1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.519 2007/06/26 11:16:54 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.
81 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
82 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
85 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
88 PH/02 In an ACL statement such as
90 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
92 if a client was not listed at all, or was listed with a value other than
93 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
94 the condition was not true (as it should be), so access was not denied.
95 The bug was that the ! inversion was incorrectly passed on to the second
96 item. This has been fixed.
98 PH/03 Added additional dnslists conditions == and =& which are different from
99 = and & when the dns lookup returns more than one IP address.
101 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
102 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
104 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
105 FSYNC, which compiles an option called disable_fsync that allows for
106 bypassing fsync(). The documentation is heavily laced with warnings.
108 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
110 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
111 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
112 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
113 including adding "make clean"; (3) Added -fPIC when compiling the test
114 dynamically loaded module, to get rid of a warning.
116 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
117 message fails, move_frozen_messages = true and ignore_bounce_errors_after
118 = 0s. The bug is otherwise harmless.
120 PH/07 There was a bug in the dovecot authenticator such that the value of
121 $auth1 could be overwritten, and so not correctly preserved, after a
122 successful authentication. This usually meant that the value preserved by
123 the server_setid option was incorrect.
125 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
127 PH/09 Installed PCRE release 7.0.
129 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
130 run for batched SMTP input. It is now run at the start of every message
131 in the batch. While fixing this I discovered that the process information
132 (output by running exiwhat) was not always getting set for -bs and -bS
133 input. This is fixed, and it now also says "batched" for BSMTP.
135 PH/11 Added control=no_pipelining.
137 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
138 patch, slightly modified), and move the expansion of helo_data till after
139 the connection is made in the smtp transport (so it can use these
142 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
144 PH/14 Added log_selector = +pid.
146 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
148 PH/16 Add ${if forany and ${if forall.
150 PH/17 Added dsn_from option to vary the From: line in DSNs.
152 PH/18 Flush SMTP output before performing a callout, unless control =
153 no_callout_flush is set.
155 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
156 was true (the default) a successful delivery failed to delete the retry
157 item, thus causing premature timeout of the address. The bug is now
160 PH/20 Added hosts_avoid_pipelining to the smtp transport.
162 PH/21 Long custom messages for fakedefer and fakereject are now split up
163 into multiline reponses in the same way that messages for "deny" and
164 other ACL rejections are.
166 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
167 with slight modification.
169 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
170 draft, changing the syntax and factoring some duplicate code".
172 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
173 for deliveries of the second and subsequent messages over the same SMTP
176 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
177 ${reduce, with only minor "tidies".
179 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
181 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
182 expansion side effects.
184 PH/27 When a message times out after an over-quota error from an Exim-imposed
185 quota, the bounce message says "mailbox is full". This message was not
186 being given when it was a system quota that was exceeded. It now should
189 MH/03 Made $recipients available in local_scan(). local_scan() already has
190 better access to the recipient list through recipients_list[], but
191 $recipients can be useful in postmaster-provided expansion strings.
193 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
194 in the case of a MAIL command with additional options following the
195 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
196 were accidentally chopped off.
198 PH/29 SMTP synchronization checks are implemented when a command is read -
199 there is a check that no more input is waiting when there shouldn't be
200 any. However, for some commands, a delay in an ACL can mean that it is
201 some time before the response is written. In this time, more input might
202 arrive, invalidly. So now there are extra checks after an ACL has run for
203 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
204 pipelining has not been advertised.
206 PH/30 MH's patch to allow iscntrl() characters to be list separators.
208 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
209 returned in the SMTP response when smtp_return_error_details was false.
212 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
213 instead of the C I/O that was originally supplied, because problems were
216 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
217 Exim which did not show up earlier: it was assuming that a call to
218 SSL_CTX_set_info_callback() might give an error value. In fact, there is
219 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
220 was a macro that became an assignment, so it seemed to work. This has
221 changed to a proper function call with a void return, hence the compile
222 error. Exim's code has been fixed.
224 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
227 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
229 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
231 PH/37 If a message is not accepted after it has had an id assigned (e.g.
232 because it turns out to be too big or there is a timeout) there is no
233 "Completed" line in the log. When some messages of this type were
234 selected by exigrep, they were listed as "not completed". Others were
235 picked up by some special patterns. I have improved the selection
236 criteria to be more general.
238 PH/38 The host_find_failed option in the manualroute router can now be set
239 to "ignore", to completely ignore a host whose IP address cannot be
240 found. If all hosts are ignored, the behaviour is controlled by the new
241 host_all_ignored option.
243 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
244 homing or because of multiple MX records with /mx) generated more than
245 one IP address, and the following item turned out to be the local host,
246 all the secondary addresses of the first item were incorrectly removed
247 from the list, along with the local host and any following hosts (which
248 is what is supposed to happen).
250 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
251 whoever called Exim into the -H file. In the case of the daemon it was
252 behaving confusingly. When first started, it used values for whoever
253 started the daemon, but after a SIGHUP it used the Exim user (because it
254 calls itself on a restart). I have changed the code so that it now always
257 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
258 message are rejected with the same error (e.g. no authentication or bad
259 sender address), and a DATA command is nevertheless sent (as can happen
260 with PIPELINING or a stupid MUA), the error message that was given to the
261 RCPT commands is included in the rejection of the DATA command. This is
262 intended to be helpful for MUAs that show only the final error to their
265 PH/42 Another patch from the Sieve maintainer.
267 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
268 Eximstats - Fixed some broken HTML links and added missing column headers
270 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
271 columns for Rejects, Temp Rejects, Ham, and Spam rows.
273 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
275 PH/43 Yet another patch from the Sieve maintainer.
277 PH/44 I found a way to check for a TCP/IP connection going away before sending
278 the response to the final '.' that terminates a message, but only in the
279 case where the client has not sent further data following the '.'
280 (unfortunately, this is allowed). However, in many cases there won't be
281 any further data because there won't be any more messages to send. A call
282 to select() can be used: if it shows that the input is "ready", there is
283 either input waiting, or the socket has been closed. An attempt to read
284 the next input character can distinguish the two cases. Previously, Exim
285 would have sent an OK response which the client would never have see.
286 This could lead to message repetition. This fix should cure that, at
287 least in a lot of common cases.
289 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
290 advertised in response to EHLO.
296 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
297 fixed by 4.65/MH/01 (is this a record?) are fixed:
299 (i) An empty string was always treated as zero by the numeric comparison
300 operators. This behaviour has been restored.
302 (ii) It is documented that the numeric comparison operators always treat
303 their arguments as decimal numbers. This was broken in that numbers
304 starting with 0 were being interpreted as octal.
306 While fixing these problems I realized that there was another issue that
307 hadn't been noticed. Values of message_size_limit (both the global option
308 and the transport option) were treated as octal if they started with 0.
309 The documentation was vague. These values are now always treated as
310 decimal, and I will make that clear in the documentation.
316 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
317 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
320 MH/01 Don't check that the operands of numeric comparison operators are
321 integers when their expansion is in "skipping" mode (fixes bug
322 introduced by 4.64-PH/07).
324 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
325 child addresses, Exim now panics and dies. Previously, because the count
326 is held in a short int, deliveries were likely to be lost. As such a
327 large number of recipients for a single message is ridiculous
328 (performance will be very, very poor), I have chosen to impose a limit
329 rather than extend the field.
335 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
336 leftover -K file (the existence of which was triggered by #402).
337 While we were at it, introduced process PID as part of the -K
338 filename. This should rule out race conditions when creating
341 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
342 processing considerably. Previous code took too long for large mails,
343 triggering a timeout which in turn triggers #401.
345 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
346 in the DK code in transports.c. sendfile() is not really portable,
347 hence the _LINUX specificness.
349 TF/01 In the add_headers option to the mail command in an Exim filter,
350 there was a bug that Exim would claim a syntax error in any
351 header after the first one which had an odd number of characters
354 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
355 callout verification, Exim cached a "reject" for the entire domain. This
356 is correct for most verifications, but it is not correct for a recipient
357 verification with use_sender or use_postmaster set, because in that case
358 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
359 case of MAIL FROM:<> rejection from other early rejections (e.g.
360 rejection of HELO). When verifying a recipient using a non-null MAIL
361 address, the cache is ignored if it shows MAIL FROM:<> rejection.
362 Whatever the result of the callout, the value of the domain cache is
363 left unchanged (for any other kind of callout, getting as far as trying
364 RCPT means that the domain itself is ok).
366 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
369 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
370 manifest itself as EPIPE rather than ECONNECT. When tidying away a
371 session, the daemon ignores ECONNECT errors and logs others; it now
372 ignores EPIPE as well.
374 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
375 (quoted-printable decoding).
377 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
378 later the small subsequent patch to fix an introduced bug.
380 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
382 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
384 PH/08 An error is now given if message_size_limit is specified negative.
386 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
387 to be given (somewhat) arbitrary names.
389 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
392 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
393 miscellaneous code fixes
395 PH/10 Added the log_reject_target ACL modifier to specify where to log
398 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
399 hostname. This is wrong, because it relates to the incoming message (and
400 probably the interface on which it is arriving) and not to the outgoing
401 callout (which could be using a different interface). This has been
402 changed to use the value of the helo_data option from the smtp transport
403 instead - this is what is used when a message is actually being sent. If
404 there is no remote transport (possible with a router that sets up host
405 addresses), $smtp_active_hostname is used.
407 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
408 tweaks were necessary in order to get it to work (see also 21 below):
409 (a) The code assumed that strncpy() returns a negative number on buffer
410 overflow, which isn't the case. Replaced with Exim's string_format()
412 (b) There were several signed/unsigned issues. I just did the minimum
413 hacking in of casts. There is scope for a larger refactoring.
414 (c) The code used strcasecmp() which is not a standard C function.
415 Replaced with Exim's strcmpic() function.
416 (d) The code set only $1; it now sets $auth1 as well.
417 (e) A simple test gave the error "authentication client didn't specify
418 service in request". It would seem that Dovecot has changed its
419 interface. Fortunately there's a specification; I followed it and
420 changed what the client sends and it appears to be working now.
422 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
425 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
426 address A is aliased to B and C, where B exists and C does not. Without
427 -v the output is "A verified" because verification stops after a
428 successful redirection if more than one address is generated. However,
429 with -v the child addresses are also verified. Exim was outputting "A
430 failed to verify" and then showing the successful verification for C,
431 with its parentage. It now outputs "B failed to verify", showing B's
432 parentage before showing the successful verification of C.
434 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
435 look up a TXT record in a specific list after matching in a combined
438 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
439 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
440 they consult the DNS. I had assumed they would set it the way they
441 wanted; and indeed my experiments on Linux seem to show that in some
442 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
443 To be on the safe side, however, I have now made the interface to
444 host_find_byname() similar to host_find_bydns(), with an argument
445 containing the DNS resolver options. The host_find_byname() function now
446 sets these options at its start, just as host_find_bydns() does. The smtp
447 transport options dns_qualify_single and dns_search_parents are passed to
448 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
449 of host_find_byname() use the default settings of RES_DEFNAMES
450 (qualify_single) but not RES_DNSRCH (search_parents).
452 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
453 spool_read_header() do less string testing, by means of a preliminary
454 switch on the second character of optional "-foo" lines. (This is
455 overdue, caused by the large number of possibilities that now exist.
456 Originally there were few.) While I was there, I also converted the
457 str(n)cmp tests so they don't re-test the leading "-" and the first
458 character, in the hope this might squeeze out yet more improvement.
460 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
461 flag allowing group syntax was set by the header_syntax check but not
462 turned off, possible causing trouble later; (2) The flag was not being
463 set at all for the header_verify test, causing "group"-style headers to
464 be rejected. I have now set it in this case, and also caused header_
465 verify to ignore an empty address taken from a group. While doing this, I
466 came across some other cases where the code for allowing group syntax
467 while scanning a header line wasn't quite right (mostly, not resetting
468 the flag correctly in the right place). These bugs could have caused
469 trouble for malformed header lines. I hope it is now all correct.
471 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
472 with the "reply" argument non-NULL. The code, however (which originally
473 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
474 but it didn't always do it. This confused somebody who was copying the
475 code for some other use. I have removed all the tests.
477 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
478 feature that was used to support insecure browsers during the U.S. crypto
479 embargo. It requires special client support, and Exim is probably the
480 only MTA that supported it -- and would never use it because real RSA is
481 always available. This code has been removed, because it had the bad
482 effect of slowing Exim down by computing (never used) parameters for the
483 RSA_EXPORT functionality.
485 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
486 authenticator to fail if there's a tab character in the incoming data
487 (there should never be unless someone is messing about, as it's supposed
488 to be base64-encoded). Also added, on Timo's advice, the "secured" option
489 if the connection is using TLS or if the remote IP is the same as the
490 local IP, and the "valid-client-cert option" if a client certificate has
493 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
494 authenticators. This can be used for authorization after authentication
495 succeeds. (In the case of plaintext, it servers for both authentication
498 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
499 if any retry times were supplied.
501 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
502 connection, where there is no client IP address to check. In this
503 situation, the verify now always succeeds.
505 PH/25 Applied John Jetmore's -Mset patch.
507 PH/26 Added -bem to be like -Mset, but loading a message from a file.
509 PH/27 In a string expansion for a processed (not raw) header when multiple
510 headers of the same name were present, leading whitespace was being
511 removed from all of them, but trailing whitespace was being removed only
512 from the last one. Now trailing whitespace is removed from each header
513 before concatenation. Completely empty headers in a concatenation (as
516 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
517 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
519 PH/29 [Removed. This was a change that I later backed out, and forgot to
520 correct the ChangeLog entry (that I had efficiently created) before
521 committing the later change.]
523 PH/30 Exim was sometimes attempting to deliver messages that had suffered
524 address errors (4xx response to RCPT) over the same connection as other
525 messages routed to the same hosts. Such deliveries are always "forced",
526 so retry times are not inspected. This resulted in far too many retries
527 for the affected addresses. The effect occurred only when there were more
528 hosts than the hosts_max_try setting in the smtp transport when it had
529 the 4xx errors. Those hosts that it had tried were not added to the list
530 of hosts for which the message was waiting, so if all were tried, there
531 was no problem. Two fixes have been applied:
533 (i) If there are any address or message errors in an SMTP delivery, none
534 of the hosts (tried or untried) are now added to the list of hosts
535 for which the message is waiting, so the message should not be a
536 candidate for sending over the same connection that was used for a
537 successful delivery of some other message. This seems entirely
538 reasonable: after all the message is NOT "waiting for some host".
539 This is so "obvious" that I'm not sure why it wasn't done
540 previously. Hope I haven't missed anything, but it can't do any
541 harm, as the worst effect is to miss an optimization.
543 (ii) If, despite (i), such a delivery is accidentally attempted, the
544 routing retry time is respected, so at least it doesn't keep
545 hammering the server.
547 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
548 in ${readsocket because some servers need this prod.
550 PH/32 Added some extra debug output when updating a wait-xxx database.
552 PH/33 The hint "could be header name not terminated by colon", which has been
553 given for certain expansion errors for a long time, was not being given
554 for the ${if def:h_colon_omitted{... case.
556 PH/34 The spec says: "With one important exception, whenever a domain list is
557 being scanned, $domain contains the subject domain." There was at least
558 one case where this was not true.
560 PH/35 The error "getsockname() failed: connection reset by peer" was being
561 written to the panic log as well as the main log, but it isn't really
562 panic-worthy as it just means the connection died rather early on. I have
563 removed the panic log writing for the ECONNRESET error when getsockname()
566 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
567 runs only) independently of the message's sender address. This meant
568 that, if the 4xx error was in fact related to the sender, a different
569 message to the same recipient with a different sender could confuse
570 things. In particualar, this can happen when sending to a greylisting
571 server, but other circumstances could also provoke similar problems.
572 I have changed the default so that the retry time for these errors is now
573 based a combination of the sender and recipient addresses. This change
574 can be overridden by setting address_retry_include_sender=false in the
577 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
578 remote server are returned as part of bounce messages. This was not
579 happening for LMTP over a pipe (the lmtp transport), but now it is the
580 same for both kinds of LMTP.
582 PH/38 Despite being documented as not happening, Exim was rewriting addresses
583 in header lines that were in fact CNAMEs. This is no longer the case.
585 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
586 and queue runs started by the daemon processed all messages. This has
587 been fixed so that -R and -S can now usefully be given with -q<time>.
589 PH/40 Import PCRE release 6.7 (fixes some bugs).
591 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
593 PH/42 Give an error if -q is specified more than once.
595 PH/43 Renamed the variables $interface_address and $interface_port as
596 $received_ip_address and $received_port, to make it clear that these
597 values apply to message reception, and not to the outgoing interface when
598 a message is delivered. (The old names remain recognized, of course.)
600 PH/44 There was no timeout on the connect() call when using a Unix domain
601 socket in the ${readsocket expansion. There now is.
603 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
604 be meaningful with "accept".
606 SC/01 Eximstats V1.43
607 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
609 SC/02 Eximstats V1.44
610 Use a glob alias rather than an array ref in the generated
611 parser. This improves both readability and performance.
613 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
614 Collect SpamAssassin and rejection statistics.
615 Don't display local sender or destination tables unless
616 there is data to show.
