57. It seems that at least some releases of Reiserfs (which does not have the
3756 concept of a fixed number of inodes) returns zero and not -1 for the
3757 number of available inodes. This interacted badly with check_spool_inodes,
3758 which assumed that -1 was the "no such thing" setting. What I have done is
3759 to check that the total number of inodes is greater than zero before doing
3760 the test of how many are available.
3762 58. When a "warn" ACL statement has a log_message modifier, the message is
3763 remembered, and not repeated. This is to avoid a lot of repetition when a
3764 message has many recipients that cause the same warning to be written.
3765 Howewer, Exim was preserving the list of already written lines for an
3766 entire SMTP session, which doesn't seem right. The memory is now reset if a
3767 new message is started.
3769 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3770 debugging output unless log_rewrite was also set.
3772 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3773 of (int)(handle) when we know that handle contains (void *)(-1).
3775 61. The Exim daemon panic-logs an error return when it closes the incoming
3776 connection. However "connection reset by peer" seems to be common, and
3777 isn't really an error worthy of noting specially, so that particular error
3780 62. When Exim is trying to find all the local interfaces, it used to panic and
3781 die if the ioctl to get the interface flags failed. However, it seems that
3782 on at least one OS (Solaris 9) it is possible to have an interface that is
3783 included in the list of interfaces, but for which you get a failure error
3784 for this call. This happens when the interface is not "plumbed" into a
3785 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3786 failure of the "get flags" call assumes that the interface is down.
3788 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3789 makes life easier for people who are doing arithmetic on fields extracted
3790 from dates, where you often get leading zeros that should not be
3791 interpreted as octal.
3793 64. Added qualify_domain to the redirect router, to override the global
3796 65. If a pathologically long header line contained very many addresses (the
3797 report of this problem mentioned 10 000) and each of them was rewritten,
3798 Exim could use up a very large amount of memory. (It kept on making new
3799 copies of the header line as it rewrote, and never released the old ones.)
3800 At the expense of a bit more processing, the header rewriting function has
3801 been changed so that it no longer eats memory in this way.
3803 66. The generation of the Received: header has been moved from the time that a
3804 message starts to be received, to the time that it finishes. The timestamp
3805 in the Received: header should now be very close to that of the <= log
3806 line. There are two side-effects of this change:
3808 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3809 logged header lines no longer include the local Received: line, because
3810 it has not yet been created. The same applies to a copy of the message
3811 that is returned to a non-SMTP sender when a message is rejected.
3813 (b) When a filter file is tested using -bf, no additional Received: header
3814 is added to the test message. After some thought, I decided that this
3817 This change does not affect the value of $received_for. It is still set
3818 after address rewriting, but before local_scan() is called.
3820 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3822 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3823 gave an unhelpful panic error message, and a defer error. I have managed to
3824 change this behaviour so that it now rejects any supplied certificate,
3825 which seems right, as the list of acceptable certificates is empty.
3827 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3828 gave an unhelpful defer error. I have not managed to make this reject any
3829 supplied certificates, but the error message it gives is "no certificate
3830 supplied", which is not helpful.
3832 70. exigrep's output now also includes lines that are not associated with any
3833 message, but which match the given pattern. Implemented by a patch from
3834 Martin Sluka, which also tidied up the Perl a bit.
3836 71. Recipient callout verification, like sender verification, was using <> in
3837 the MAIL FROM command. This isn't really the right thing, since the actual
3838 sender may affect whether the remote host accepts the recipient or not. I
3839 have changed it to use the actual sender in the callout; this means that
3840 the cache record is now keyed on a recipient/sender pair, not just the
3841 recipient address. There doesn't seem to be a real danger of callout loops,
3842 since a callout by the remote host to check the sender would use <>.
3843 [SEE ABOVE: changed after hitting problems.]
3845 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3846 temporary errors. However, in the case of such a code being given after
3847 the end of a data transmission (i.e. after ".") Exim was failing to write
3848 a retry record for the message. (Yes, there was some broken host that was
3849 actually sending 8xx at this point.)
3851 73. An unknown lookup type in a host list could cause Exim to panic-die when
3852 the list was checked. (An example that provoked this was putting <; in the
3853 middle of a list instead of at the start.) If this happened during a DATA
3854 ACL check, a -D file could be left lying around. This kind of configuration
3855 error no longer causes Exim to die; instead it causes a defer errror. The
3856 incident is still logged to the main and panic logs.
3858 74. Buglet left over from Exim 3 conversion. The message "too many messages
3859 in one connection" was written to the rejectlog but not the mainlog, except
3860 when address rewriting (yes!) was being logged.
3862 75. Added write_rejectlog option.
3864 76. When a system filter was run not as root (that is, when system_filter_user
3865 was set), the values of the $n variables were not being returned to the
3866 main process; thus, they were not subsequently available in the $sn
3869 77. Added +return_path_on_delivery log selector.