617 Added average volumes into the top table text output.
619 SC/04 Eximstats V1.46
620 Collect data on the number of addresses (recipients)
621 as well as the number of messages.
623 SC/05 Eximstats V1.47
624 Added 'Message too big' to the list of mail rejection
625 reasons (thanks to Marco Gaiarin).
627 SC/06 Eximstats V1.48
628 Mainlog lines which have GMT offsets and are too short to
629 have a flag are now skipped.
631 SC/07 Eximstats V1.49 (Alain Williams)
632 Added the -emptyok flag.
634 SC/08 Eximstats V1.50
635 Fixes for obtaining the IP address from reject messages.
637 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
638 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
639 whitesspace changes from 4.64-PH/27
641 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
644 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
645 are found, allow negative numbers in numeric criteria)
647 JJ/06 exipick.20061117.2, added new $message_body_missing variable
649 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
650 to match changes made in 4.64-PH/43
652 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
654 PH/47 Put in an explicit test for a DNS lookup of an address record where the
655 "domain" is actually an IP address, and force a failure. This locks out
656 those revolvers/nameservers that support "A-for-A" lookups, in
657 contravention of the specifications.
659 PH/48 When a host name was looked up from an IP address, and the subsequent
660 forward lookup of the name timed out, the host name was left in
661 $sender_host_name, contrary to the specification.
663 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
664 restricted to single-key lookups, Exim was not diagnosing an error if
665 * or *@ was used with a query-style lookup.
667 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
669 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
670 long ago, but noone interested enough thought of it. Let's just say that
671 the "1.1" means that there are some new functions that weren't there at
672 some point in the past.
674 PH/51 Error processing for expansion failure of helo_data from an smtp
675 transport during callout processing was broken.
677 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
678 tested/used via the -bh/-bhc/-bs options.
680 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
681 bug, fixed in subsequent PCRE releases).
683 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
684 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
686 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
692 SC/01 Use a glob alias rather than an array ref in eximstats generated
693 parser. This improves both readability and performance.
695 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
696 Don't display local sender or destination tables in eximstats unless
697 there is data to show.
698 Added average volumes into the eximstats top table text output.
700 SC/03 Collect data on the number of addresses (recipients) as well
701 as the number of messages in eximstats.
703 TF/01 Correct an error in the documentation for the redirect router. Exim
704 does (usually) call initgroups() when daemonizing.
706 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
707 with consistent privilege compared to when running as a daemon.
709 TF/03 Note in the spec that $authenticated_id is not set for local
710 submissions from trusted users.
712 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
713 Thanks to Dean Brooks <dean@iglou.com> for the patch.
715 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
716 by adding some example configuration directives to the default
717 configuration file. A little bit of work is required to uncomment the
718 directives and define how usernames and passwords are checked, but
719 there is now a framework to start from.
721 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
722 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
723 without this. I don't know how relevant this is to other LDAP libraries.
725 PH/02 Add the verb name to the "unknown ACL verb" error.
727 PH/03 Magnus Holmgren's patch for filter_prepend_home.
729 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
731 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
732 directory not expanded when it should be if an expanded home directory
733 was set for the address (which is overridden by the transport).
735 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
738 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
739 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
740 because it is too late at that time, and has no effect.
742 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
743 security issue with \' (bugzilla #107). I could not use the
744 PQescapeStringConn() function, because it needs a PGconn value as one of
747 PH/08 When testing addresses using -bt, indicate those final addresses that
748 are duplicates that would not cause an additional delivery. At least one
749 person was confused, thinking that -bt output corresponded to deliveries.
750 (Suppressing duplicates isn't a good idea as you lose the information
751 about possibly different redirections that led to the duplicates.)
753 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
754 systems where poll() doesn't work, in particular OS X.
756 PH/10 Added more information to debugging output for retry time not reached.
758 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
759 operations in malware.c.
761 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
764 PH/13 If write_rejectlog was set false when logging was sent to syslog with
765 syslog_duplication set false, log lines that would normally be written
766 both the the main log and to the reject log were not written to syslog at
769 PH/14 In the default configuration, change the use of "message" in ACL warn
770 statements to "add_header".
772 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
773 not followed by a command (e.g. "seen endif").
775 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
776 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
779 PH/17 Added extra conditions to the default value of delay_warning_condition
783 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
784 { match{$h_precedence:}{(?i)bulk|list|junk} } \
785 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
788 The Auto-Submitted: and various List- headers are standardised, whereas I
789 don't think Precedence: ever was.
791 PH/18 Refactored debugging code in route_finduser() to show more information,
792 in particular, the error code if getpwnam() issues one.
794 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
795 This is apparently needed in addition to the PH/07 change above to avoid
796 any possible encoding problems.
798 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
799 but not after initializing Perl.
801 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
802 output them only if debugging. By default they are written stderr,
803 apparently, which is not desirable.
805 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
808 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
811 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
813 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
814 authenticated or an ident call has been made. Suppress the default
815 values for $authenticated_id and $authenticated_sender (but permit -oMai
816 and -oMas) when testing with -bh.
818 PH/24 Re-jigged the order of the tests in the default configuration so that the
819 tests for valid domains and recipients precede the DNS black list and CSA
820 tests, on the grounds that those ones are more expensive.
822 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
823 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
824 This bug exists in every version of Exim that I still have, right back to
827 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
828 However, an attempt to turn on case-sensitivity in a regex key by
829 including (?-i) didn't work because the subject string was already
830 lowercased, and the effects were non-intuitive. It turns out that a
831 one-line patch can be used to allow (?-i) to work as expected.
837 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
838 other effects) broke the use of negated acl sub-conditions.
840 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
843 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
844 "Deny" causes Exim to reject the incoming connection with a 554 error.
845 Unfortunately, if there is a major crisis, such as a disk failure,
846 tcp-wrappers gives "deny", whereas what one would like would be some
847 kind of temporary error. A kludge has been added to help with this.
848 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
849 554 error is used if errno is still zero or contains ENOENT (which occurs
850 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
853 PH/03 Add -lutil to the default FreeBSD LIBS setting.
855 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
856 errors. Otherwise a message that provokes a temporary error (when other
857 messages do not) can cause a whole host to time out.
859 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
860 addresses were routed directly to files or pipes from a redirect router.
861 File deliveries just didn't batch; pipe deliveries might have suffered
864 PH/06 A failure to get a lock for a hints database would erroneously always say
865 "Failed to get write lock", even when it was really a read lock.
867 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
868 of 0600. This has been changed to use the value of the lockfile_mode
869 option (which defaults to 0600).
871 PH/08 Applied small patch from the Sieve maintainer.
873 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
874 folder from quota calculations, a direct delivery into this folder messed
875 up the contents of the maildirsize file. This was because the regex was
876 used only to exclude .Trash (or whatever) when the size of the mailbox
877 was calculated. There was no check that a delivery was happening into an
878 excluded directory. This bug has been fixed by ignoring all quota
879 processing for deliveries into excluded directories.
881 PH/10 Added the maildirfolder_create_regex option to appendfile.
887 PH/01 The code for finding all the local interface addresses on a FreeBSD
888 system running IPv6 was broken. This may well have applied to all BSD
889 systems, as well as to others that have similar system calls. The broken
890 code found IPv4 interfaces correctly, but gave incorrect values for the
891 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
892 that it would not match correctly against @[] and not recognize the IPv6
895 PH/02 The ipliteral router was not recognizing addresses of the form user@
896 [ipv6:....] because it didn't know about the "ipv6:" prefix.
898 PH/03 Added disable_ipv6.
900 PH/04 Changed $reply_address to use the raw form of the headers instead of the
901 decoded form, because it is most often used to construct To: headers
902 lines in autoreplies, and the decoded form may well be syntactically
903 invalid. However, $reply_address has leading white space removed, and all
904 newlines turned into spaces so that the autoreply transport does not
907 PH/05 If group was specified without a user on a router, and no group or user
908 was specified on a transport, the group from the router was ignored.
910 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
911 for visible compile-time settings that can be used to change these
912 numbers, for those that want even more. Backwards compatibility with old
913 spool files has been maintained. However, going back to a previous Exim
914 release will lost any variables that are in spool files.
916 PH/07 Two small changes when running in the test harness: increase delay when
917 passing a TCP/IP connection to a new process, in case the original
918 process has to generate a bounce, and remove special handling of
919 127.0.0.2 (sic), which is no longer necessary.
921 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
922 be the same on different OS.
924 PH/09 Moved a debug statement in filter processing to avoid a race problem when
927 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
928 whether --show-vars was specified or not
930 JJ/02 exipick: Added support for new ACL variable spool format introduced
933 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
934 syntactically invalid From: or Reply-to: line, and a filter used this to
935 generate an autoreply, and therefore failed to obtain an address for the
936 autoreply, Exim could try to deliver to a non-existent relative file
937 name, causing unrelated and misleading errors. What now happens is that
938 it logs this as a hard delivery error, but does not attempt to create a
941 PH/11 The exinext utility has a -C option for testing purposes, but although
942 the given file was scanned by exinext itself; it wasn't being passed on
943 when Exim was called.
945 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
946 an end-of-file indication when reading a command response.
948 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
949 compiled. In many other places in Exim, IPv6 addresses are always
950 recognized, so I have changed this. It also means that IPv4 domain
951 literals of the form [IPV4:n.n.n.n] are now always recognized.
953 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
954 used if the router is not running as root, for example, when verifying at
955 ACL time, or when using -bh. The debugging output from this situation was
956 non-existent - all you got was a failure to exec. I have made two
959 (a) Failures to set uid/gid, the current directory, or a process leader
960 in a subprocess such as that created by queryprogram now generate
961 suitable debugging ouput when -d is set.
963 (b) The queryprogram router detects when it is not running as root,
964 outputs suitable debugging information if -d is set, and then runs
965 the subprocess without attempting to change uid/gid.
967 PH/15 Minor change to Makefile for building test_host (undocumented testing
970 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
971 additional section of a DNS packet that returns MX or SRV records.
972 Instead, it always explicitly searches for A/AAAA records. This avoids
973 major problems that occur when a DNS server includes only records of one
974 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
975 fixed another bug: if SRV records were looked up and the corresponding
976 address records were *not* found in the additional section, the port
977 values from the SRV records were lost.
979 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
980 using the correct key (the original address) when searching the retry
981 rules in order to find which one to use for generating the retry hint.
983 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
984 adding the default one. Similarly, if it contains a Reply-To: header, the
985 errors_reply_to option, if set, is not used.
987 PH/19 When calculating a retry time, Exim used to measure the "time since
988 failure" by looking at the "first failed" field in the retry record. Now
989 it does not use this if it is later than than the arrival time of the
990 message. Instead it uses the arrival time. This makes for better
991 behaviour in cases where some deliveries succeed, thus re-setting the
992 "first failed" field. An example is a quota failure for a huge message
993 when small messages continue to be delivered. Without this change, the
994 "time since failure" will always be short, possible causing more frequent
995 delivery attempts for the huge message than are intended.
996 [Note: This change was subsequently modified - see PH/04 for 4.62.]
998 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
999 $1, $2, $3) because the numerical variables can be reset during some
1000 expansion items (e.g. "match"), thereby losing the authentication data.
1002 PH/21 Make -bV show the size of off_t variables so that the test suite can
1003 decide whether to run tests for quotas > 2G.
1005 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1006 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1007 filecount value is greater than 2G or if a quota value is greater than 2G
1008 on a system where the size of off_t is not greater than 4, a panic error
1011 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1012 never match. The debug and -bh output now contains an explicit error
1013 message indicating a malformed IPv4 address or mask.
1015 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1016 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1017 PH/23 above applies.
1019 PH/25 Do not write to syslog when running in the test harness. The only
1020 occasion when this arises is a failure to open the main or panic logs
1021 (for which there is an explicit test).
1023 PH/26 Added the /no_tell option to "control=freeze".
1025 PH/27 If a host name lookup failed very early in a connection, for example, if
1026 the IP address matched host_lookup and the reverse lookup yielded a name
1027 that did not have a forward lookup, an error message of the form "no IP
1028 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1029 could be logged. Now it outputs the IP address instead of "NULL".
1031 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1032 allows the sender and the authenticated sender to be set when
1033 submitting a message from within Exim. Since child_open_exim() is
1034 documented for local_scan(), the new function should be too.
1036 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1037 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1038 results in an empty string is now treated as unset.
1040 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1042 PH/31 Added sender_verify_fail logging option.
1044 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1045 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1046 tidied the source and removed it altogether.
1048 PH/33 When a queue run was abandoned because the load average was too high, a
1049 log line was always written; now it is written only if the queue_run log
1050 selector is set. In addition, the log line for abandonment now contains
1051 information about the queue run such as the pid. This is always present
1052 in "start" and "stop" lines but was omitted from the "abandon" line.
1054 PH/34 Omit spaces between a header name and the colon in the error message that
1055 is given when verify = headers_syntax fails (if there are lots of them,
1056 the message gets confusing).
1058 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1059 names, as there are now some PTR records that contain slashes. This check
1060 is only to protect against broken name servers that fall over on strange
1061 characters, so the fact that it applies to all lookups doesn't matter.
1063 PH/36 Now that the new test suite is complete, we can remove some of the
1064 special code in Exim that was needed for the old test suite. For example,
1065 sorting DNS records because real resolvers return them in an arbitrary
1066 order. The new test suite's fake resolver always returns records in the
1069 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1070 bounces) except when queue_only is set, to avoid logging races between
1071 the different processes.
1073 PH/38 Panic-die if .include specifies a non-absolute path.
1075 PH/39 A tweak to the "H" retry rule from its user.
1077 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1078 a label. They prevented compilation on older perls.
1080 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1081 a warning to be raised on newish perls.
1083 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1084 on queue. Changes to match documented behaviour of showing count of
1085 messages matching specified criteria.
1087 PH/40 Changed the default ident timeout from 30s to 5s.
1089 PH/41 Added support for the use of login_cap features, on those BSD systems
1090 that have them, for controlling the resources used by pipe deliveries.
1092 PH/42 The content-scanning code uses fopen() to create files in which to put
1093 message data. Previously it was not paying any attention to the mode of
1094 the files. Exim runs with umask(0) because the rest of the code creates
1095 files with open(), and sets the required mode explicitly. Thus, these
1096 files were ending up world-writeable. This was not a big issue, because,
1097 being within the spool directory, they were not world-accessible. I have
1098 created a function called modefopen, which takes an additional mode
1099 argument. It sets umask(777), creates the file, chmods it to the required
1100 mode, then resets the umask. All the relevant calls to fopen() in the
1101 content scanning code have been changed to use this function.
1103 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1104 to 24 hours. This avoids potential overflow problems when processing G
1105 and H retry rules. I suspect nobody ever tinkers with this value.
1107 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1109 PH/45 When the plaintext authenticator is running as a client, the server's
1110 challenges are checked to ensure they are valid base64 strings. By
1111 default, the authentication attempt is cancelled if an invalid string is
1112 received. Setting client_ignore_invalid_base64 true ignores these errors.
1113 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1114 they are received. Thus, the responses can be made to depend on the
1115 challenges. If an invalid string is ignored, an empty string is placed in
1118 PH/46 Messages that are created by the autoreply transport now contains a
1119 References: header, in accordance with RFCs 2822 and 3834.
1121 PH/47 Added authenticated_sender_force to the smtp transport.
1123 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1125 PH/49 Installed latest patch from the Sieve maintainer.
1127 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1128 was also set, the appendfile transport was unnecessarily scanning a
1129 directory of message files (e.g. for maildir delivery) to find the count
1130 of files (along with the size), even though it did not need this
1131 information. It now does the scan only if it needs to find either the
1132 size of the count of files.
1134 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1136 PH/52 Two bugs concerned with error handling when the smtp transport is
1139 (i) Exim was not creating retry information for temporary errors given
1140 for individual recipients after the DATA command when the smtp transport
1141 was used in LMTP mode. This meant that they could be retried too
1142 frequently, and not timed out correctly.
1144 (ii) Exim was setting the flag that allows error details to be returned
1145 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1146 recipients that were returned after the DATA command.
1148 PH/53 This is related to PH/52, but is more general: for any failing address,
1149 when detailed error information was permitted to be returned to the
1150 sender, but the error was temporary, then after the final timeout, only
1151 "retry timeout exceeded" was returned. Now it returns the full error as
1152 well as "retry timeout exceeded".
1154 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1155 do this, and (what is worse) MTAs that accept it.
1157 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1158 will now be deprecated.
1160 PH/56 New os.c-cygwin from the Cygwin maintainer.
1162 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1163 formats (previously only available in exim formats via -bpr, -bpru,
1164 and -bpra. Now also available in native and exiqgrep formats)
1166 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1167 with very large, slow to parse queues
1169 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1171 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1173 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1174 responses to authentication challenges, though it was showing the
1175 challenges; (ii) I've removed the CR characters from the debug output for
1178 PH/58 Allow for the insertion of a newline as well as a space when a string
1179 is turned into more than one encoded-word during RFC 2047 encoding. The
1180 Sieve code now uses this.
1182 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1183 data_4xx, lost_connection, tls_required.
1185 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1186 message was being sent as an SMTP response.