3871 78. A connection timeout was being treated differently from recipients deferred
3872 when testing hosts_max_try with a message that was older than the host's
3873 retry timeout. (The host should not be counted, thus allowing all hosts to
3874 be tried at least once before bouncing.) This may have been the cause of an
3875 occasionally reported bug whereby a message would remain on the queue
3876 longer than the retry timeout, but would be bounced if a delivery was
3877 forced. I say "may" because I never totally pinned down the problem;
3878 setting up timeout/retry tests is difficult. See also the next item.
3880 79. The ultimate address timeout was not being applied to errors that involved
3881 a combination of host plus message (for example, a timeout on a MAIL
3882 command). When an address resolved to a number of possible hosts, and they
3883 were not all tried for each delivery (e.g. because of hosts_max_try), a
3884 message could remain on the queue longer than the retry timeout.
3886 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3889 81. Fixed an obscure SMTP outgoing bug which required at least the following
3890 conditions: (a) there was another message waiting for the same server;
3891 (b) the server returned 5xx to all RCPT commands in the first message so
3892 that the message was not completed; (c) the server dropped the connection
3893 or gave a negative response to the RSET that Exim sends to abort the
3894 transaction. The observed case was a dropped connection after DATA that had
3895 been sent in pipelining mode. That is, the server had advertised PIPELINING
3896 but was not implementing it correctly. The effect of the bug was incorrect
3897 behaviour, such as trying another host, and this could lead to a crash.
3903 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3904 and daemon.c were passed as pointers to ints; they should have been
3905 pointers to socklen_t variables (which are typically unsigned ints).
3907 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3910 3. Fixed a really odd bug that affected only the testing scheme; patching a
3911 certain fixed string in the binary changed the value of another string that
3912 happened to be identical to the end of the original first string.
3914 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3915 name", it returns that address as the IP address. On some operating
3916 systems (e.g. Solaris), it also passes back the IP address string as the
3917 "host name". However, on others (e.g. Linux), it passes back an empty
3918 string. Exim wasn't checking for this, and was changing the host name to an
3919 empty string, assuming it had been canonicized.
3921 5. Although rare, it is permitted to have more than one PTR record for a given
3922 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3923 all the names associated with an address, because they do in Solaris.
3924 However, it seems that they do not in Linux for data that comes from the
3925 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3926 I found this out when I moved to a new Linux workstation and tried to run
3927 the Exim test suite.
3929 To get round this problem I have changed the code so that it now does its
3930 own call to the DNS to look up PTR records when searching for a host name.
3931 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3932 addresses that are only in /etc/hosts are still found.
3934 This behaviour is, however, controlled by an option called host_lookup_
3935 order, which defaults to "bydns:byaddr". If people want to use the other
3936 order, or indeed, just use one or the other means of lookup, they can
3937 specify it in this variable.
3939 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3940 some operating systems, this comes back from gethostbyaddr() as an empty
3941 string, and this is what Exim used to test for. However, it seems that in
3942 other systems, "." is yielded. Exim now tests for this case too.
3944 7. The values of check_spool_space and check_log_space are now held internally
3945 as a number of kilobytes instead of an absolute number of bytes. If a
3946 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3947 kilobyte. This means that much larger values can be stored.
3949 8. Exim monitor: an attempt to get the action menu when not actually pointing
3950 at a message produces an empty menu entitled "No message selected". This
3951 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3952 no entries in it ("Shell widget menu has zero width and/or height"). So I
3953 have added a single, blank menu entry in this case.
3955 9. Added ${quote_local_part.
3957 10. MIME decoding is now applied to the contents of Subject: header lines when
3960 11. Now that a reference to $sender_host_address automatically causes a reverse
3961 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3962 host lookup before query-style lookups in lists that might use this
3963 variable. This has therefore been abolished, and the "net-" prefix is no
3964 longer necessary for query-style lookups.
3966 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3967 have been a typo for LFLAGS, so it has been changed.
3969 13. The install script calls Exim with "-C /dev/null" in order to find the
3970 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3971 to be output. Howeve, since Exim outputs its version number before the
3972 error, it didn't break the script. It just looked ugly. I fixed this by
3973 always allowing "-C /dev/null" if the caller is root.
3975 14. Ignore overlarge ACL variable number when reading spool file - insurance
3976 against a later release with more variables having written the file.
3978 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3979 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3980 this, as well as the form without the "IPv6" on the front.
3982 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3983 OS/Makefile-Darwin file.
3985 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3986 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3988 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3989 variable. However, some OS use ints for inodes. I've added cast to long int
3990 to get rid of the compiler warning.
3992 19. I had forgotten to lock out "/../" in configuration file names when
3993 ALT_CONFIG_PREFIX was set.
3995 20. Routers used for verification do not need to specify transports. However,
3996 if such a router generated a host list, and callout was configured, Exim
3997 crashed, because it could not find a port number from the (non-existent)
3998 transport. It now assumes port 25 in this circumstance.