1188 PH/61 Add -l and -k options to exicyclog.
1190 PH/62 When verifying, if an address was redirected to one new address, so that
1191 verification continued, and the new address failed or deferred after
1192 having set something in $address_data, the value of $address_data was not
1193 passed back to the ACL. This was different to the case when no
1194 redirection occurred. The value is now passed back in both cases.
1196 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1197 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1198 use login_cap.h, so on its own it isn't the distinguishing feature. The
1199 new name refers directly to the setclassresources() function.
1201 PH/65 Added configuration files for NetBSD3.
1203 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1205 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1206 is preferred over IPv4.
1208 PH/68 The bounce_return_message and bounce_return_body options were not being
1209 honoured for bounces generated during the reception of non-SMTP messages.
1210 In particular, this applied to messages rejected by the ACL. This bug has
1211 been fixed. However, if bounce_return_message is true and bounce_return_
1212 body is false, the headers that are returned for a non-SMTP message
1213 include only those that have been read before the error was detected.
1214 (In the case of an ACL rejection, they have all been read.)
1216 PH/69 The HTML version of the specification is now built in a directory called
1217 spec_html instead of spec.html, because the latter looks like a path with
1218 a MIME-type, and this confuses some software.
1220 PH/70 Catch two compiler warnings in sieve.c.
1222 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1223 function verify_get_ident() calls ip_connect() to connect a socket, but
1224 if the "connect()" function timed out, ip_connect() used to close the
1225 socket. However, verify_get_ident() also closes the socket later, and in
1226 between Exim writes to the log, which may get opened at this point. When
1227 the socket was closed in ip_connect(), the log could get the same file
1228 descriptor number as the socket. This naturally causes chaos. The fix is
1229 not to close the socket in ip_connect(); the socket should be closed by
1230 the function that creates it. There was only one place in the code where
1231 this was missing, in the iplookup router, which I don't think anybody now
1232 uses, but I've fixed it anyway.
1234 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1235 well as to direct DNS lookups. Otherwise the handling of names in host
1236 lists is inconsistent and therefore confusing.
1242 PH/01 Two changes to the default runtime configuration:
1244 (1) Move the checks for relay_from_hosts and authenticated clients from
1245 after to before the (commented out) DNS black list checks.
1247 (2) Add control=submission to the relay_from_hosts and authenticated
1248 clients checks, on the grounds that messages accepted by these
1249 statements are most likely to be submissions.
1251 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1253 (1) Generate an error if the third argument for the ${prvs expansion is
1256 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1259 (3) Reset the variables that are obtained from the first argument of
1260 ${prvscheck and used in the second argument before leaving the code,
1261 because their memory is reclaimed, so using them afterwards may do
1264 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1265 one (it's much easier than Tom thought :-).
1267 (5) Because of (4), we can now allow for the use of $prvscheck_result
1268 inside the third argument.
1270 PH/03 For some reason, the default setting of PATH when running a command from
1271 a pipe transport was just "/usr/bin". I have changed it to
1274 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1275 anything to be listed in the output from -bV.
1277 PH/05 When a filter generated an autoreply, the entire To: header line was
1278 quoted in the delivery log line, like this:
1280 => >A.N.Other <ano@some.domain> <original@ddress> ...
1282 This has been changed so that it extracts the operative address. There
1283 may be more than one such address. If so, they are comma-separated, like
1286 => >ano@some.domain,ona@other.domain <original@ddress> ...
1288 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1289 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1290 not being looked up in the rDNS to get a host name, Exim was showing the
1291 IP address twice in Received: lines, even though the IP addresses were
1292 identical. For example:
1294 Received: from [1.2.3.4] (helo=[1.2.3.4])
1296 However, if the real host name was known, it was omitting the HELO data
1297 if it matched the actual IP address. This has been tidied up so that it
1298 doesn't show the same IP address twice.
1300 PH/07 When both +timestamp and +memory debugging was on, the value given by
1301 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1302 called by the debug printing, thereby overwriting the timestamp buffer.
1303 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1305 PH/08 When the original message was included in an autoreply transport, it
1306 always said "this is a copy of the message, including all the headers",
1307 even if body_only or headers_only was set. It now gives an appropriate
1310 PH/09 Applied a patch from the Sieve maintainer which:
1312 o fixes some comments
1313 o adds the (disabled) notify extension core
1314 o adds some debug output for the result of if/elsif tests
1315 o points to the current vacation draft in the documentation
1316 and documents the missing references header update
1320 o fixes a bug in processing the envelope test (when testing
1321 multiple envelope elements, the last element determinted the
1324 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1325 Electronic Mail") by including:
1327 Auto-submitted: auto-generated
1329 in the messages that it generates (bounce messages and others, such as
1330 warnings). In the case of bounce messages for non-SMTP mesages, there was
1331 also a typo: it was using "Auto_submitted" (underscore instead of
1332 hyphen). Since every message generated by Exim is necessarily in response
1333 to another message, thes have all been changed to:
1335 Auto-Submitted: auto-replied
1337 in accordance with these statements in the RFC:
1339 The auto-replied keyword:
1341 - SHOULD be used on messages sent in direct response to another
1342 message by an automatic process,
1344 - MUST NOT be used on manually-generated messages,
1346 - MAY be used on Delivery Status Notifications (DSNs) and Message
1347 Disposition Notifications (MDNs),
1349 - MUST NOT be used on messages generated by automatic or periodic
1350 processes, except for messages which are automatic responses to
1353 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1354 to the default Received: header definition.
1356 PH/12 Added log selector acl_warn_skipped (default on).
1358 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1359 variables because (a) they are in the wrong storage pool and (b) even if
1360 they were copied, it wouldn't work properly because of the caching.
1362 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1363 checking when decoding. Apparently there are clients that generate
1364 overlong encoded strings. Why am I not surprised?
1366 PH/15 If the first argument of "${if match_address" was not empty, but did not
1367 contain an "@" character, Exim crashed. Now it writes a panic log message
1368 and treats the condition as false.
1370 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1372 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1373 extension "envelope-auth". The code is finished and in agreement with
1374 other implementations, but there is no documentation so far and in fact,
1375 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1376 not changing the active code.
1378 Print executed "if" and "elsif" statements when debugging is used. This
1379 helps a great deal to understand what a filter does.
1381 Document more things not specified clearly in RFC3028. I had all this
1382 sorted out, when out of a sudden new issues came to my mind. Oops."
1384 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1387 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1388 canonical form (as documented). However, after a host name lookup from
1389 the IP address, check_host() was doing a simple string comparison with
1390 addresses acquired from the DNS when checking that the found name did
1391 have the original IP as one of its addresses. Since any found IPv6
1392 addresses are likely to be in abbreviated form, the comparison could
1393 fail. Luckily, there already exists a function for doing the comparison
1394 by converting both addresses to binary, so now that is used instead of
1395 the text comparison.
1397 PH/20 There was another similar case to PH/19, when a complete host name was
1398 given in a host list; looking up its IP address could give an abbreviated
1399 form, whereas the current host's name might or might not be abbreviated.
1400 The same fix has been applied.
1406 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1407 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1410 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1411 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1413 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1415 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1416 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1417 Also, for "Auto-Submitted", treat anything other than "no" as
1418 non-personal, in accordance with RFC 3834. (Previously it treated
1419 anything starting "auto-" as non-personal.)
1421 TF/01 The control=submission/name=... option had a problem with syntax
1422 errors if the name included a slash character. The /name= option
1423 now slurps the rest of the string, so it can include any characters
1424 but it must come last in the list of options (after /sender_retain
1427 PH/05 Some modifications to the interface to the fake nameserver for the new
1435 TK/01 Added the "success_on_redirect" address verification option. See
1436 NewStuff for rationale and an example.
1438 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1440 PH/02 Patch to exigrep to allow it to work on syslog lines.
1442 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1443 fread() to skip over the body file's header line, because in Cygwin the
1444 header line is locked and is inaccessible.
1446 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1447 co-exist for some time) to make it clear that it is the Exim ID that is
1448 referenced, not the Message-ID: header line.
1450 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1451 string_format() function, because snprintf() does not exist on all
1454 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1455 ${stat: expansion item.
1457 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1458 protocol synchronization error", to keep the pedants happy.
1460 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1461 well as for IRIX systems, when gcc is being used. See the host.c source
1464 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1466 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1469 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1470 transport and to the smtp transport in LMTP mode.
1472 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1474 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1475 run a filter in a subprocess. This could lead to confusion in subsequent
1476 lookups in the parent process. There should also be a search_tidyup() at
1477 the end of the subprocess.
1479 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1480 only if the host matched helo_try_verify_hosts, which caused the
1481 verification to occur when the EHLO/HELO command was issued. The ACL just
1482 tested the remembered result. Now, if a previous verification attempt has
1483 not happened, "verify = helo" does it there and then.
1485 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1487 TK/03 Fix log output including CR from clamd.
1489 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1490 exist provoked a memory error which could cause a segfault.
1492 PH/15 Installed PCRE 6.2
1494 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1496 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1497 of the problem. Specifically, suggested +O2 rather than +O1 for the
1500 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1502 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1503 record, the port was not correctly logged when the outgoing_port log
1504 selector was set (it logged the transort's default port).
1506 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1507 fallback_hosts, and "hosts" in the smtp transport.
1509 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1510 host errors such as "Connection refused".
1512 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1513 authentication with radiusclient 0.4.9:
1515 - Error returned from rc_read_config was caught wrongly
1516 - Username/password not passed on to radius server due to wrong length.
1518 The presumption is that some radiusclient API changes for 4.51/PH/17
1519 were not taken care of correctly. The code is still untested by me (my
1520 Linux distribution still has 0.3.2 of radiusclient), but it was
1521 contributed by a Radius user.
1523 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1524 expanding the "port" option of the smtp transport.
1526 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1527 while reading a MIME header. Thanks to Tom Hughes for a patch.
1529 PH/24 Include config.h inside local_scan.h so that configuration settings are
1532 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1533 that in an ACL for RCPT (for example), you can examine exactly what was
1536 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1537 commands, but it was not correctly comparing the address with the actual
1538 client host address. Thus, it would show the EHLO address in Received:
1539 header lines when this was not necessary.
1541 PH/27 Added the % operator to ${eval:}.
1543 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1544 it should be ignoring failures (because with -C, for example, it has lost
1545 privilege). It wasn't ignoring creation failures other than "already
1548 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1551 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1552 because an input line was too long, either on its own, or by virtue of
1553 too many continuations, the temporary file was not being removed, and the
1554 return code was incorrect.
1556 PH/31 Missing "BOOL" in function definition in filtertest.c.
1558 PH/32 Applied Sieve patches from the maintainer.
1560 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1562 PH/33 Added "verify = not_blind".
1564 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1565 Local/Makefile (with some defaults set). These are used in built scripts
1566 such as exicyclog, but they have never been used in the exim_install
1567 script (though there are many overriding facilities there). I have
1568 arranged that the exim_install script now takes note of these two
1571 PH/35 Installed configuration files for Dragonfly.
1573 PH/36 When a locally submitted message by a trusted user did not contain a
1574 From: header, and the sender address was obtained from -f or from an SMTP
1575 MAIL command, and the trusted user did not use -F to supply a sender
1576 name, $originator_name was incorrectly used when constructing a From:
1577 header. Furthermore, $originator_name was used for submission mode
1578 messages from external hosts without From: headers in a similar way,
1579 which is clearly wrong.
1581 PH/37 Added control=suppress_local_fixups.
1583 PH/38 When log_selector = +received_sender was set, and the addition of the
1584 sender made the log line's construction buffer exactly full, or one byte
1585 less than full, an overflow happened when the terminating "\n" was
1588 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1589 when the result of a list match is failure because a DNS lookup failed.
1591 PH/40 RM_COMMAND is now used in the building process.
1593 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1594 the "build-* directories that it finds.
1596 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1597 address in a domain literal was a prefix of an interface address.
1599 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1600 when verifying a sender address, unless rewrite_headers is false.
1602 PH/44 Wrote a long comment about why errors_to addresses are verified as
1603 recipients, not senders.
1605 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1606 the ratelimit ACL was added.
1608 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1610 PH/46 Added extra information about PostgreSQL errors to the error string.
1612 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1613 avoiding the need to install special zones in a real server. This is
1614 backwards compatible; if it can't find the fake resolver, it drops back.
1615 Thus, both old and new test suites can be run.
1617 TF/02 Added util/ratelimit.pl
1619 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1620 clock is set back in time.
1622 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1623 Candler <B.Candler@pobox.com>.
1625 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1626 OK for addresses that are the result of redirections.
1628 PH/48 A number of further additions for the benefit of the new test suite,
1629 including a fake gethostbyname() that interfaces to the fake DNS resolver
1632 TF/06 The fix for widen_domains has also been applied to qualify_single and
1633 search_parents which are the other dnslookup options that can cause
1636 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1639 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1641 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1642 service" if it encountered an MX record with an empty target hostname.
1643 The message is now "an MX or SRV record indicated no SMTP service".
1645 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1646 if the DNS of the sending site is misconfigured. This is quite a
1647 common situation. This change restores the behaviour of treating a
1648 helo verification defer as a failure.
1650 PH/51 If self=fail was set on a router, the bounce message did not include the
1651 actual error message.
1657 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1659 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1660 command itself ended in error, the underlying message about the transport
1661 filter timeout was being overwritten with the pipe command error. Now the
1662 underlying error message should be appended to the second error message.
1664 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1666 PH/02 Reduce the amount of output that "make" produces by default. Full output
1667 can still be requested.
1669 PH/03 The warning log line about a condition test deferring for a "warn" verb
1670 was being output only once per connection, rather than after each
1671 occurrence (because it was using the same function as for successful
1672 "warn" verbs). This seems wrong, so I have changed it.
1674 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1675 it should not have, which might have caused a crash in the right
1676 circumstances, but probably never did.
1678 PH/04 Installed a modified version of Tony Finch's patch to make submission
1679 mode fix the return path as well as the Sender: header line, and to
1680 add a /name= option so that you can make the user's friendly name appear
1683 TF/03 Added the control = fakedefer ACL modifier.
1685 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1686 Mark Lowes for thorough testing.
1688 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1690 TK/03 Merged latest SRS patch from Miles Wilton.
1692 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1693 with the definition in sysexits.h (which is #included earlier).
1694 Fortunately, Exim does not actually use EX_OK. The code used to try to
1695 preserve the sysexits.h value, by assumimg that macro definitions were
1696 scanned for macro replacements. I have been disabused of this notion,
1697 so now the code just undefines EX_OK before #including unistd.h.
1699 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1700 in the smtp transport. When a block could not be written in a single
1701 write() function, the timeout was being re-applied to each part-write.
1702 This seems wrong - if the receiver was accepting one byte at a time it
1703 would take for ever. The timeout is now adjusted when this happens. It
1704 doesn't have to be particularly precise.
1706 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1707 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1709 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1710 without a domain if the check to <postmaster@domain> fails.
1712 SC/01 Eximstats: added -xls and the ability to specify output files
1713 (patch written by Frank Heydlauf).
1715 SC/02 Eximstats: use FileHandles for outputing results.
1717 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1719 SC/04 Eximstats: fixed display of large numbers with -nvr option
1721 SC/05 Eximstats: fixed merging of reports with empty tables.
1723 SC/06 Eximstats: added the -include_original_destination flag
1725 SC/07 Eximstats: removed tabs and trailing whitespace.
1727 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1729 TK/06 MBOX spool code: Add real "From " MBOX separator line
1730 so the .eml file is really in mbox format (even though
1731 most programs do not really care). Patch from Alex Miller.
1733 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1734 The latter is generated from $received_to and is only set if the
1735 message has one envelope recipient. SA can use these headers,
1736 obviously out-of-the-box. Patch from Alex Miller.
1738 PH/08 The ${def test on a variable was returning false if the variable's
1739 value was "0", contrary to what the specification has always said!
1740 The result should be true unless the variable is empty.
1742 PH/09 The syntax error of a character other than { following "${if
1743 def:variable_name" (after optional whitespace) was not being diagnosed.
1744 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1745 accidental colon was present, for example, could give incorrect results.
1747 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1748 result is used (not including appendfile, where other changes are about
1751 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1752 This involved changing a lot of size variables from int to off_t. It
1753 should work with maildirs and everything.
1755 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1756 spamd dying while we are connected to it.
1758 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1761 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1762 function for BDB 4.3.
1764 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1766 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1767 the sources for PCRE have been re-arranged and more files are now
1770 PH/15 The code I had for printing potentially long long variables in PH/11
1771 above was not the best (it lost precision). The length of off_t variables
1772 is now inspected at build time, and an appropriate printing format (%ld
1773 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1774 to be "long long int" or "long int". This is needed for the internal
1775 formatting function string_vformat().
1777 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1778 the configuration file to be ":syslog", then the script "guesses" where
1779 the logs files are, rather than using the compiled in default. In our
1780 case the guess is not the same as the compiled default, so the script
1781 suddenly stopped working when I started to use syslog. The patch checks
1782 to see if log_file_path is "". If so, it attempts to read it from exim
1783 with no configuration file to get the compiled in version, before it
1784 falls back to the previous guessing code."
1786 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1787 implementing BATV in an Exim configuration. See NewStuff for the gory
1790 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1791 Makefile that are specific to HP-UX.
1793 PH/18 If the "use_postmaster" option was set for a recipient callout together
1794 with the "random" option, the postmaster address was used as the MAIL
1795 FROM address for the random test, but not for the subsequent recipient
1796 test. It is now used for both.
1798 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1799 patch removes a few documentation additions to RFC 3028, because the
1800 latest draft now contains them. It adds the new en;ascii-case comparator
1801 and a new error check for 8bit text in MIME parts. Comparator and
1802 require names are now matched exactly. I enabled the subaddress
1803 extension, but it is not well tested yet (read: it works for me)."
1805 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1806 rework some of the code of TK/09 above to avoid the hardwired use of
1807 "%lld" and "long long". Replaced the call to snprintf() with a call to
1810 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1811 records point to non-existent hosts", "retry timeout exceeded", and
1812 "retry time not reached for any host after a long failure period".
1814 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1815 experimental DomainKeys support:
1817 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1818 (2) On an error such as an illegally used "control", the wrong name for
1819 the control was given.
1821 These problems did NOT occur unless DomainKeys support was compiled.
1823 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1825 PH/24 Added ${if match_ip condition.
1827 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1828 databases so that it will be absolutely obvious if a crash occurs in the
1829 DB library. This is a regular occurrence (often caused by mis-matched
1832 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1833 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1834 picked up on a user's system that detects such things. There doesn't seem
1835 to be a gcc warning option for this - only an attribute that has to be
1836 put on the function's prototype. It seems that in Fedora Core 4 they have
1837 set this on a number of new functions. No doubt there will be more in due
1840 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1841 specify a transport. However, if an address that was verified by such a
1842 router was the subject of a callout, Exim crashed because it tried to
1843 read the rcpt_include_affixes from the non-existent transport. Now it
1844 just assumes that the setting of that option is false. This bug was
1845 introduced by 4.51/PH/31.
1847 PH/28 Changed -d+all to exclude +memory, because that information is very
1848 rarely of interest, but it makes the output a lot bigger. People tend to
1849 do -d+all out of habit.
1851 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1852 code in os-type was giving problems when libc.so lives in lib64, like on
1855 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1856 aren't the modern standard, and it seems that some systems' include files
1857 don't always have them. Exim was already checking for some of the newer
1858 ones like T_AAAA, and defining it itself. I've added checks for all the
1859 record types that Exim uses.
1861 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1862 not automatically generating a new one, as it is supposed to. This
1863 prevented TLS from working. If the file did exist, but contained invalid
1864 data, a new version was generated, as expected. It was only the case of a
1865 non-existent file that was broken.
1867 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1868 with a change in libdomainkeys > 0.64.
1870 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1871 from DNS. If the selector record carries the flag, it now has
1872 precedence over the domain-wide flag.
1874 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1876 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1877 the use of a port name that isn't defined in /etc/services) occurred, the
1878 message was deferred as in a normal delivery, and thus remained on the
1879 spool, instead of being failed because of the mua_wrapper setting. This
1880 is now fixed, and I tidied up some of the mua_wrapper messages at the
1883 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1884 the messages in a hash of arrays rather than using individual hashes.
1885 This is a bit cleaner and results in dramatic memory savings, albeit
1886 at a slight CPU cost.
1888 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1889 as requested by Marc Sherman.
1891 SC/10 Eximstats: added histograms for user specified patterns as requested
1894 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1896 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1897 fopen() in the content-scanning modules that did not already have it.
1903 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1904 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1906 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1908 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1910 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1911 to test Sieve filters that use "vacation".
1913 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1914 that changes the way the GnuTLS parameters are stored in the cache file.
1915 The new format can be generated externally. For backward compatibility,
1916 if the data in the cache doesn't make sense, Exim assumes it has read an
1917 old-format file, and it generates new data and writes a new file. This
1918 means that you can't go back to an older release without removing the
1921 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1922 work if there are any delivery delays because "one_time" forces the
1923 parent to be marked "delivered", so its unseen clone is never tried
1924 again. For this reason, Exim now forbids the simultaneous setting of
1927 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1928 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1929 entry if you want to know the details. The fix, however, neglected to
1930 consider the case when local delivery batching is involved. The test for
1931 "previously delivered" was not happening when checking to see if an
1932 address could be batched with a previous (undelivered) one; under
1933 certain circumstances this could lead to multiple deliveries to the same
1936 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1937 in its include files, and this causes problems building Exim.
1939 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1940 header_syntax) but Exim was just ignoring anything given after a slash.
1941 In particular, this caused confusion with an attempt to use "verify =
1942 reverse_host_lookup/defer_ok". An error is now given when options are
1943 supplied for verify items that do not have them. (Maybe reverse_host_
1944 lookup should have a defer_ok option, but that's a different point.)
1946 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1947 defined by RFC 821) to 2048, because there were problems with some AUTH
1948 commands, and RFC 1869 says the size should be increased for extended
1949 SMTP commands that take arguments.
1951 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1954 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1955 "unknown" error; now it says that the functionality isn't in the binary.
1957 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1958 an address' error message when a string expansion fails (syntax or
1959 whatever). Otherwise the password may appear in the log. Following change
1960 PH/42 below, there is no longer a chance of it appearing in a bounce
1963 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1965 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1966 of its addresses was ever tried. (Bugzilla bug #2.)
1968 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1969 the result incorrectly in the debug output. (It correctly added a newline
1970 to what was transported.)
1972 TF/01 Added $received_time.
1974 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1975 commented out examples of how to interface to a virus scanner and to
1976 SpamAssassin. Also added commented examples of av_scanner and
1977 spamd_address settings.
1979 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1980 and controls are allowed in which ACLs. There were a couple of minor
1981 errors. Some of the entries in the conditions table (which is a table of
1982 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1983 negation of where the condition IS allowed.
1985 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1987 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1988 header file does not have a version number, so I've had to invent a new
1989 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1990 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1991 radiusclient), but it was contributed by a Radius user.
1993 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1994 files or directories, for OpenSSL.
1996 PH/19 When an Exim process that is running as root has to create an Exim log
1997 file, it does so in a subprocess that runs as exim:exim so as to get the
1998 ownership right at creation (otherwise, other Exim processes might see
1999 the file with the wrong ownership). There was no test for failure of this
2000 fork() call, which would lead to the process getting stuck as it waited
2001 for a non-existent subprocess. Forks do occasionally fail when resources
2002 run out. I reviewed all the other calls to fork(); they all seem to check
2005 PH/20 When checking for unexpected SMTP input at connect time (before writing
2006 the banner), Exim was not dealing correctly with a non-positive return
2007 from the read() function. If the client had disconnected by this time,
2008 the result was a log entry for a synchronization error with an empty
2009 string after "input=" when read() returned zero. If read() returned -1
2010 (an event I could not check), uninitialized data bytes were printed.
2011 There were reports of junk text (parts of files, etc) appearing after
2014 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2016 PH/22 Added support for macro redefinition, and (re)definition in between
2017 driver and ACL definitions.
2019 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2020 forgetting to use the resulting value; it was using the unexpanded value.
2022 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2023 hadn't been configured. The fix is from Juergen Kreileder, who
2024 understands it better than I do:
2026 "Here's what I see happening with three configured cyrus_sasl
2027 authenticators configured (plain, login, cram-md5):
2029 On startup auth_cyrus_sasl_init() gets called for each of these.
2030 This means three calls to sasl_listmech() without a specified mech_list.
2031 => SASL tests which mechs of all available mechs actually work
2032 => three warnings about OTP not working
2033 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2035 With the patch, sasl_listmech() also gets called three times. But now
2036 SASL's mech_list option is set to the server_mech specified in the the
2037 authenticator. Or in other words, the answer from sasl_listmech()
2038 gets limited to just the mech you're testing for (which is different
2040 => the return list contains just 'plain' or 'login', 'cram-md5' or
2041 nothing depending on the value of ob->server_mech.
2043 I've just tested the patch: Authentication still works fine,
2044 unavailable mechs specified in the exim configuration are still
2045 caught, and the auth.log warnings about OTP are gone."
2047 PH/25 When debugging is enabled, the contents of the command line are added
2048 to the debugging output, even when log_selector=+arguments is not
2051 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2052 answer is "GNU", and only if the return is "GNU/something" is the answer
2055 PH/27 $acl_verify_message is now set immediately after the failure of a
2056 verification in an ACL, and so is available in subsequent modifiers. In
2057 particular, the message can be preserved by coding like this:
2059 warn !verify = sender
2060 set acl_m0 = $acl_verify_message
2062 Previously, $acl_verify_message was set only while expanding "message"
2063 and "log_message" when a very denied access.
2065 PH/28 Modified OS/os.c-Linux with
2067 -#ifndef OS_LOAD_AVERAGE
2068 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2070 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2071 nomenclature these days.)
2073 PH/29 Installed patch from the Sieve maintainer that adds the options
2074 sieve_useraddress and sieve_subaddress to the redirect router.
2076 PH/30 In these circumstances:
2077 . Two addresses routed to the same list of hosts;
2078 . First host does not offer TLS;
2079 . First host accepts first address;
2080 . First host gives temporary error to second address;
2081 . Second host offers TLS and a TLS session is established;
2082 . Second host accepts second address.
2083 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2084 and peerdn, if requested) that were in fact used only for the second
2087 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2088 attention to any local part prefix or suffix that was matched by the
2089 router that accepted the address. It now behaves in the same way as it
2090 does for delivery: the affixes are removed from the local part unless
2091 rcpt_include_affixes is set on the transport.
2093 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2094 timeout during the DATA phase of an incoming message.
2096 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2097 applied a patch sanctioned by the Sieve maintainer.
2099 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2100 the uid or gid is negative. A case of a negative gid caused this to be
2101 noticed. The fix allows for either to be negative.
2103 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2104 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2107 PH/36 The change PH/12 above was broken. Fixed it.
2109 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2110 the grounds that routing the same address twice would always produce the
2111 same answer. This might have been true once, but it is certainly no
2112 longer true now. Routing a child address may depend on the previous
2113 routing that produced that child. Some complicated redirection strategies
2114 went wrong when messages had multiple recipients, and made Exim's
2115 behaviour dependent on the order in which the addresses were given.
2117 I have moved the duplicate checking until after the routing is complete.
2118 Exim scans the addresses that are assigned to local and remote
2119 transports, and removes any duplicates. This means that more work will be
2120 done, as duplicates will always all be routed, but duplicates are
2121 presumably rare, so I don't expect this is of any significance.
2123 For deliveries to pipes, files, and autoreplies, the duplicate checking
2124 still happens during the routing process, since they are not going to be
2127 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2128 It corrects a timeout issue with spamd. This is Ian's comment: "The
2129 background is that sometimes spamd either never reads data from a
2130 connection it has accepted, or it never writes response data. The exiscan
2131 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2132 blindly assumes that writes won't block so it may never time out."
2134 PH/39 Allow G after quota size as well as K and M.
2136 PH/40 The value set for $authenticated_id in an authenticator may not contain
2137 binary zeroes or newlines because the value is written to log lines and
2138 to spool files. There was no check on this. Now the value is run through
2139 the string_printing() function so that such characters are converted to
2140 printable escape sequences.
2142 PH/41 $message_linecount is a new variable that contains the total number of
2143 lines in the message. Compare $body_linecount, which is the count for the
2146 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2147 bounce and delay warning messages, except in certain special cases, which
2150 (a) An SMTP error message from a remote host;
2151 (b) A message specified in a :fail: redirection;
2152 (c) A message specified in a "fail" command in a system filter;
2153 (d) A message specified in a FAIL return from the queryprogram router;
2154 (e) A message specified by the cannot_route_message router option.
2156 In these cases only, Exim does include the error details in bounce and
2157 warning messages. There are also a few cases where bland messages such
2158 as "unrouteable address" or "local delivery error" are given.
2160 PH/43 $value is now also set for the "else" part of a ${run expansion.
2162 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2163 being worked on, but at least Exim now implements the latest version to
2166 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2167 process to complete was treated as a delivery failure, a timeout while
2168 writing the message to the pipe was logged, but erroneously treated as a
2169 successful delivery. Such timeouts include transport filter timeouts. For
2170 consistency with the overall process timeout, these timeouts are now
2171 treated as errors, giving rise to delivery failures by default. However,
2172 there is now a new Boolean option for the pipe transport called
2173 timeout_defer, which, if set TRUE, converts the failures into defers for
2174 both kinds of timeout. A transport filter timeout is now identified in
2177 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2178 systems where "make" and "gmake" are different, calling "gmake" at top
2179 level broke things. I've arranged for the value of $(MAKE) to be passed
2180 from the Makefile to this script so that it can call the same version of
2184 A note about Exim versions 4.44 and 4.50
2185 ----------------------------------------
2187 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2188 changes of various kinds. As a consequence, a big documentation update was
2189 needed. This delayed the release for rather longer than seemed good, especially
2190 in the light of a couple of (minor) security issues. Therefore, the changes
2191 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2192 release. So 4.44 and 4.50 are in effect two different branches that both start
2195 I have left the 4.50 change log unchanged; it contains all the changes since
2196 4.43. The change log for 4.44 is below; many of its items are identical to
2197 those for 4.50. This seems to be the most sensible way to preserve the
2198 historical information.
2204 1. Minor wording change to the doc/README.SIEVE file.
2206 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2207 computation of the current number of files was incorrect.
2209 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2210 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2213 4. Give more explanation in the error message when the command for a transport
2214 filter fails to execute.
2216 5. There are several places where Exim runs a non-Exim command in a
2217 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2218 was being done only for the command run by the queryprogram router. It is
2219 now done for all such subprocesses. The other cases are: ${run, transport
2220 filters, and the commands run by the lmtp and pipe transports.
2222 6. Added CONFIGURE_GROUP build-time option.
2224 7. Some older OS have a limit of 256 on the maximum number of file
2225 descriptors. Exim was using setrlimit() to set 1000 as a large value
2226 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2227 systems. I've change it so that if it can't get 1000, it tries for 256.
2229 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2230 was an oversight, and furthermore, ever since the addition of extra
2231 controls (e.g. 4.43/32), the checks on when to allow different forms of
2232 "control" were broken. There should now be diagnostics for all cases when a
2233 control that does not make sense is encountered.
2235 9. Added the /retain_sender option to "control=submission".
2237 10. $recipients is now available in the predata ACL (oversight).
2239 11. Tidy the search cache before the fork to do a delivery from a message
2240 received from the command line. Otherwise the child will trigger a lookup
2241 failure and thereby defer the delivery if it tries to use (for example) a
2242 cached ldap connection that the parent has called unbind on.
2244 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2245 of $address_data from the recipient verification was clobbered by the
2246 sender verification.
2248 13. The value of address_data from a sender verification is now available in
2249 $sender_address_data in subsequent conditions in the ACL statement.
2251 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2253 15. Added a new option "connect=<time>" to callout options, to set a different
2256 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2257 was its contents. (It was OK if the option was not defined at all.)
2259 17. A "Completed" log line is now written for messages that are removed from
2260 the spool by the -Mrm option.
2262 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2263 information about exactly what failed.
2265 19. Added -dd to debug only the daemon process.
2267 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2268 handles timeouts, both on the server side and network timeouts. Renamed the
2269 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2271 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2272 It is now set to "smtps".
2274 22. $host_address is now set to the target address during the checking of
2275 ignore_target_hosts.
2277 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2278 being passed; this would have caused $sender_host_name to have been used if
2279 matching the list had actually called for a host name (not very likely,
2280 since this list is usually IP addresses). A host name is now passed as
2283 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2284 code that specifies a non-restarting handler (typically sigaction() in
2285 modern systems) in an attempt to fix a rare and obscure crash bug.
2287 25. Narrowed the window for a race in the daemon that could cause it to ignore
2288 SIGCHLD signals. This is not a major problem, because they are used only to
2289 wake it up if nothing else does.
2291 26. A malformed maildirsize file could cause Exim to calculate negative values
2292 for the mailbox size or file count. Odd effects could occur as a result.
2293 The maildirsize information is now recalculated if the size or filecount
2296 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2297 support for a long time. Removed HAVE_SYS_VFS_H.
2299 28. Installed the latest version of exipick from John Jetmore.
2301 29. In an address list, if the pattern was not a regular expression, an empty
2302 subject address (from a bounce message) matched only if the pattern was an
2303 empty string. Non-empty patterns were not even tested. This was the wrong
2304 because it is perfectly reasonable to use an empty address as part of a
2305 database query. An empty address is now tested by patterns that are
2306 lookups. However, all the other forms of pattern expect the subject to
2307 contain a local part and a domain, and therefore, for them, an empty
2308 address still always fails if the pattern is not itself empty.
2310 30. Exim went into a mad DNS loop when attempting to do a callout where the
2311 host was specified on an smtp transport, and looking it up yielded more
2312 than one IP address.
2314 31. Re-factored the code for checking spool and log partition space into a
2315 function that finds that data and another that does the check. The former
2316 is then used to implement four new variables: $spool_space, $log_space,
2317 $spool_inodes, and $log_inodes.