4000 21. Added the -t option to exigrep.
4002 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4003 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4004 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4005 them is set, LOOKUP_LSEARCH is forced.
4007 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4009 24. Added sender and host information to the "rejected by local_scan()" log
4010 line; previously there was no indication of these.
4012 25. Added .include_if_exists.
4014 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4015 Linux. It turns out that not all file systems support this. Apparently some
4016 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4017 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4018 not supported on the file descriptor, is now ignored when Exim is trying to
4019 sync a directory. This applies only to Linux.
4021 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4023 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4024 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4027 29. Added support for maildirsize files from supplied patch (modified a bit).
4029 30. The use of :fail: followed by an empty string could lead Exim to respond to
4030 sender verification failures with (e.g.):
4032 550 Verification failed for <xxx>
4033 550 Sender verify failed
4035 where the first response line was missing the '-' that indicates it is not
4036 the final line of the response.
4038 31. The loop for finding the name of the user that called Exim had a hardwired
4039 limit of 10; it now uses the value of finduser_retries, which is used for
4040 all other user lookups.
4042 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4045 33. Exim was neglecting to zero errno before one call of strtol() when
4046 expanding a string and expecting an integer value. On some systems this
4047 resulted in spurious "integer overflow" errors. Also, it was casting the
4048 result into an int without checking.
4050 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4051 did not work. The code looks as if it has *never* worked, though it appears
4052 to have been documented since at least releast 1.62. I have made it work.
4054 35. The "timeout_DNS" error in retry rules, also documented since at least
4055 1.62, also never worked. As it isn't clear exactly what this means, and
4056 clearly it isn't a major issue, I have abolished the feature by treating it
4057 as "timeout", and writing a warning to the main and panic logs.
4059 36. The display of retry rules for -brt wasn't always showing the error code
4062 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4063 timeout_connect_A, timeout_connect_MX.
4065 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4066 to the empty sender.
4068 39. The daemon was not analysing the content of -oX till after it had closed
4069 stderr and disconnected from the controlling terminal. This meant that any
4070 syntax errors were only noted on the panic log, and the return code from
4071 the command was 0. By re-arranging the code a little, I've made the
4072 decoding happen first, so such errors now appear on stderr, and the return
4073 code is 1. However, the actual setting up of the sockets still happens in
4074 the disconnected process, so errors there are still only recorded on the
4077 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4078 connections (as happens on some IP stacks) was logged at start up time as
4079 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4080 it from "IPv6 and IPv4", which means that two separate sockets are being
4083 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4084 says whether AF_INET or AF_INET6 was passed as an argument.
4086 42. Exiwhat output was messed up when time zones were included in log
4089 43. Exiwhat now gives more information about the daemon's listening ports,
4090 and whether -tls-on-connect was used.
4092 44. The "port" option of the smtp transport is now expanded.
4094 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4095 silently ignored. Now an error message is written to the main and panic
4098 46. There's a new ACL modifier called "logwrite" which writes to a log file
4099 as soon as it is encountered.
4101 47. Added $local_user_uid and $local_user_gid at routing time.
4103 48. Exim crashed when trying to verify a sender address that was being
4106 49. Exim was recognizing only a space character after ".include". It now also
4107 recognizes a tab character.
4109 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4110 extracting the relevant information from the specification. The man page no
4111 longer contains scrambled data for the -d option, and I've added a section
4112 at the front about calling Exim under different names.
4114 51. Added "extra_headers" argument to the "mail" command in filter files.
4116 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4119 53. Installed eximstats 1.29.
4121 54. Added transport_filter_timeout as a generic transport option.
4123 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4124 Cc: header lines. This was required by RFC 822, but it not required by RFC
4127 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4128 incoming messages that did not have them. Now it does so only if the
4129 message originates locally, that is, if there is no associated remote host
4130 address. When Resent- header lines are present, this applies to the Resent-
4131 lines rather than the non-Resent- lines.
4133 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4134 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4136 58. Messages for configuration errors now include the name of the main
4137 configuration file - useful now that there may be more than one file in a
4138 list (.included file names were always shown).
4140 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4141 for those rare installations that do not start the daemon as root or run it
4142 setuid root. I've cut out the call to initgroups() if the daemon is not
4145 60. The Exim user and group can now be bound into the binary as text strings
4146 that are looked up at the start of Exim's processing.
4148 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4150 62. Added $mailstore_basename variable.
4152 63. Installed patch to sieve.c from Michael Haardt.
4154 64. When Exim failed to open the panic log after failing to open the main log,
4155 the original message it was trying to log was written to stderr and debug
4156 output, but if they were not available (the usual case in production), it
4157 was lost. Now it is written to syslog before the two lines that record the
4158 failures to open the logs.