2319 32. The RFC2047 encoding function was originally intended for short strings
2320 such as real names; it was not keeping to the 75-character limit for
2321 encoded words that the RFC imposes. It now respects the limit, and
2322 generates multiple encoded words if necessary. To be on the safe side, I
2323 have increased the buffer size for the ${rfc2047: expansion operator from
2326 33. It is now permitted to omit both strings after an "if" condition; if the
2327 condition is true, the result is "true". As before, when the second string
2328 is omitted, a false condition yields an empty string. This makes it less
2329 cumbersome to write custom ACL and router conditions.
2331 34. Failure to deliver a bounce message always caused it to be frozen, even if
2332 there was an errors_to setting on the router. The errors_to setting is now
2335 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2336 canonical form (fully expanded) before being placed in
2337 $sender_host_address.
2339 36. The table in the code that translates DNS record types into text (T_A to
2340 "A" for instance) was missing entries for NS and CNAME. It is just possible
2341 that this could have caused confusion if both these types were looked up
2342 for the same domain, because the text type is used as part of Exim's
2343 per-process caching. But the chance of anyone hitting this buglet seems
2346 37. The dnsdb lookup has been extended in a number of ways.
2348 (1) There is a new type, "zns", which walks up the domain tree until it
2349 finds some nameserver records. It should be used with care.
2351 (2) There is a new type, "mxh", which is like "mx" except that it returns
2352 just the host names, not the priorities.
2354 (3) It is now possible to give a list of domains (or IP addresses) to be
2355 looked up. The behaviour when one of the lookups defers can be
2356 controlled by a keyword.
2358 (4) It is now possible to specify the separator character for use when
2359 multiple records are returned.
2361 38. The dnslists ACL condition has been extended: it is now possible to supply
2362 a list of IP addresses and/or domains to be looked up in a particular DNS
2365 39. Added log_selector=+queue_time_overall.
2367 40. When running the queue in the test harness, wait just a tad after forking a
2368 delivery process, to get repeatability of debugging output.
2370 41. Include certificate and key file names in error message when GnuTLS fails
2371 to set them up, because the GnuTLS error message doesn't include the name
2372 of the failing file when there is a problem reading it.
2374 42. Allow both -bf and -bF in the same test run.
2376 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2378 44. The "Exiscan patch" is now merged into the mainline Exim source.
2380 45. Sometimes the final signoff response after QUIT could fail to get
2381 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2382 < 0 before doing a fflush(). This bug looks as though it goes back to the
2383 introduction of TLS in release 3.20, but "sometimes" must have been rare
2384 because the tests only now provoked it.
2386 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2387 (this can affect the format of dates).
2389 47. exim_tidydb, when checking for the continued existence of a message for
2390 which it has found a message-specific retry record, was not finding
2391 messages that were in split spool directories. Consequently, it was
2392 deleting retry records that should have stayed in existence.
2394 48. Steve fixed some bugs in eximstats.
2396 49. The SPA authentication driver was not abandoning authentication and moving
2397 on to the next authenticator when an expansion was forced to fail,
2398 contradicting the general specification for all authenticators. Instead it
2399 was generating a temporary error. It now behaves as specified.
2401 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2402 (the order specifies the preference for clients). The order is now AES256,
2403 AES128, 3DES, ARCFOUR128.
2405 51. Small patch to Sieve code - explicitly set From: when generating an
2408 52. Exim crashed if a remote delivery caused a very long error message to be
2409 recorded - for instance if somebody sent an entire SpamAssassin report back
2410 as a large number of 550 error lines. This bug was coincidentally fixed by
2411 increasing the size of one of Exim's internal buffers (big_buffer) that
2412 happened as part of the Exiscan merge. However, to be on the safe side, I
2413 have made the code more robust (and fixed the comments that describe what
2416 53. Now that there can be additional text after "Completed" in log lines (if
2417 the queue_time_overall log selector is set), a one-byte patch to exigrep
2418 was needed to allow it to recognize "Completed" as not the last thing in
2421 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2422 patch that reportedly fixes this has been added. I am not expert enough to
2423 create a test for it. This is what the patch creator wrote:
2425 "I found a little strange behaviour of ldap code when working with
2426 Windows 2003 AD Domain, where users was placed in more than one
2427 Organization Units. When I tried to give exim partial DN, the exit code
2428 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2429 But simultaneously result of request was absolutely normal ldap result,
2430 so I produce this patch..."
2432 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2433 REFERENCE, so I have modified the code to exclude the patch when that macro
2436 55. Some experimental protocols are using DNS PTR records for new purposes. The
2437 keys for these records are domain names, not reversed IP addresses. The
2438 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2439 leaves it alone. Component reversal etc. now happens only for IP addresses.
2442 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2444 57. Double the size of the debug message buffer (to 2048) so that more of very
2445 long debug lines gets shown.
2447 58. The exicyclog utility now does better if the number of log files to keep
2448 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2450 59. Two changes related to the smtp_active_hostname option:
2452 (1) $smtp_active_hostname is now available as a variable.
2453 (2) The default for smtp_banner uses $smtp_active_hostname instead
2454 of $primary_hostname.
2456 60. The host_aton() function is supposed to be passed a string that is known
2457 to be a valid IP address. However, in the case of IPv6 addresses, it was
2458 not checking this. This is a hostage to fortune. Exim now panics and dies
2459 if the condition is not met. A case was found where this could be provoked
2460 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2461 components; fortuitously, this particular loophole had already been fixed
2462 by change 4.50/55 above.
2464 If there are any other similar loopholes, the new check in host_aton()
2465 itself should stop them being exploited. The report I received stated that
2466 data on the command line could provoke the exploit when Exim was running as
2467 exim, but did not say which command line option was involved. All I could
2468 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2469 running as the user.
2472 61. There was a buffer overflow vulnerability in the SPA authentication code
2473 (which came originally from the Samba project). I have added a test to the
2474 spa_base64_to_bits() function which I hope fixes it.
2477 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2478 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2480 63. The daemon start-up calls getloadavg() while still root for those OS that
2481 need the first call to be done as root, but it missed one case: when
2482 deliver_queue_load_max is set with deliver_drop_privilege. This is
2483 necessary for the benefit of the queue runner, because there is no re-exec
2484 when deliver_drop_privilege is set.
2486 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2487 This has been fixed.
2489 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2490 was in use, was not putting the data itself into the right store pool;
2491 consequently, it could be overwritten for a subsequent message in the same
2492 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2495 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2497 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2498 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2499 the function were treating the return as a boolean value, which happened to
2500 work because 0=false and not-0=true, but is not correct code.
2502 68. The host_aton() function was not handling scoped IPv6 addresses (those
2503 with, for example, "%eth0" on the end) correctly.
2505 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2506 negated items (that is, ~something) in unsigned ints. Some compilers
2507 apparently mutter when there is no cast.
2509 70. If an address verification called from an ACL failed, and did not produce a
2510 user-specific message (i.e. there was only a "system" message), nothing was
2511 put in $acl_verify_message. In this situation, it now puts the system
2514 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2515 session; change 4.31/43 added the unwanted input to the log line - except
2516 that it did not do this in the start of session case. It now does.
2518 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2519 This is wrong and can cause the other end to generate a synchronization
2520 error if it is another Exim or anything else that does the synchronization
2521 check. A QUIT command is no longer sent after a timeout.
2523 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2524 during host lookups.
2526 74. The defer_ok option of callout verification was not working if it was used
2527 when verifying addresses in header lines, that is, for this case:
2529 verify = header_sender/callout=defer_ok
2531 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2532 those file descriptors could be used for SMTP connections. If anything
2533 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2534 could be sent to the SMTP client, causing chaos. The daemon now opens
2535 stdin, stdout, and stderr to /dev/null when it puts itself into the
2538 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2539 log by default. The user can override this with suitable Perl magic.
2541 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2542 the log message when discard triggers, was not working for the DATA ACL or
2543 for the non-SMTP ACL.
2545 78. Error message wording change in sieve.c.
2547 79. If smtp_accept_max_per_host was set, the number of connections could be
2548 restricted to fewer than expected, because the daemon was trying to set up
2549 a new connection before checking whether the processes handling previous
2550 connections had finished. The check for completed processes is now done
2551 earlier. On busy systems, this bug wouldn't be noticed because something
2552 else would have woken the daemon, and it would have reaped the completed
2555 80. If a message was submitted locally by a user whose login name contained one
2556 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2557 It caused a spool format error. I have fixed the spool reading code. A
2558 related problem was that the "from" clause in the Received: line became
2559 illegal because of the space(s). It is now covered by ${quote_local_part.
2561 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2564 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2566 83. Patch from Sieve maintainer to fix unterminated string problem in
2567 "vacation" handling.
2569 84. Some minor changes to the Linux configuration files to help with other
2570 OS variants using glibc.
2572 85. One more patch for Sieve to update vacation handling to latest spec.
2575 ----------------------------------------------------
2576 See the note above about the 4.44 and 4.50 releases.
2577 ----------------------------------------------------
2583 1. Change 4.43/35 introduced a bug that caused file counts to be
2584 incorrectly computed when quota_filecount was set in an appendfile
2587 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2588 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2591 3. Give more explanation in the error message when the command for a transport
2592 filter fails to execute.
2594 4. There are several places where Exim runs a non-Exim command in a
2595 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2596 was being done only for the command run by the queryprogram router. It is
2597 now done for all such subprocesses. The other cases are: ${run, transport
2598 filters, and the commands run by the lmtp and pipe transports.
2600 5. Some older OS have a limit of 256 on the maximum number of file
2601 descriptors. Exim was using setrlimit() to set 1000 as a large value
2602 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2603 systems. I've change it so that if it can't get 1000, it tries for 256.
2605 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2606 was an oversight, and furthermore, ever since the addition of extra
2607 controls (e.g. 4.43/32), the checks on when to allow different forms of
2608 "control" were broken. There should now be diagnostics for all cases when a
2609 control that does not make sense is encountered.
2611 7. $recipients is now available in the predata ACL (oversight).
2613 8. Tidy the search cache before the fork to do a delivery from a message
2614 received from the command line. Otherwise the child will trigger a lookup
2615 failure and thereby defer the delivery if it tries to use (for example) a
2616 cached ldap connection that the parent has called unbind on.
2618 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2619 of $address_data from the recipient verification was clobbered by the
2620 sender verification.
2622 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2623 was its contents. (It was OK if the option was not defined at all.)
2625 11. A "Completed" log line is now written for messages that are removed from
2626 the spool by the -Mrm option.
2628 12. $host_address is now set to the target address during the checking of
2629 ignore_target_hosts.
2631 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2632 being passed; this would have caused $sender_host_name to have been used if
2633 matching the list had actually called for a host name (not very likely,
2634 since this list is usually IP addresses). A host name is now passed as
2637 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2638 code that specifies a non-restarting handler (typically sigaction() in
2639 modern systems) in an attempt to fix a rare and obscure crash bug.
2641 15. Narrowed the window for a race in the daemon that could cause it to ignore
2642 SIGCHLD signals. This is not a major problem, because they are used only to
2643 wake it up if nothing else does.
2645 16. A malformed maildirsize file could cause Exim to calculate negative values
2646 for the mailbox size or file count. Odd effects could occur as a result.
2647 The maildirsize information is now recalculated if the size or filecount
2650 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2651 support for a long time. Removed HAVE_SYS_VFS_H.
2653 18. Updated exipick to current release from John Jetmore.
2655 19. Allow an empty sender to be matched against a lookup in an address list.
2656 Previously the only cases considered were a regular expression, or an
2659 20. Exim went into a mad DNS lookup loop when doing a callout where the
2660 host was specified on the transport, if the DNS lookup yielded more than
2663 21. The RFC2047 encoding function was originally intended for short strings
2664 such as real names; it was not keeping to the 75-character limit for
2665 encoded words that the RFC imposes. It now respects the limit, and
2666 generates multiple encoded words if necessary. To be on the safe side, I
2667 have increased the buffer size for the ${rfc2047: expansion operator from
2670 22. Failure to deliver a bounce message always caused it to be frozen, even if
2671 there was an errors_to setting on the router. The errors_to setting is now
2674 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2675 canonical form (fully expanded) before being placed in
2676 $sender_host_address.
2678 24. Updated eximstats to version 1.33
2680 25. Include certificate and key file names in error message when GnuTLS fails
2681 to set them up, because the GnuTLS error message doesn't include the name
2682 of the failing file when there is a problem reading it.
2684 26. Expand error message when OpenSSL has problems setting up cert/key files.
2687 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2688 (this can affect the format of dates).
2690 28. exim_tidydb, when checking for the continued existence of a message for
2691 which it has found a message-specific retry record, was not finding
2692 messages that were in split spool directories. Consequently, it was
2693 deleting retry records that should have stayed in existence.
2695 29. eximstats updated to version 1.35
2696 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2697 1.35 - bugfix such that pie charts by volume are generated correctly
2699 30. The SPA authentication driver was not abandoning authentication and moving
2700 on to the next authenticator when an expansion was forced to fail,
2701 contradicting the general specification for all authenticators. Instead it
2702 was generating a temporary error. It now behaves as specified.
2704 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2705 (the order specifies the preference for clients). The order is now AES256,
2706 AES128, 3DES, ARCFOUR128.
2708 31. Small patch to Sieve code - explicitly set From: when generating an
2711 32. Exim crashed if a remote delivery caused a very long error message to be
2712 recorded - for instance if somebody sent an entire SpamAssassin report back
2713 as a large number of 550 error lines. This bug was coincidentally fixed by
2714 increasing the size of one of Exim's internal buffers (big_buffer) that
2715 happened as part of the Exiscan merge. However, to be on the safe side, I
2716 have made the code more robust (and fixed the comments that describe what
2719 33. Some experimental protocols are using DNS PTR records for new purposes. The
2720 keys for these records are domain names, not reversed IP addresses. The
2721 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2722 leaves it alone. Component reversal etc. now happens only for IP addresses.
2725 34. The host_aton() function is supposed to be passed a string that is known
2726 to be a valid IP address. However, in the case of IPv6 addresses, it was
2727 not checking this. This is a hostage to fortune. Exim now panics and dies
2728 if the condition is not met. A case was found where this could be provoked
2729 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2730 components; fortuitously, this particular loophole had already been fixed
2731 by change 4.50/55 or 4.44/33 above.
2733 If there are any other similar loopholes, the new check in host_aton()
2734 itself should stop them being exploited. The report I received stated that
2735 data on the command line could provoke the exploit when Exim was running as
2736 exim, but did not say which command line option was involved. All I could
2737 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2738 running as the user.
2741 35. There was a buffer overflow vulnerability in the SPA authentication code
2742 (which came originally from the Samba project). I have added a test to the
2743 spa_base64_to_bits() function which I hope fixes it.
2746 36. The daemon start-up calls getloadavg() while still root for those OS that
2747 need the first call to be done as root, but it missed one case: when
2748 deliver_queue_load_max is set with deliver_drop_privilege. This is
2749 necessary for the benefit of the queue runner, because there is no re-exec
2750 when deliver_drop_privilege is set.
2752 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2753 was in use, was not putting the data itself into the right store pool;
2754 consequently, it could be overwritten for a subsequent message in the same
2755 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2758 38. Sometimes the final signoff response after QUIT could fail to get
2759 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2760 < 0 before doing a fflush(). This bug looks as though it goes back to the
2761 introduction of TLS in release 3.20, but "sometimes" must have been rare
2762 because the tests only now provoked it.
2768 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2769 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2770 Now it returns an integer. A number of calls were still expecting a T/F
2771 return. Fortuitously, in all cases, the tests worked in OK situations,
2772 which is the norm. However, things would have gone wrong on any write
2773 failures on the smtp file descriptor. This function is used when sending
2774 messages over SMTP and also when doing verify callouts.
2776 2. When Exim is called to do synchronous delivery of a locally submitted
2777 message (the -odf or -odi options), it no longer closes stderr before doing
2780 3. Implemented the mua_wrapper option.
2782 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2784 5. Implemented the functions header_remove(), header_testname(),
2785 header_add_at_position(), and receive_remove_recipient(), and exported them
2788 6. If an ACL "warn" statement specified the addition of headers, Exim already
2789 inserted X-ACL-Warn: at the start if there was no header name. However, it
2790 was not making this test for the second and subsequent header lines if
2791 there were newlines in the string. This meant that an invalid header could
2792 be inserted if Exim was badly configured.
2794 7. Allow an ACL "warn" statement to add header lines at the start or after all
2795 the Received: headers, as well as at the end.
2797 8. Added the rcpt_4xx retry error code.
2799 9. Added postmaster_mailfrom=xxx to callout verification option.
2801 10. Added mailfrom=xxxx to the callout verification option, for verify=
2804 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2805 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2807 12. Inserted some casts to stop certain compilers warning when using pointer
2808 differences as field lengths or precisions in printf-type calls (mostly
2809 affecting debugging statements).
2811 13. Added optional readline() support for -be (dynamically loaded).
2813 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2814 same clock tick as a message's arrival, so that its received time was the
2815 same as the "first fail" time on the retry record, and that message
2816 remained on the queue past the ultimate address timeout, every queue runner
2817 would try a delivery (because it was past the ultimate address timeout) but
2818 after another failure, the ultimate address timeout, which should have then
2819 bounced the address, did not kick in. This was a "< instead of <=" error;
2820 in most cases the first failure would have been in the next clock tick
2821 after the received time, and all would be well.
2823 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2824 being recognized when the domain list was tested by the match_domain
2825 condition in an expansion string.
2827 16. Added the ${str2b64: operator.
2829 17. Exim was always calling setrlimit() to set a large limit for the number of
2830 processes, without checking whether the existing limit was already
2831 adequate. (It did check for the limit on file descriptors.) Furthermore,
2832 errors from getrlimit() and setrlimit() were being ignored. Now they are
2833 logged to the main and panic logs, but Exim does carry on, to try to do its
2834 job under whatever limits there are.
2836 18. Imported PCRE 5.0.
2838 19. Trivial typo in log message " temporarily refused connection" (the leading
2841 20. If the log selector return_path_on_delivery was set and an address was
2842 redirected to /dev/null, the delivery process crashed because it assumed
2843 that a return path would always be set for a "successful" delivery. In this
2844 case, the whole delivery is bypassed as an optimization, and therefore no
2847 21. Internal re-arrangement: the function for sending a challenge and reading
2848 a response while authentication was assuming a zero-terminated challenge
2849 string. It's now changed to take a pointer and a length, to allow for
2850 binary data in such strings.
2852 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2854 23. Exim was not respecting finduser_retries when seeking the login of the
2855 uid under which it was called; it was always trying 10 times. (The default
2856 setting of finduser_retries is zero.) Also, it was sleeping after the final
2857 failure, which is pointless.
2859 24. Implemented tls_on_connect_ports.
2861 25. Implemented acl_smtp_predata.
2863 26. If the domain in control=submission is set empty, Exim assumes that the
2864 authenticated id is a complete email address when it generates From: or
2865 Sender: header lines.
2867 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2868 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2869 chown and chgrp in /bin and hostname in /usr/bin.
2871 28. Exim was keeping the "process log" file open after each use, just as it
2872 does for the main log. This opens the possibility of it remaining open for
2873 long periods when the USR1 signal hits a daemon. Occasional processlog
2874 errors were reported, that could have been caused by this. Anyway, it seems
2875 much more sensible not to leave this file open at all, so that is what now
2878 29. The long-running daemon process does not normally write to the log once it
2879 has entered its main loop, and it closes the log before doing so. This is
2880 so that log files can straightforwardly be renamed and moved. However,
2881 there are a couple of unusual error situations where the daemon does write
2882 log entries, and I had neglected to close the log afterwards.
2884 30. The text of an SMTP error response that was received during a remote
2885 delivery was being truncated at 512 bytes. This is too short for some of
2886 the long messages that one sometimes sees. I've increased the limit to
2889 31. It is now possible to make retry rules that apply only when a message has a
2890 specific sender, in particular, an empty sender.
2892 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2893 it possible to be selective about when SMTP synchronization is enforced.
2895 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2897 32. Implemented hosts_connection_nolog.
2899 33. Added an ACL for QUIT.
2901 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2904 35. Added mailbox_size and mailbox_filecount to appendfile.
2906 36. Added control = no_multiline_responses to ACLs.
2908 37. There was a bug in the logic of the code that waits for the clock to tick
2909 in the case where the clock went backwards by a substantial amount such
2910 that the microsecond fraction of "now" was more than the microsecond
2911 fraction of "then" (but the whole seconds number was less).
2913 38. Added support for the libradius Radius client library this is found on
2914 FreeBSD (previously only the radiusclient library was supported).
2920 1. When certain lookups returned multiple values in the form name=value, the
2921 quoting of the values was not always being done properly. Specifically:
2922 (a) If the value started with a double quote, but contained no whitespace,
2924 (b) If the value contained whitespace other than a space character (i.e.
2925 tabs or newlines or carriage returns) it was not quoted.
2926 This fix has been applied to the mysql and pgsql lookups by writing a
2927 separate quoting function and calling it from the lookup code. The fix
2928 should probably also be applied to nisplus, ibase and oracle lookups, but
2929 since I cannot test any of those, I have not disturbed their existing code.
2931 2. A hit in the callout cache for a specific address caused a log line with no
2932 reason for rejecting RCPT. Now it says "Previous (cached) callout
2933 verification failure".
2935 3. There was an off-by-one bug in the queryprogram router. An over-long
2936 return line was truncated at 256 instead of 255 characters, thereby
2937 overflowing its buffer with the terminating zero. As well as fixing this, I
2938 have increased the buffer size to 1024 (and made a note to document this).
2940 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2941 when Exim is waiting for an SMTP response from a remote server, Exim
2942 restarts its select() call on the socket, thereby resetting its timeout.
2943 This is not a problem when such interrupts are rare. Somebody set up a cron
2944 job to run exiwhat every 2 minutes, which is less than the normal select()
2945 timeout (5 or 10 minutes). This meant that the select() timeout never
2946 kicked in because it was always reset. I have fixed this by comparing the
2947 time when an interrupt arrives with the time at the start of the first call
2948 to select(). If more time than the timeout has elapsed, the interrupt is
2949 treated as a timeout.
2951 5. Some internal re-factoring in preparation for the addition of Sieve
2952 extensions (by MH). In particular, the "personal" test is moved to a
2953 separate function, and given an option for scanning Cc: and Bcc: (which is
2954 not set for Exim filters).
2956 6. When Exim created an email address using the login of the caller as the
2957 local part (e.g. when creating a From: or Sender: header line), it was not
2958 quoting the local part when it contained special characters such as @.
2960 7. Installed new OpenBSD configuration files.
2962 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2963 try to make them clearer.
2965 9. Callout options, other than the timeout value, were being ignored when
2966 verifying sender addresses in header lines. For example, when using
2968 verify = header_sender/callout=no_cache
2970 the cache was (incorrectly) being used.
2972 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2973 only the Cygwin environment.
2975 11. When return_path_on_delivery was set as a log selector, if different remote
2976 addresses in the same message used different return paths and parallel
2977 remote delivery occurred, the wrong values would sometimes be logged.
2978 (Whenever a remote delivery process finished, the return path value from
2979 the most recently started remote delivery process was logged.)
2981 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2982 lines when AUTH and/or TLS are in use. This is the "received protocol"
2983 field. Exim used to use "asmtp" for authenticated SMTP, without any
2984 indication (in the protocol name) for TLS use. Now it follows the RFC and
2985 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2986 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2987 names appear in log lines as well as in Received: header lines.
2989 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2990 extensions, and comparison tests, and to fix some bugs.
2992 14. Changes to the "personal" filter test:
2994 (1) The test was buggy in that it was just doing the equivalent of
2995 "contains" tests on header lines. For example, if a user's address was
2996 anne@some.where, the "personal" test would incorrectly be true for
2998 To: susanne@some.where
3000 This test is now done by extracting each address from the header in turn,
3001 and checking the entire address. Other tests that are part of "personal"
3002 are now done using regular expressions (for example, to check local parts
3003 of addresses in From: header lines).
3005 (2) The list of non-personal local parts in From: addresses has been
3006 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3007 taken from the Sieve specification recommendations.
3009 (3) If the message contains any header line starting with "List-" it is
3010 treated as non-personal.
3012 (4) The test for "circular" in the Subject: header line has been removed
3013 because it now seems ill-conceived.
3015 15. Minor typos in src/EDITME comments corrected.
3017 16. Installed latest exipick from John Jetmore.
3019 17. If headers_add on a router specified a text string that was too long for
3020 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3021 of string_sprintf() is now avoided.
3023 18. $message_body_size was not set (it was always zero) when running the DATA
3024 ACL and the local_scan() function.
3026 19. For the "mail" command in an Exim filter, no default was being set for
3027 the once_repeat time, causing a random time value to be used if "once" was
3028 specified. (If the value happened to be <= 0, no repeat happened.) The
3029 default is now 0s, meaning "never repeat". The "vacation" command was OK
3030 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3031 (I found it when inspecting the code).
3033 20. There is now an overall timeout for performing a callout verification. It
3034 defaults to 4 times the callout timeout, which applies to individual SMTP
3035 commands during the callout. The overall timeout applies when there is more
3036 than one host that can be tried. The timeout is checked before trying the
3037 next host. This prevents very long delays if there are a large number of
3038 hosts and all are timing out (e.g. when the network connections are timing
3039 out). The value of the overall timeout can be changed by specifying an
3040 additional sub-option for "callout", called "maxwait". For example:
3042 verify = sender/callout=5s,maxwait=20s
3044 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3045 to the end before writing, but this should make it even safer).
3047 22. Exim was forgetting that it had advertised PIPELINING for the second and
3048 subsequent messages on an SMTP connection. It was also not resetting its
3049 memory on STARTTLS and an internal HELO.
3051 23. When Exim logs an SMTP synchronization error within a session, it now
3052 records whether PIPELINING has been advertised or not.
3054 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3055 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3056 rather than long int.
3058 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3060 26. Added the never_mail option to autoreply.
3066 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3067 crash if the getsockname() call failed; this can happen if a connection is
3068 closed very soon after it is established. The problem was simply in the
3069 order in which certain operations were done, causing Exim to try to write
3070 to the SMTP stream before it had set up the file descriptor. The bug has
3071 been fixed by making things happen in the correct order.
3077 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3078 before the connection was closed, thus losing the rejection response.
3080 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3081 some early Solaris releases, but causes trouble in current releases where
3082 socklen_t is defined.
3084 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3087 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3090 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3091 that releasing the top of it at the end releases what was used for sub-
3092 expansions (unless the block got too big). However, discard this block if
3093 the first thing is a variable or header, so that we can use its block when
3094 it is dynamic (useful for very large $message_headers, for example).
3096 6. Lookups now cache *every* query, not just the most recent. A new, separate
3097 store pool is used for this. It can be recovered when all lookup caches are
3098 flushed. Lookups now release memory at the end of their result strings.
3099 This has involved some general refactoring of the lookup sources.
3101 7. Some code has been added to the store_xxx() functions to reduce the amount
3102 of flapping under certain conditions.
3104 8. log_incoming_interface used to affect only the <= reception log lines. Now
3105 it causes the local interface and port to be added to several more SMTP log
3106 lines, for example "SMTP connection from", and rejection lines.
3108 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3110 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3112 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3113 could be overwritten at the end of the current message (or the start of a
3114 new message if it was set in a HELO ACL). The value is now preserved for
3115 the duration of the SMTP connection.
3117 12. If a transport had a headers_rewrite setting, and a matching header line
3118 contained an unqualified address, that address was qualified, even if it
3119 did not match any rewriting rules. The underlying bug was that the values
3120 of the flags that permit the existence of unqualified sender and recipient
3121 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3122 non-local messages, and by -bnq for local messages) were not being
3123 preserved with the message after it was received.
3125 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3126 "next input=" as part of the text comprising the host identity instead of
3127 the correct text. The code was using the same buffer for two different
3128 strings. However, depending on which order the printing function evaluated
3129 its arguments, the bug did not always show up. Under Linux, for example, my
3130 test suite worked just fine.
3132 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3133 doesn't work with some older versions of Perl. It has been changed to "my",
3134 which in any case is probably the better facility to use.
3136 15. A really picky compiler found some instances of statements for creating
3137 error messages that either had too many or two few arguments for the format
3140 16. The size of the buffer for calls to the DNS resolver has been increased
3141 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3142 for addresses that have a lot of PTR records. This alleviates a problem; it
3143 does not fully solve it.
3145 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3146 buffer now truncates the list and logs the incident, which is the same
3147 action as happens when Exim is looking up a host name and its aliases.
3148 Previously in this situation something unpredictable would happen;
3149 sometimes it was "internal error: store_reset failed".
3151 18. If a server dropped the connection unexpectedly when an Exim client was
3152 using GnuTLS and trying to read a response, the client delivery process
3153 crashed while trying to generate an error log message.
3155 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3156 string, for example:
3158 warn message = H1: something\nH2: something
3160 the text was added as a single header line from Exim's point of view
3161 though it ended up OK in the delivered message. However, searching for the
3162 second and subsequent header lines using $h_h2: did not work. This has been
3163 fixed. Similarly, if a system filter added multiple headers in this way,
3164 the routers could not see them.
3166 20. Expanded the error message when iplsearch is called with an invalid key to
3167 suggest using net-iplsearch in a host list.
3169 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3170 ACLs are no longer actually imposed (and a message to that effect is
3173 22. If a "gecos" field in a passwd entry contained escaped characters, in
3174 particular, if it contained a \" sequence, Exim got it wrong when building
3175 a From: or a Sender: header from that name. A second bug also caused
3176 incorrect handling when an unquoted " was present following a character
3177 that needed quoting.
3179 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3180 was not being matched caselessly.
3182 24. Arranged for all hyphens in the exim.8 source to be escaped with
3185 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3186 properly. Recipient callout cache records were still being keyed to include
3187 the sender, even when use_sender was set false. This led to far more
3188 callouts that were necessary. The sender is no longer included in the key
3189 when use_sender is false.
3191 26. Added "control = submission" modifier to ACLs.
3193 27. Added the ${base62d: operator to decode base 62 numbers.
3195 28. dnsdb lookups can now access SRV records.
3197 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3198 the configuration file.
3200 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3201 (-v) mode. This makes the output for a verbose queue run more intelligible.
3203 31. Added a use_postmaster feature to recipient callouts.
3205 32. Added the $body_zerocount variable, containing the number of binary zero
3206 bytes in the message body.
3208 33. The time of last modification of the "new" subdirectory is now used as the
3209 "mailbox time last read" when there is a quota error for a maildir
3212 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3214 35. Added +ignore_unknown as a special item in host lists.
3216 36. Code for decoding IPv6 addresses in host lists is now included, even if
3217 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3218 address was recognized as an IP address, but was then not correctly decoded
3219 into binary, causing unexpected and incorrect effects when compared with
3226 1. Very minor rewording of debugging text in manualroute to say "list of
3227 hosts" instead of "hostlist".
3229 2. If verify=header_syntax was set, and a header line with an unqualified
3230 address (no domain) and a large number of spaces between the end of the
3231 name and the colon was received, the reception process suffered a buffer
3232 overflow, and (when I tested it) crashed. This was caused by some obsolete
3233 code that should have been removed. The fix is to remove it!
3235 3. When running in the test harness, delay a bit after writing a bounce
3236 message to get a bit more predictability in the log output.
3238 4. Added a call to search_tidyup() just before forking a reception process. In
3239 theory, someone could use a lookup in the expansion of smtp_accept_max_
3240 per_host which, without the tidyup, could leave open a database connection.
3242 5. Added the variables $recipient_data and $sender_data which get set from a
3243 lookup success in an ACL "recipients" or "senders" condition, or a router
3244 "senders" option, similar to $domain_data and $local_part_data.
3246 6. Moved the writing of debug_print from before to after the "senders" test
3249 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3250 problems for message scanning, either using a data ACL, or using
3251 local_scan() because the Received: header was not generated till after they
3252 were called (in order to set the time as the time of reception completion).
3253 I have revised the way this works. The header is now generated after the
3254 body is received, but before the ACL or local_scan() are called. After they
3255 are run, the timestamp in the header is updated.
3261 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3262 before starting a queue runner without re-exec. This happened only when
3263 deliver_drop_privilege was set or when the Exim user was set to root. The
3264 effect of the bug was that timeouts during subsequent deliveries caused
3265 crashes instead of being properly handled. The handler is now left at its
3266 default (and expected) setting.
3268 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3269 message, again when deliver_drop_privilege is set or Exim is run as root.
3270 The bug described in (1) was not present in this case, but the tidying up
3271 of the other signals was missing. I have made the two cases consistent.
3273 3. The ignore_target_hosts setting on a manualroute router was being ignored
3274 for hosts that were looked up using the /MX notation.
3276 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3279 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3280 operated on the sender address. After changing the $sender_address to <>
3281 for the sender address verify, Exim was re-instated it as the original
3282 (before rewriting) address, but remembering that it had rewritten it, so it
3283 wasn't rewriting it again. This bug also had the effect of breaking the
3284 sender address verification caching when the sender address was rewritten.
3286 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3287 This has been changed so that if the ip literal address matches
3288 ignore_target_hosts, the router declines.
3290 7. Added expansion conditions match_domain, match_address, and match_local_
3291 part (NOT match_host).
3293 8. The placeholder for the Received: header didn't have a length field set.
3295 9. Added code to Exim itself and to exim_lock to test for a specific race
3296 condition that could lead to file corruption when using MBX delivery. The
3297 issue is with the lockfile that is created in /tmp. If this file is removed
3298 after a process has opened it but before that process has acquired a lock,
3299 there is the potential for a second process to recreate the file and also
3300 acquire a lock. This could lead to two Exim processes writing to the file
3301 at the same time. The added code performs the same test as UW imapd; it
3302 checks after acquiring the lock that its file descriptor still refers to
3303 the same named file.
3305 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3306 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3309 11. Added the smtp_active_hostname option. If used, this will typically be made
3310 to depend on the incoming interface address. Because $interface_address is
3311 not set up until the daemon has forked a reception process, error responses
3312 that can happen earlier (such as "too many connections") no longer contain
3315 12. If an expansion in a condition on a "warn" statement fails because a lookup
3316 defers, the "warn" statement is abandoned, and the next ACL statement is
3317 processed. Previously this caused the whole ACL to be aborted.
3319 13. Added the iplsearch lookup type.
3321 14. Added ident_timeout as a log selector.
3323 15. Added tls_certificate_verified as a log selector.