4160 65. Users' Exim filters run in subprocesses under the user's uid. It is
4161 possible for a "deliver" command or an alias in a "personal" command to
4162 provoke an address rewrite. If logging of address rewriting is configured,
4163 this fails because the process is not running as root or exim. There may be
4164 a better way of dealing with this, but for the moment (because 4.30 needs
4165 to be released), I have disabled address rewrite logging when running a
4166 filter in a non-root, non-exim process.
4172 1. The buildconfig auxiliary program wasn't quoting the value set for
4173 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4174 not defined. This bug was masked in 4.22 by the effect that was fixed in
4177 2. Some messages that were rejected after a message id was allocated were
4178 shown as "incomplete" by exigrep. It no longer does this for messages that
4179 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4181 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4182 have allow_domain_literals set, the ID did not get logged in the <= line.
4183 Domain literals are now always recognized in Message-ID: header lines.
4185 4. The first argument for a ${extract expansion item is the key name or field
4186 number. Leading and trailing spaces in this item were not being ignored,
4187 causing some misleading effects.
4189 5. When deliver_drop_privilege was set, single queue runner processes started
4190 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4191 same command in the process it spins off) were not dropping privilege.
4193 6. When the daemon running as "exim" started a queue runner, it always
4194 re-executed Exim in the spun-off process. This is a waste of effort when
4195 deliver_drop_privilege is set. The new process now just calls the
4196 queue-runner function directly.
4202 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4205 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4206 ignored. Though the use of -oP was forcing the writing of a pid file, it
4207 was always written to the default place.
4209 3. If the message "no IP address found for host xxxx" is generated during
4210 incoming verification, it is now followed by identification of the incoming
4211 connection (so you can more easily find what provoked it).
4213 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4215 5. Added some features to "harden" Exim a bit more against certain attacks:
4217 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4218 be put in Local/Makefile. This is like the never_users runtime option,
4219 but it cannot be overridden. The default setting is "root".
4221 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4222 prefix string with which any file named in a -C command line option
4225 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4226 is retained for -C and -D only if the caller of Exim is root. Without
4227 it, the exim user may also use -C and -D and retain privilege.
4229 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4230 command line option is disabled.
4232 6. Macro names set by the -D option must start with an upper case letter, just
4233 like macro names defined in the configuration file.
4235 7. Added "dereference=" facility to LDAP.
4237 8. Two instances of the typo "uknown" in the source files are fixed.
4239 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4240 the Configure-Makefile script screwed up while processing it.
4242 10. Incorporated PCRE 4.4.
4244 11. The SMTP synchronization check was not operating right at the start of an
4245 SMTP session. For example, it could not catch a HELO sent before the client
4246 waited for the greeting. There is now a check for outstanding input at the
4247 point when the greeting is written. Because of the duplex, asynchronous
4248 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4249 way, but not yet received, when the check is performed.
4251 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4252 on TCP/IP sockets, because this apparently causes some broken clients to
4255 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4256 unchanged) from the Cygwin maintainer.
4258 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4259 was supported instead of testing for mbx. Effectively a typo.
4261 15. The spa authenticator server code was not checking that the input it
4262 received was valid base64.
4264 16. The debug output line for the "set" modifier in ACLs was not showing the
4265 name of the variable that was being set.
4267 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4269 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4270 lookup on its own. Something else was needed to trigger the lookup. For
4271 example, a match in host_lookup or the need for a host name in a host list.
4272 Now, if $sender_host_name is referenced and the host name has not yet been
4273 looked up, a lookup is performed. If the lookup fails, the variable remains
4274 empty, and $host_lookup_failed is set to "1".
4276 19. Added "eqi" as a case-independent comparison operator.
4278 20. The saslauthd authentication condition could segfault if neither service
4279 nor realm was specified.
4281 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4282 error message that was logged was misleading, and incoming SMTP
4283 connections were dropped. The message is now more accurate, and temporary
4284 errors are given to SMTP connections.
4286 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4287 (except RSET) with a 421 error, until QUIT is received. However, it was
4288 failing to send a response to QUIT.
4290 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4291 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4292 ACL is now run after the helo_try_verify_hosts code.
4294 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4295 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4296 case-independent, so other case variants are also recognized). Apparently
4297 some systems use these upper case variants.
4299 25. If more than two messages were waiting for the same host, and a transport
4300 filter was specified for the transport, Exim sent two messages over the
4301 same TCP/IP connection, and then failed with "socket operation on non-
4302 socket" when it tried to send the third.
4304 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4306 27. The extern definition of crypt16() in expand.c was not being excluded when
4307 the OS had its own crypt16() function.
4309 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4310 as a preferred synonym for return_size_limit, both as an option and as an
4313 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4315 30. Changed the default configuration ACL to relax the local part checking rule
4316 for addresses that are not in any local domains. For these addresses,
4317 slashes and pipe symbols are allowed within local parts, but the sequence
4318 /../ is explicitly forbidden.
4320 31. SPA server authentication was not clearing the challenge buffer before
4323 32. log_message in a "warn" ACL statement was writing to the reject log as
4324 well as to the main log, which contradicts the documentation and doesn't
4325 seem right (because no rejection is happening). So I have stopped it.
4327 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4328 However, I am unable to do any testing of this.
4330 34. Fixed an infelicity in the appendfile transport. When checking directories
4331 for a mailbox, to see if any needed to be created, it was accidentally
4332 using path names with one or more superfluous leading slashes; tracing
4333 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4335 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4336 contents are added to the log line that is written for every discarded
4337 recipient. (Previously a log_message setting was ignored.)
4339 36. The ${quote: operator now quotes the string if it is empty.
4341 37. The install script runs exim in order to find its version number. If for
4342 some reason other than non-existence or emptiness, which it checks, it
4343 could not run './exim', it was installing it with an empty version number,
4344 i.e. as "exim-". This error state is now caught, and the installation is
4347 38. An argument was missing from the function that creates an error message
4348 when Exim fails to connect to the socket for saslauthd authentication.
4349 This could cause Exim to crash, or give a corrupted message.
4351 39. Added isip, isip4, and isip6 to ${if conditions.
4353 40. The ACL variables $acl_xx are now saved with the message, and can be
4354 accessed later in routers, transports, and filters.
4356 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4357 strings in the file are not string-expanded.
4359 42. If a MAIL command specified a SIZE value that was too large to fit into an
4360 int variable, the check against message_size_limit failed. Such values are
4361 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4362 day this will have to be increased, but I don't think I want to be around
4363 when emails are that large.
4370 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4371 iconv() is not standard in FreeBSD.
4373 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4374 IPv6 enabled. The observed symptom was a segmentation fault on return from
4375 the function os_common_find_running_interfaces() in src/os.c.
4377 3. In the check_special_case() function in daemon.c I had used "errno" as an
4378 argument name, which causes warnings on some systems. This was basically a
4379 typo, since it was named "eno" in the comments!
4381 4. The code that waits for the clock to tick (at a resolution of some fraction
4382 of a second) so as to ensure message-id uniqueness was always waiting for
4383 at least one whole tick, when it could have waited for less. [This is
4384 almost certainly not relevant at current processor speeds, where it is
4385 unlikely to ever wait at all. But we try to future-proof.]
4387 5. The function that sleeps for a time interval that includes fractions of a
4388 second contained a race. It did not block SIGALRM between setting the
4389 timer, and suspending (a couple of lines later). If the interval was short
4390 and the sigsuspend() was delayed until after it had expired, the suspension
4391 never ended. On busy systems this could lead to processes getting stuck for
4394 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4395 process, before it forks any delivery processes. The open lookup caching
4396 mechanism meant that the open file or database connection was passed into
4397 the delivery process. The problem was that delivery processes always tidy
4398 up cached lookup data. This could cause a problem for the next delivery
4399 process started by the queue runner, because the external queue runner
4400 process does not know about the closure. So the next delivery process
4401 still has data in the lookup cache. In the case of a file lookup, there was
4402 no problem because closing a file descriptor in a subprocess doesn't affect
4403 the parent. However, if the lookup was caching a connection to a database,
4404 the connection was closed, and the second delivery process was likely to
4405 see errors such as "PGSQL: query failed: server closed the connection
4406 unexpectedly". The problem has been fixed by closing all cached lookups
4407 in a queue runner before running a delivery process.
4409 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4410 seem to have the "const" qualifier which it has on other OS. I've
4413 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4414 *of the same type* (client or server) with the same public name that an
4415 error should be diagnosed.
4417 9. When Exim looked up a host name for an IP address, but failed to find the
4418 original IP address when looking up the host name (a safety check), it
4419 output the message "<ip address> does not match any IP for NULL", which was
4420 confusing, to say the least. The bug was that the host name should have
4421 appeared instead of "NULL".
4423 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4424 user that is built into the binary, and the -C or -D options is used, root
4425 privilege is dropped before the configuration file is read. In addition,
4426 logging is switched to stderr instead of the normal log files. If the
4427 configuration then re-defines the Exim user, the unprivileged environment
4428 is probably not what is expected, so Exim logs a panic warning message (but
4431 However, if deliver_drop_privilege is set, the unprivileged state may well
4432 be exactly what is intended, so the warning has been cut out in that case,
4433 and Exim is allowed to try to write to its normal log files.
4439 1. smtp_return_error_details was not giving details for temporary sender
4440 or receiver verification errors.
4442 2. Diagnose a configuration error if two authenticators have the same public
4445 3. Exim used not to create the message log file for a message until the first
4446 delivery attempt. This could be confusing when incoming messages were held
4447 for policy or load reasons. The message log file is now created at the time
4448 the message is received, and an initial "Received" line is written to it.