3325 16. Added a global option tls_require_ciphers (compare the smtp transport
3326 option of the same name). This controls incoming TLS connections.
3328 17. I finally figured out how to make tls_require_ciphers do a similar thing
3329 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3330 before starting the TLS session.
3332 18. Tabs are now shown as \t in -bP output.
3334 19. If the log selector return_path_on_delivery was set, Exim crashed when
3335 bouncing a message because it had too many Received: header lines.
3337 20. If two routers both had headers_remove settings, and the first one included
3338 a superfluous trailing colon, the final name in the first list and the
3339 first name in the second list were incorrectly joined into one item (with a
3340 colon in the middle).
3346 1. Added -C and -D options to the exinext utility, mainly to make it easier
3347 to include in the automated testing, but these could be helpful when
3348 multiple configurations are in use.
3350 2. The exinext utility was not formatting the output nicely when there was
3351 an alternate port involved in the retry record key, nor when there was a
3352 message id as well (for retries that were specific to a specific message
3353 and a specific host). It was also confused by IPv6 addresses, because of
3354 the additional colons they contain. I have fixed the IPv4 problem, and
3355 patched it up to do a reasonable job for IPv6.
3357 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3358 delivery, the log line now contains "pipelined" if PIPELINING was used.
3360 4. An SMTP transport process used to panic and die if the bind() call to set
3361 an explicit outgoing interface failed. This has been changed; it is now
3362 treated in the same way as a connect() failure.
3364 5. A reference to $sender_host_name in the part of a conditional expansion
3365 that was being skipped was still causing a DNS lookup. This no longer
3368 6. The def: expansion condition was not recognizing references to header lines
3369 that used bh_ and bheader_.
3371 7. Added the _cache feature to named lists.
3373 8. The code for checking quota_filecount in the appendfile transport was
3374 allowing one more file than it should have been.
3376 9. For compatibility with Sendmail, the command line option
3384 and sets the incoming protocol and host name (for trusted callers). The
3385 host name and its colon can be omitted when only the protocol is to be set.
3386 Note the Exim already has two private options, -pd and -ps, that refer to
3387 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3388 "s", but I don't think that's a major issue.
3390 10. A number of refactoring changes to the code, none of which should affect
3393 (a) The number of logging options was getting close to filling up the
3394 32-bit word that was used as a bit map. I have split them into two classes:
3395 those that are passed in the argument to log_write(), and those that are
3396 only ever tested independently outside of that function. These are now in
3397 separate 32-bit words, so there is plenty of room for expansion again.
3398 There is no change in the user interface or the logging behaviour.
3400 (b) When building, for example, log lines, the code previously used a
3401 macro that called string_cat() twice, in order to add two strings. This is
3402 not really sufficiently general. Furthermore, there was one instance where
3403 it was actually wrong because one of the argument was used twice, and in
3404 one call a function was used. (As it happened, calling the function twice
3405 did not affect the overall behaviour.) The macro has been replaced by a
3406 function that can join an arbitrary number of extra strings onto a growing
3409 (c) The code for expansion conditions now uses a table and a binary chop
3410 instead of a serial search (which was left over from when there were very
3411 few conditions). Also, it now recognizes conditions like "pam" even when
3412 the relevant support is not compiled in: a suitably worded error message is
3413 given if an attempt is made to use such a condition.
3415 11. Added ${time_interval:xxxxx}.
3417 12. A bug was causing one of the ddress fields not to be passed back correctly
3418 from remote delivery subprocesses. The field in question was not being
3419 subsequently used, so this caused to problems in practice.
3421 13. Added new log selectors queue_time and deliver_time.
3423 14. Might have fixed a bug in maildirsizefile handling that threw up
3424 "unexpected character" debug warnings, and recalculated the data
3425 unnecessarily. In any case, I expanded the warning message to give more
3428 15. Added the message "Restricted characters in address" to the statements in
3429 the default ACL that block characters like @ and % in local parts.
3431 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3432 Three changes have been made:
3434 (a) There was a serious bug; a negative response to MAIL caused the whole
3435 recipient domain to be cached as invalid, thereby blocking all messages
3436 to all local parts at the same domain, from all senders. This bug has
3437 been fixed. The domain is no longer cached after a negative response to
3438 MAIL if the sender used is not empty.
3440 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3443 (c) A new callout option, "use_sender" has been added for people who want
3444 the modified behaviour.
3450 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3453 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3454 indeed breaks things for older releases.
3456 3. Added additional logging to the case where there is a problem reading data
3457 from a filter that is running in a subprocess using a pipe, in order to
3458 try to track down a specific problem.
3460 4. Testing facility fudge: when running in the test harness and attempting
3461 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3462 getting "No route to host". Convert this to a timeout.
3464 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3467 6. Some OS don't have socklen_t but use size_t instead. This affects the
3468 fifth argument of getsockopt() amongst other things. This is now
3469 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3470 can be set for individual OS. I have set it for SunOS5, OSF1, and
3471 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3472 some earlier ones do not.
3474 7. Change 4.30/15 was not doing the test caselessly.
3476 8. The standard form for an IPv6 address literal was being rejected by address
3477 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3478 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3479 this, as well as the form without the "IPv6" on the front (but only when
3480 address literals are enabled, of course).
3482 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3484 10. Exim crashed if a message with an empty sender address specified by -f
3485 encountered a router with an errors_to setting. This could be provoked only
3486 by a command such as
3490 where an empty string was supplied; "<>" did not hit this bug.
3492 11. Installed PCRE release 4.5.
3494 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3495 remained set. It is now erased.
3497 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3498 times from message ids (which are base 36 rather than the normal 62).
3500 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3501 were being counted as actual protocol errors, and logged if the log
3502 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3503 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3504 and DATA following a set of rejected RCPTs do not count as protocol errors.
3505 In other words, Exim assumes they were pipelined, though this may not
3506 actually be the case. Of course, in all cases the client gets an
3507 appropriate error code.
3509 15. If a lookup fails in an ACL condition, a message about the failure may
3510 be available; it is used if testing the ACL cannot continue, because most
3511 such messages specify what the cause of the deferral is. However, some
3512 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3513 that caused an old message to be retained and used if a later statement
3514 caused a defer, replacing the real cause of the deferral.
3516 16. If an IP address had so many PTR records that the DNS lookup buffer
3517 was not large enough to hold them, Exim could crash while trying to process
3518 the truncated data. It now detects and logs this case.
3520 17. Further to 4.21/58, another change has been made: if (and only if) the
3521 first line of a message (the first header line) ends with CRLF, a bare LF
3522 in a subsequent header line has a space inserted after it, so as not to
3523 terminate the header.
3525 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3526 unnecessarily, used atoi() instead of strtol(), and didn't check the
3527 termination when getting file sizes from file names by regex.
3529 19. Completely re-implemented the support for maildirsize files, in the light
3530 of a number of problems with the previous contributed implementation
3531 (4.30/29). In particular:
3533 . If the quota is zero, the maildirsize file is maintained, but no quota is
3536 . If the maildir directory does not exist, it is created before any attempt
3537 to write a maildirsize file.
3539 . The quota value in the file is just a cache; if the quota is changed in
3540 the transport, the new value overrides.
3542 . A regular expression is available for excluding directories from the
3545 20. The autoreply transport checks the characters in options that define the
3546 message's headers; it allows continued headers, but it was checking with
3547 isspace() after an embedded newline instead of explicitly looking for a
3550 21. If all the "regular" hosts to which an address was routed had passed their
3551 expiry times, and had not reached their retry times, the address was
3552 bounced, even if fallback hosts were defined. Now Exim should go on to try
3555 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3556 equivalent code in the SMTP transport. Some hosts send humungous responses
3557 to HELO/EHLO, more than 1024 it seems.
3559 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3560 gives compiler warnings in some environments. I've now done it "properly",
3563 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3564 (because of problems with the built-in one) was declared to return uschar *
3565 instead of char *, causing compiler failure.
3567 25. Fixed a file descriptor leak when processing alias/forward files.
3569 26. Fixed a minor format string issue in dbfn.c.
3571 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3573 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3574 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3577 29. When a sender address is verified, it is cached, to save repeating the test
3578 when there is more than one recipient in a message. However, when the
3579 verification involves a callout, it is possible for different callout
3580 options to be set for different recipients. It is too complicated to keep
3581 track of this in the cache, so now Exim always runs a verification when a
3582 callout is required, relying on the callout cache for the optimization.
3583 The overhead is duplication of the address routing, but this should not be
3586 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3587 to be verified with callout=postmaster, and the main callout worked but the
3588 postmaster check failed, the verification correctly failed. However, if a
3589 subsequent RCPT command asked for sender verification *without* the
3590 postmaster check, incorrect caching caused this verification also to fail,
3593 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3594 it was not caching the DNS options (qualify_single, search_parents) that
3595 were used when the lookup failed. A subsequent lookup with different
3596 options therefore always gave the same answer, though there were cases
3597 where it should not have. (Example: a "domains = !$mx_any" option on a
3598 dnslookup router: the "domains" option is always processed without any
3599 widening, but the router might have qualify_single set.) Now Exim uses the
3600 cached value only when the same options are set.
3602 32. Added John Jetmore's "exipick" utility to the distribution.
3604 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3605 than a timeout (e.g. a certificate is required, and is not provided), an
3606 Exim server now closes the connection immediately. Previously it waited for
3607 the client to close - but if the client is SSL, it seems that they each
3608 wait for each other, leading to a delay before one of them times out.
3610 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3611 maintained 0.8.x compatibility because I don't think many are using it, and
3612 it is clearly obsolete.
3614 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3617 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3618 client certificate was expired. A simple patch fixes this, though I don't
3619 understand the full logic of why the verify callback is called multiple
3622 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3623 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3624 which causes problems with some clients (such as the Certicom SSL Plus
3625 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3626 disables the coutermeasure allowing Eudora to connect."
3628 38. Exim was not checking that a write() to a log file succeeded. This could
3629 lead to Bad Things if a log got too big, in particular if it hit a file
3630 size limit. Exim now panics and dies if it cannot write to a log file, just
3631 as it does if it cannot open a log file.
3633 39. Modified OS/Makefile-Linux so that it now contains
3635 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3637 The two -D definitions ensure that Exim is compiled with large file
3638 support, which makes it possible to handle log files that are bigger than
3641 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3642 instance) a domain was checked against a named list that involved a lookup,
3643 causing $domain_data to be set, then another domain was checked against the
3644 same list, then the first domain was re-checked, the value of $domain_data
3645 after the final check could be wrong. In particular, if the second check
3646 failed, it could be set empty. This bug probably also applied to
3649 41. The strip_trailing_dot option was not being applied to the address given
3650 with the -f command-line option.
3652 42. The code for reading a message's header from the spool was incrementing
3653 $received_count, but never initializing it. This meant that the value was
3654 incorrect (doubled) while delivering a message in the same process in which
3655 it was received. In the most common configuration of Exim, this never
3656 happens - a fresh exec is done - but it can happen when
3657 deliver_drop_privilege is set.
3659 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3660 it now includes up to 150 characters of the unexpected data in the log
3663 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3664 and building data strings. The size of both of these buffers was 10 000
3665 bytes - far larger than anybody would *ever* want, thought I. Needless to
3666 say, somebody hit the limit. I have increased the maximum line length to
3667 20 000 and the maximum data length of concatenated lines to 100 000. I have
3668 also fixed two bugs, because there was no checking on these buffers. Tsk,
3669 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3670 buffer is too small.
3672 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3673 lsearch lookups. Now it does.
3675 46. When parsing a route_list item in a manualroute router, a fixed-length
3676 buffer was used for the list of hosts. I made this 1024 bytes long,
3677 thinking that nobody would ever have a list of hosts that long. Wrong.
3678 Somebody had a whole pile of complicated expansion conditions, and the
3679 string was silently truncated, leading to an expansion error. It turns out
3680 that it is easier to change to an unlimited length (owing to other changes
3681 that have happened since this code was originally written) than to build
3682 structure for giving a limitation error. The length of the item that
3683 expands into the list of hosts is now unlimited.
3685 47. The lsearch lookup could not handle data where the length of text line was
3686 more than 4095 characters. Such lines were truncated, leading to shortened
3687 data being returned. It should now handle lines of any length.
3689 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3690 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3693 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3694 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3695 no longer happens in comment lines. A list of replacements is now placed
3696 at the head of all of the source files, except those whose only change is
3697 to replace PERL_COMMAND in the very first #! line.
3699 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3700 messages on the queue, with a bottom-up merge sort, using code contributed
3701 by Michael Haardt. This should make operations like -bp somewhat faster on
3702 large queues. It won't affect queue runners, except when queue_run_in_order
3705 51. Installed eximstats 1.31 in the distribution.
3707 52. Added support for SRV lookups to the dnslookup router.
3709 53. If an ACL referred to $message_body or $message_body_end, the value was not
3710 reset for any messages that followed in the same SMTP session.
3712 54. The store-handling optimization for building very long strings was not
3713 differentiating between the different store pools. I don't think this
3714 actually made any difference in practice, but I've tidied it.
3716 55. While running the routers to verify a sender address, $sender_address
3717 was still set to the sender address. This is wrong, because when routing to
3718 send a bounce to the sender, it would be empty. Therefore, I have changed
3719 it so that, while verifying a sender address, $sender_address is set to <>.
3720 (There is no change to what happens when verifying a recipient address.)
3722 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3723 target A or AAAA records (if not already returned) without resetting the
3724 qualify_single or search_parents options of the DNS resolver. These are
3725 inappropriate in this case because the targets of MX and SRV records must
3726 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3727 target that, when qualified, matched something in the local domain. These
3728 two options are now turned off when doing these lookups.
3730 57. It seems that at least some releases of Reiserfs (which does not have the
3731 concept of a fixed number of inodes) returns zero and not -1 for the
3732 number of available inodes. This interacted badly with check_spool_inodes,
3733 which assumed that -1 was the "no such thing" setting. What I have done is
3734 to check that the total number of inodes is greater than zero before doing
3735 the test of how many are available.
3737 58. When a "warn" ACL statement has a log_message modifier, the message is
3738 remembered, and not repeated. This is to avoid a lot of repetition when a
3739 message has many recipients that cause the same warning to be written.
3740 Howewer, Exim was preserving the list of already written lines for an
3741 entire SMTP session, which doesn't seem right. The memory is now reset if a
3742 new message is started.
3744 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3745 debugging output unless log_rewrite was also set.
3747 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3748 of (int)(handle) when we know that handle contains (void *)(-1).
3750 61. The Exim daemon panic-logs an error return when it closes the incoming
3751 connection. However "connection reset by peer" seems to be common, and
3752 isn't really an error worthy of noting specially, so that particular error
3755 62. When Exim is trying to find all the local interfaces, it used to panic and
3756 die if the ioctl to get the interface flags failed. However, it seems that
3757 on at least one OS (Solaris 9) it is possible to have an interface that is
3758 included in the list of interfaces, but for which you get a failure error
3759 for this call. This happens when the interface is not "plumbed" into a
3760 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3761 failure of the "get flags" call assumes that the interface is down.
3763 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3764 makes life easier for people who are doing arithmetic on fields extracted
3765 from dates, where you often get leading zeros that should not be
3766 interpreted as octal.
3768 64. Added qualify_domain to the redirect router, to override the global
3771 65. If a pathologically long header line contained very many addresses (the
3772 report of this problem mentioned 10 000) and each of them was rewritten,
3773 Exim could use up a very large amount of memory. (It kept on making new
3774 copies of the header line as it rewrote, and never released the old ones.)
3775 At the expense of a bit more processing, the header rewriting function has
3776 been changed so that it no longer eats memory in this way.
3778 66. The generation of the Received: header has been moved from the time that a
3779 message starts to be received, to the time that it finishes. The timestamp
3780 in the Received: header should now be very close to that of the <= log
3781 line. There are two side-effects of this change:
3783 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3784 logged header lines no longer include the local Received: line, because
3785 it has not yet been created. The same applies to a copy of the message
3786 that is returned to a non-SMTP sender when a message is rejected.
3788 (b) When a filter file is tested using -bf, no additional Received: header
3789 is added to the test message. After some thought, I decided that this
3792 This change does not affect the value of $received_for. It is still set
3793 after address rewriting, but before local_scan() is called.
3795 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3797 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3798 gave an unhelpful panic error message, and a defer error. I have managed to
3799 change this behaviour so that it now rejects any supplied certificate,
3800 which seems right, as the list of acceptable certificates is empty.
3802 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3803 gave an unhelpful defer error. I have not managed to make this reject any
3804 supplied certificates, but the error message it gives is "no certificate
3805 supplied", which is not helpful.
3807 70. exigrep's output now also includes lines that are not associated with any
3808 message, but which match the given pattern. Implemented by a patch from
3809 Martin Sluka, which also tidied up the Perl a bit.
3811 71. Recipient callout verification, like sender verification, was using <> in
3812 the MAIL FROM command. This isn't really the right thing, since the actual
3813 sender may affect whether the remote host accepts the recipient or not. I
3814 have changed it to use the actual sender in the callout; this means that
3815 the cache record is now keyed on a recipient/sender pair, not just the
3816 recipient address. There doesn't seem to be a real danger of callout loops,
3817 since a callout by the remote host to check the sender would use <>.