4450 4. The automatically generated man page for command line options had a minor
4451 bug that caused no ill effects; however, a more serious problem was that
4452 the procedure for building the man page automatically didn't always
4453 operate. Consequently, release 4.20 contains an out-of-date version. This
4454 shouldn't happen again.
4456 5. When building Exim with embedded Perl support, the script that builds the
4457 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4458 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4460 6. The freeze_tell option was not being used for messages that were frozen on
4461 arrival, either by an ACL or by local_scan().
4463 7. Added the smtp_incomplete_transaction log selector.
4465 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4466 was accepting AUTH without a new EHLO.
4468 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4469 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4472 10. Logging of TCP/IP connections (when configured) now happens in the main
4473 daemon process instead of the child process, so that the TCP/IP connection
4474 count is more accurate (but it can never be perfect).
4476 11. The use of "drop" in a nested ACL was not being handled correctly in the
4477 outer ACL. Now, if condition failure induced by the nested "drop" causes
4478 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4479 or "require"), the connection is dropped.
4481 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4482 that yield "discard" can now be used with an "accept" or a "discard" verb,
4483 but an error is generated for any others (because I can't see a useful way
4484 to define what should happen).
4486 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4487 are now processed in the same way as lines in the Exim configuration file.
4488 In particular, continuation lines are supported.
4490 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4492 15. Added -ti meaning -t -i.
4494 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4495 domains, and warn by logging if others are found.
4497 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4498 returned from GIFCONF when Exim is trying to find the list of interfaces on
4499 a host. The code in os.c has been modified to copy each ifreq to an aligned
4500 structure in all cases.
4502 Also, in some cases, the returned ifreq's were being copied to a 'struct
4503 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4504 means the last couple of bytes of an IPv6 address could be chopped if the
4505 ifreq contained only a normal sockaddr (14 bytes storage).
4507 18. Named domain lists were not supported in the hosts_treat_as_local option.
4508 An entry such as +xxxx was not recognized, and was treated as a literal
4511 19. Ensure that header lines added by a DATA ACL are included in the reject log
4512 if the ACL subsequently rejects the message.
4514 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4515 MD5 (which is deprecated).
4517 21. When testing a filter file using -bf, Exim was writing a message when it
4518 took the sender from a "From " line in the message, but it was not doing so
4519 when it took $return_path from a Return-Path: header line. It now does.
4521 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4522 with a valid header line field name (a series of printing characters
4523 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4525 23. Changed "disc" in the source to "disk" to conform to the documentation and
4526 the book and for uniformity.
4528 24. Ignore Sendmail's -Ooption=value command line item.
4530 25. When execve() failed while trying to run a command in a pipe transport,
4531 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4532 could be confused with a return value of 69 from the command itself. This
4533 has been changed to 127, the value the shell returns if it is asked to run
4534 a non-existent command. The wording for the related log line suggests a
4535 non-existent command as the problem.
4537 26. If received_header_text expands to an empty string, do not add a Received:
4538 header line to the message. (Well, it adds a token one on the spool, but
4539 marks it "old" so that it doesn't get used or transmitted.)
4541 27. Installed eximstats 1.28 (addition of -nt option).
4543 28. There was no check for failure on the call to getsockname() in the daemon
4544 code. This can fail if there is a shortage of resources on the system, with
4545 ENOMEM, for example. A temporary error is now given on failure.
4547 29. Contrary to the C standard, it seems that in some environments, the
4548 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4549 program. Exim now does this explicitly; it affects the formatting of
4550 timestamps using strftime().
4552 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4553 complaints. I've now initialized all the variables, to avoid this.
4555 32. Header lines added by a system filter were not being "seen" during
4556 transport-time rewrites.
4558 33. The info_callback() function passed to OpenSSL is set up with type void
4559 (*)(SSL *, int, int), as described somewhere. However, when calling the
4560 function (actually a macro) that sets it up, the type void(*)() is
4561 expected. I've put in a cast to prevent warnings from picky compilers.
4563 34. If a DNS black list lookup found a CNAME record, but there were no A
4564 records associated with the domain it pointed at, Exim crashed.
4566 35. If a DNS black list lookup returned more than one A record, Exim ignored
4567 all but the first. It now scans all returned addresses if a particular IP
4568 value is being sought. In this situation, the contents of the
4569 $dnslist_value variable are a list of all the addresses, separated by a
4572 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4573 to accept a host name and all its aliases if the forward lookup for any of
4574 them yielded the IP address of the incoming connection. Now it accepts only
4575 those names whose forward lookup yields the correct IP address. Any other
4576 names are discarded. This closes a loophole whereby a rogue DNS
4577 administrator could create reverse DNS records to break through a
4578 wildcarded host restriction in an ACL.
4580 37. If a user filter or a system filter that ran in a subprocess used any of
4581 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4582 the wrong values were passed to the pipe command ($thisaddress had the
4583 value of $0, $0 had the value of $1, etc). This bug was introduced by
4584 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4586 38. Improved the line breaking for long SMTP error messages from ACLs.
4587 Previously, if there was no break point between 40 and 75 characters, Exim
4588 left the rest of the message alone. Two changes have been made: (a) I've
4589 reduced the minimum length to 35 characters; (b) if it can't find a break
4590 point between 35 and 75 characters, it looks ahead and uses the first one
4591 that it finds. This may give the occasional overlong line, but at least the
4592 remaining text gets split now.
4594 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4595 file descriptors might be low, and that setting 1000 would always raise it.
4596 It turns out that in some environments, the limit is already over 1000 and
4597 that lowering it causes trouble. So now Exim takes care not to decrease it.
4599 40. When delivering a message, the value of $return_path is set to $sender_
4600 address at the start of routing (routers may change the value). By an
4601 oversight, this default was not being set up when an address was tested by
4602 -bt or -bv, which affected the outcome if any router or filter referred to
4605 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4606 log only when "message" or "log_message" are set. However, if one of the
4607 conditions was an address verification, or a call to a nested ACL, the
4608 messages generated by the underlying test were being passed through. This
4609 no longer happens. The underlying message is available in $acl_verify_
4610 message for both "message" and "log_message" expansions, so it can be
4611 passed through if needed.
4613 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4614 new expansion $bh_ to give the encoded byte string without charset
4615 translation. Translation happens only if iconv() is available; HAVE_ICONV
4616 indicates this at build time. HEADERS_CHARSET gives the charset to
4617 translate to; headers_charset can change it in the configuration, and
4618 "headers charset" can change it in an individual filter file.
4620 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4621 that creates RFC 2047 encoded "words" labels them as that charset instead
4622 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4623 expansion operator; (ii) when Exim creates a From: line for a local
4624 message; (iii) when a header line is rewritten to include a "phrase" part.
4626 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4627 buggy, causing it to skip the first lines of messages whose message ID
4628 ended in 'D'. This would not have bitten before Exim release 4.14, because
4629 message IDs were unlikely to end in 'D' before then. The effect was to have
4630 incorrect size information for certain domains.
4632 45. #include "config.h" was missing at the start of the crypt16.c module. This
4633 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4636 46. If there was a timeout during a "random" callout check, Exim treated it as
4637 a failure of the random address, and carried on sending RSET and the real
4638 address. If the delay was just some slowness somewhere, the response to the
4639 original RCPT would be taken as a response to RSET and so on, causing
4640 mayhem of various kinds.
4642 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4643 when I implemented it. It didn't allow for the fact that some option values
4644 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4645 the right test for positive values.
4647 48. Domain names in DNS records are case-independent. Exim always looks them up
4648 in lower case. Some resolvers return domain names in exactly the case they
4649 appear in the zone file, that is, they may contain uppercase letters. Not
4650 all resolvers do this - some return always lower case. Exim was treating a
4651 change of case by a resolver as a change of domain, similar to a widening
4652 of a domain abbreviation. This triggered its re-routing code and so it was
4653 trying to route what was effectively the same domain again. This normally
4654 caused routing to fail (because the router wouldn't handle the domain
4655 twice). Now Exim checks for this case specially, and just changes the
4656 casing of the domain that it ultimately uses when it transmits the message
4659 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4662 50. If a filter generated a file delivery with a non-absolute name (possible if
4663 no home directory exists for the router), the forbid_file option was not
4666 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4667 the existing equality matching.
4669 52. Exim was using ints instead of ino_t variables in some places where it was
4670 dealing with inode numbers.
4672 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4673 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4674 called TMPDIR, and if it finds it is different, it changes its value.
4676 54. The smtp_printf() function is now made available to local_scan() so
4677 additional output lines can be written before returning. There is also an
4678 smtp_fflush() function to enable the detection of a dropped connection.
4679 The variables smtp_input and smtp_batched_input are exported to
4682 55. Changed the default runtime configuration: the message "Unknown user"
4683 has been removed from the ACL, and instead placed on the localuser router,
4684 using the cannot_route_message feature. This means that any verification
4685 failures that generate their own messages won't get overridden. Similarly,
4686 the "Unrouteable address" message that was in the ACL for unverifiable
4687 relay addresses has also been removed.
4689 56. Added hosts_avoid_esmtp to the smtp transport.
4691 57. The exicyclog script was not checking for the esoteric option
4692 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4693 will work only if exicyclog is run under the appropriate euid.
4695 58. Following a discussion on the list, the rules by which Exim recognises line
4696 endings on incoming messages have been changed. The -dropcr and drop_cr
4697 options are now no-ops, retained only for backwards compatibility. The
4698 following line terminators are recognized: LF CRLF CR. However, special
4699 processing applies to CR:
4701 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4702 nor a local message in the state where . is a terminator.