3818 [SEE ABOVE: changed after hitting problems.]
3820 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3821 temporary errors. However, in the case of such a code being given after
3822 the end of a data transmission (i.e. after ".") Exim was failing to write
3823 a retry record for the message. (Yes, there was some broken host that was
3824 actually sending 8xx at this point.)
3826 73. An unknown lookup type in a host list could cause Exim to panic-die when
3827 the list was checked. (An example that provoked this was putting <; in the
3828 middle of a list instead of at the start.) If this happened during a DATA
3829 ACL check, a -D file could be left lying around. This kind of configuration
3830 error no longer causes Exim to die; instead it causes a defer errror. The
3831 incident is still logged to the main and panic logs.
3833 74. Buglet left over from Exim 3 conversion. The message "too many messages
3834 in one connection" was written to the rejectlog but not the mainlog, except
3835 when address rewriting (yes!) was being logged.
3837 75. Added write_rejectlog option.
3839 76. When a system filter was run not as root (that is, when system_filter_user
3840 was set), the values of the $n variables were not being returned to the
3841 main process; thus, they were not subsequently available in the $sn
3844 77. Added +return_path_on_delivery log selector.
3846 78. A connection timeout was being treated differently from recipients deferred
3847 when testing hosts_max_try with a message that was older than the host's
3848 retry timeout. (The host should not be counted, thus allowing all hosts to
3849 be tried at least once before bouncing.) This may have been the cause of an
3850 occasionally reported bug whereby a message would remain on the queue
3851 longer than the retry timeout, but would be bounced if a delivery was
3852 forced. I say "may" because I never totally pinned down the problem;
3853 setting up timeout/retry tests is difficult. See also the next item.
3855 79. The ultimate address timeout was not being applied to errors that involved
3856 a combination of host plus message (for example, a timeout on a MAIL
3857 command). When an address resolved to a number of possible hosts, and they
3858 were not all tried for each delivery (e.g. because of hosts_max_try), a
3859 message could remain on the queue longer than the retry timeout.
3861 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3864 81. Fixed an obscure SMTP outgoing bug which required at least the following
3865 conditions: (a) there was another message waiting for the same server;
3866 (b) the server returned 5xx to all RCPT commands in the first message so
3867 that the message was not completed; (c) the server dropped the connection
3868 or gave a negative response to the RSET that Exim sends to abort the
3869 transaction. The observed case was a dropped connection after DATA that had
3870 been sent in pipelining mode. That is, the server had advertised PIPELINING
3871 but was not implementing it correctly. The effect of the bug was incorrect
3872 behaviour, such as trying another host, and this could lead to a crash.
3878 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3879 and daemon.c were passed as pointers to ints; they should have been
3880 pointers to socklen_t variables (which are typically unsigned ints).
3882 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3885 3. Fixed a really odd bug that affected only the testing scheme; patching a
3886 certain fixed string in the binary changed the value of another string that
3887 happened to be identical to the end of the original first string.
3889 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3890 name", it returns that address as the IP address. On some operating
3891 systems (e.g. Solaris), it also passes back the IP address string as the
3892 "host name". However, on others (e.g. Linux), it passes back an empty
3893 string. Exim wasn't checking for this, and was changing the host name to an
3894 empty string, assuming it had been canonicized.
3896 5. Although rare, it is permitted to have more than one PTR record for a given
3897 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3898 all the names associated with an address, because they do in Solaris.
3899 However, it seems that they do not in Linux for data that comes from the
3900 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3901 I found this out when I moved to a new Linux workstation and tried to run
3902 the Exim test suite.
3904 To get round this problem I have changed the code so that it now does its
3905 own call to the DNS to look up PTR records when searching for a host name.
3906 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3907 addresses that are only in /etc/hosts are still found.
3909 This behaviour is, however, controlled by an option called host_lookup_
3910 order, which defaults to "bydns:byaddr". If people want to use the other
3911 order, or indeed, just use one or the other means of lookup, they can
3912 specify it in this variable.
3914 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3915 some operating systems, this comes back from gethostbyaddr() as an empty
3916 string, and this is what Exim used to test for. However, it seems that in
3917 other systems, "." is yielded. Exim now tests for this case too.
3919 7. The values of check_spool_space and check_log_space are now held internally
3920 as a number of kilobytes instead of an absolute number of bytes. If a
3921 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3922 kilobyte. This means that much larger values can be stored.
3924 8. Exim monitor: an attempt to get the action menu when not actually pointing
3925 at a message produces an empty menu entitled "No message selected". This
3926 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3927 no entries in it ("Shell widget menu has zero width and/or height"). So I
3928 have added a single, blank menu entry in this case.
3930 9. Added ${quote_local_part.
3932 10. MIME decoding is now applied to the contents of Subject: header lines when
3935 11. Now that a reference to $sender_host_address automatically causes a reverse
3936 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3937 host lookup before query-style lookups in lists that might use this
3938 variable. This has therefore been abolished, and the "net-" prefix is no
3939 longer necessary for query-style lookups.
3941 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3942 have been a typo for LFLAGS, so it has been changed.
3944 13. The install script calls Exim with "-C /dev/null" in order to find the
3945 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3946 to be output. Howeve, since Exim outputs its version number before the
3947 error, it didn't break the script. It just looked ugly. I fixed this by
3948 always allowing "-C /dev/null" if the caller is root.
3950 14. Ignore overlarge ACL variable number when reading spool file - insurance
3951 against a later release with more variables having written the file.
3953 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3954 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3955 this, as well as the form without the "IPv6" on the front.
3957 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3958 OS/Makefile-Darwin file.
3960 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3961 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3963 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3964 variable. However, some OS use ints for inodes. I've added cast to long int
3965 to get rid of the compiler warning.
3967 19. I had forgotten to lock out "/../" in configuration file names when
3968 ALT_CONFIG_PREFIX was set.
3970 20. Routers used for verification do not need to specify transports. However,
3971 if such a router generated a host list, and callout was configured, Exim
3972 crashed, because it could not find a port number from the (non-existent)
3973 transport. It now assumes port 25 in this circumstance.
3975 21. Added the -t option to exigrep.
3977 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3978 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3979 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3980 them is set, LOOKUP_LSEARCH is forced.
3982 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3984 24. Added sender and host information to the "rejected by local_scan()" log
3985 line; previously there was no indication of these.
3987 25. Added .include_if_exists.
3989 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3990 Linux. It turns out that not all file systems support this. Apparently some
3991 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3992 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3993 not supported on the file descriptor, is now ignored when Exim is trying to
3994 sync a directory. This applies only to Linux.
3996 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3998 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3999 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4002 29. Added support for maildirsize files from supplied patch (modified a bit).
4004 30. The use of :fail: followed by an empty string could lead Exim to respond to
4005 sender verification failures with (e.g.):
4007 550 Verification failed for <xxx>
4008 550 Sender verify failed
4010 where the first response line was missing the '-' that indicates it is not
4011 the final line of the response.
4013 31. The loop for finding the name of the user that called Exim had a hardwired
4014 limit of 10; it now uses the value of finduser_retries, which is used for
4015 all other user lookups.
4017 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4020 33. Exim was neglecting to zero errno before one call of strtol() when
4021 expanding a string and expecting an integer value. On some systems this
4022 resulted in spurious "integer overflow" errors. Also, it was casting the
4023 result into an int without checking.
4025 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4026 did not work. The code looks as if it has *never* worked, though it appears
4027 to have been documented since at least releast 1.62. I have made it work.
4029 35. The "timeout_DNS" error in retry rules, also documented since at least
4030 1.62, also never worked. As it isn't clear exactly what this means, and
4031 clearly it isn't a major issue, I have abolished the feature by treating it
4032 as "timeout", and writing a warning to the main and panic logs.
4034 36. The display of retry rules for -brt wasn't always showing the error code
4037 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4038 timeout_connect_A, timeout_connect_MX.
4040 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4041 to the empty sender.
4043 39. The daemon was not analysing the content of -oX till after it had closed
4044 stderr and disconnected from the controlling terminal. This meant that any
4045 syntax errors were only noted on the panic log, and the return code from
4046 the command was 0. By re-arranging the code a little, I've made the
4047 decoding happen first, so such errors now appear on stderr, and the return
4048 code is 1. However, the actual setting up of the sockets still happens in
4049 the disconnected process, so errors there are still only recorded on the
4052 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4053 connections (as happens on some IP stacks) was logged at start up time as
4054 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4055 it from "IPv6 and IPv4", which means that two separate sockets are being
4058 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4059 says whether AF_INET or AF_INET6 was passed as an argument.
4061 42. Exiwhat output was messed up when time zones were included in log
4064 43. Exiwhat now gives more information about the daemon's listening ports,
4065 and whether -tls-on-connect was used.
4067 44. The "port" option of the smtp transport is now expanded.
4069 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4070 silently ignored. Now an error message is written to the main and panic
4073 46. There's a new ACL modifier called "logwrite" which writes to a log file
4074 as soon as it is encountered.
4076 47. Added $local_user_uid and $local_user_gid at routing time.
4078 48. Exim crashed when trying to verify a sender address that was being
4081 49. Exim was recognizing only a space character after ".include". It now also
4082 recognizes a tab character.
4084 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4085 extracting the relevant information from the specification. The man page no
4086 longer contains scrambled data for the -d option, and I've added a section
4087 at the front about calling Exim under different names.
4089 51. Added "extra_headers" argument to the "mail" command in filter files.
4091 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4094 53. Installed eximstats 1.29.
4096 54. Added transport_filter_timeout as a generic transport option.
4098 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4099 Cc: header lines. This was required by RFC 822, but it not required by RFC
4102 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4103 incoming messages that did not have them. Now it does so only if the
4104 message originates locally, that is, if there is no associated remote host
4105 address. When Resent- header lines are present, this applies to the Resent-
4106 lines rather than the non-Resent- lines.
4108 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4109 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4111 58. Messages for configuration errors now include the name of the main
4112 configuration file - useful now that there may be more than one file in a
4113 list (.included file names were always shown).
4115 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4116 for those rare installations that do not start the daemon as root or run it
4117 setuid root. I've cut out the call to initgroups() if the daemon is not
4120 60. The Exim user and group can now be bound into the binary as text strings
4121 that are looked up at the start of Exim's processing.
4123 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4125 62. Added $mailstore_basename variable.
4127 63. Installed patch to sieve.c from Michael Haardt.
4129 64. When Exim failed to open the panic log after failing to open the main log,
4130 the original message it was trying to log was written to stderr and debug
4131 output, but if they were not available (the usual case in production), it
4132 was lost. Now it is written to syslog before the two lines that record the
4133 failures to open the logs.
4135 65. Users' Exim filters run in subprocesses under the user's uid. It is
4136 possible for a "deliver" command or an alias in a "personal" command to
4137 provoke an address rewrite. If logging of address rewriting is configured,
4138 this fails because the process is not running as root or exim. There may be
4139 a better way of dealing with this, but for the moment (because 4.30 needs
4140 to be released), I have disabled address rewrite logging when running a
4141 filter in a non-root, non-exim process.
4147 1. The buildconfig auxiliary program wasn't quoting the value set for
4148 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4149 not defined. This bug was masked in 4.22 by the effect that was fixed in
4152 2. Some messages that were rejected after a message id was allocated were
4153 shown as "incomplete" by exigrep. It no longer does this for messages that
4154 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4156 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4157 have allow_domain_literals set, the ID did not get logged in the <= line.
4158 Domain literals are now always recognized in Message-ID: header lines.
4160 4. The first argument for a ${extract expansion item is the key name or field
4161 number. Leading and trailing spaces in this item were not being ignored,
4162 causing some misleading effects.
4164 5. When deliver_drop_privilege was set, single queue runner processes started
4165 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4166 same command in the process it spins off) were not dropping privilege.
4168 6. When the daemon running as "exim" started a queue runner, it always
4169 re-executed Exim in the spun-off process. This is a waste of effort when
4170 deliver_drop_privilege is set. The new process now just calls the
4171 queue-runner function directly.
4177 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4180 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4181 ignored. Though the use of -oP was forcing the writing of a pid file, it
4182 was always written to the default place.
4184 3. If the message "no IP address found for host xxxx" is generated during
4185 incoming verification, it is now followed by identification of the incoming
4186 connection (so you can more easily find what provoked it).
4188 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4190 5. Added some features to "harden" Exim a bit more against certain attacks:
4192 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4193 be put in Local/Makefile. This is like the never_users runtime option,
4194 but it cannot be overridden. The default setting is "root".
4196 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4197 prefix string with which any file named in a -C command line option
4200 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4201 is retained for -C and -D only if the caller of Exim is root. Without
4202 it, the exim user may also use -C and -D and retain privilege.
4204 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4205 command line option is disabled.
4207 6. Macro names set by the -D option must start with an upper case letter, just
4208 like macro names defined in the configuration file.
4210 7. Added "dereference=" facility to LDAP.
4212 8. Two instances of the typo "uknown" in the source files are fixed.
4214 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4215 the Configure-Makefile script screwed up while processing it.
4217 10. Incorporated PCRE 4.4.
4219 11. The SMTP synchronization check was not operating right at the start of an
4220 SMTP session. For example, it could not catch a HELO sent before the client
4221 waited for the greeting. There is now a check for outstanding input at the
4222 point when the greeting is written. Because of the duplex, asynchronous
4223 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4224 way, but not yet received, when the check is performed.
4226 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4227 on TCP/IP sockets, because this apparently causes some broken clients to
4230 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4231 unchanged) from the Cygwin maintainer.
4233 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4234 was supported instead of testing for mbx. Effectively a typo.
4236 15. The spa authenticator server code was not checking that the input it
4237 received was valid base64.
4239 16. The debug output line for the "set" modifier in ACLs was not showing the
4240 name of the variable that was being set.
4242 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4244 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4245 lookup on its own. Something else was needed to trigger the lookup. For
4246 example, a match in host_lookup or the need for a host name in a host list.
4247 Now, if $sender_host_name is referenced and the host name has not yet been
4248 looked up, a lookup is performed. If the lookup fails, the variable remains
4249 empty, and $host_lookup_failed is set to "1".
4251 19. Added "eqi" as a case-independent comparison operator.
4253 20. The saslauthd authentication condition could segfault if neither service
4254 nor realm was specified.
4256 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4257 error message that was logged was misleading, and incoming SMTP
4258 connections were dropped. The message is now more accurate, and temporary
4259 errors are given to SMTP connections.
4261 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4262 (except RSET) with a 421 error, until QUIT is received. However, it was
4263 failing to send a response to QUIT.
4265 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4266 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4267 ACL is now run after the helo_try_verify_hosts code.
4269 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4270 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4271 case-independent, so other case variants are also recognized). Apparently
4272 some systems use these upper case variants.
4274 25. If more than two messages were waiting for the same host, and a transport
4275 filter was specified for the transport, Exim sent two messages over the
4276 same TCP/IP connection, and then failed with "socket operation on non-
4277 socket" when it tried to send the third.
4279 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4281 27. The extern definition of crypt16() in expand.c was not being excluded when
4282 the OS had its own crypt16() function.
4284 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4285 as a preferred synonym for return_size_limit, both as an option and as an
4288 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4290 30. Changed the default configuration ACL to relax the local part checking rule
4291 for addresses that are not in any local domains. For these addresses,
4292 slashes and pipe symbols are allowed within local parts, but the sequence
4293 /../ is explicitly forbidden.
4295 31. SPA server authentication was not clearing the challenge buffer before
4298 32. log_message in a "warn" ACL statement was writing to the reject log as
4299 well as to the main log, which contradicts the documentation and doesn't
4300 seem right (because no rejection is happening). So I have stopped it.
4302 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4303 However, I am unable to do any testing of this.
4305 34. Fixed an infelicity in the appendfile transport. When checking directories
4306 for a mailbox, to see if any needed to be created, it was accidentally
4307 using path names with one or more superfluous leading slashes; tracing
4308 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4310 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4311 contents are added to the log line that is written for every discarded
4312 recipient. (Previously a log_message setting was ignored.)
4314 36. The ${quote: operator now quotes the string if it is empty.
4316 37. The install script runs exim in order to find its version number. If for
4317 some reason other than non-existence or emptiness, which it checks, it
4318 could not run './exim', it was installing it with an empty version number,
4319 i.e. as "exim-". This error state is now caught, and the installation is
4322 38. An argument was missing from the function that creates an error message
4323 when Exim fails to connect to the socket for saslauthd authentication.
4324 This could cause Exim to crash, or give a corrupted message.
4326 39. Added isip, isip4, and isip6 to ${if conditions.
4328 40. The ACL variables $acl_xx are now saved with the message, and can be
4329 accessed later in routers, transports, and filters.
4331 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4332 strings in the file are not string-expanded.
4334 42. If a MAIL command specified a SIZE value that was too large to fit into an
4335 int variable, the check against message_size_limit failed. Such values are
4336 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4337 day this will have to be increased, but I don't think I want to be around
4338 when emails are that large.
4345 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4346 iconv() is not standard in FreeBSD.
4348 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4349 IPv6 enabled. The observed symptom was a segmentation fault on return from
4350 the function os_common_find_running_interfaces() in src/os.c.
4352 3. In the check_special_case() function in