4704 (ii) If a bare CR is encountered in a header line, an extra space is added
4705 after the line terminator so as not to end the header. The reasoning
4706 behind this is that bare CRs in header lines are most likely either
4707 to be mistakes, or people trying to play silly games.
4709 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4710 was being incorrectly given as 18 bytes larger than it should have been.
4711 This is a VOB (very old bug).
4713 60. This may never have affected anything current, but just in case it has:
4714 When the local host is found other than at the start of a list of hosts,
4715 the local host, those with the same MX, and any that follow, are discarded.
4716 When the list in question was part of a longer list of hosts, the following
4717 hosts (not currently being processed) were also being discarded. This no
4718 longer happens. I'm not sure if this situation could ever has previously
4721 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4724 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4725 header. This is something that is recommended in a new Internet Draft, and
4726 is something that is documented as being done by Sendmail. There are two
4727 possible values. For messages generated by the autoreply transport, Exim
4730 Auto-Submitted: auto-replied
4732 whereas for all other generated messages (e.g. bounces) it adds
4734 Auto-Submitted: auto-generated
4736 63. The "personal" condition in filters now includes a test for the
4737 Auto-Submitted: header. If it contains the string "auto-" the message it
4738 not considered personal.
4740 64. Added rcpt_include_affixes as a generic transport option.
4742 65. Added queue_only_override (default true).
4744 66. Added the syslog_duplication option.
4746 67. If what should have been the first header line of a message consisted of
4747 a space followed by a colon, Exim was mis-interpreting it as a header line.
4748 It isn't of course - it is syntactically invalid and should therefore be
4749 treated as the start of the message body. The misbehaviour could have
4750 caused a number of strange effects, including loss of data in subsequent
4751 header lines, and spool format errors.
4753 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4754 client host had authenticated. This control can now be exercised by an ACL
4755 for more flexibility.
4757 69. By default, callouts do not happen when testing with -bh. There is now a
4758 variant, -bhc, which does actually run the callout code, including
4759 consulting and updating the callout cache.
4761 70. Added support for saslauthd authentication, courtesy of Alexander
4764 71. If statvfs() failed on the spool or log directories while checking their
4765 size for availability, Exim confusingly gave the error "space shortage".
4766 Furthermore, in debugging mode it crashed with a floating point exception.
4767 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4768 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4769 serious problem, Exim now writes to the main and panic logs when this
4770 happens, with details of the failure. It then refuses to accept the
4771 incoming message, giving the message "spool directory problem" or "log
4772 directory problem" with a 421 code for SMTP messages.
4774 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4775 0, 1, and 2 exist, because some OS complain for execs without them (see
4776 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4777 descriptors. However, the code omitted to check that the open succeeded,
4778 causing mysterious errors if for some reason the permissions on /dev/null
4779 got screwed. Now Exim writes a message to the main and panic logs, and
4780 bombs out if it can't open /dev/null.
4782 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4783 interact so that it is all more flexible. It is supposed to remain
4784 backwards compatible. Also added extra_local_interfaces.
4786 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4787 to bomb out with an assertion failure - to the client this appears as a
4788 connection drop. This problem occurs in the part of the code that was taken
4789 from the Samba project. Fortunately, the assertion is in a very simple
4790 function, so I have fixed this by reproducing the function inline in the
4791 one place where it is called, and arranging for authentication to fail
4792 instead of killing the process with assert().
4794 75. The SPA client code was not working when the server requested OEM rather
4795 than Unicode encoding.
4797 76. Added code to make require_files with a specific uid setting more usable in
4798 the case where statting the file as root fails - usually a non-root-mounted
4799 NFS file system. When this happens and the failure is EACCES, Exim now
4800 forks a subprocess and does the per-uid checking as the relevant uid.
4802 77. Added process_log_path.
4804 78. If log_file_path was not explicitly set, a setting of check_log_space or
4805 check_log_inodes was ignored.
4807 79. If a space check for the spool or log partitions fails, the incident is now
4808 logged. Of course, in the latter case the data may get lost...
4810 80. Added the %p formatting code to string_format() so that it can be used to
4811 print addresses in debug_print(). Adjusted all the address printing in the
4812 debugging in store.c to use %p rather than %d.
4814 81. There was a concern that a line of code in smtp_in.c could overflow a
4815 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4816 initially expressed, the concern was not well-founded, because trailing
4817 spaces are removed early. However, if the trailing spaces were followed by
4818 a NULL, they did not get removed, so the overflow was possible. Two fixes
4821 (a) I re-wrote the offending code in a cleaner fashion.
4822 (b) If an incoming SMTP command contains a NULL character, it is rejected
4825 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4826 runs initgroups(), so that if the Exim user is in any additional groups,
4827 they will be used during message reception.
4833 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob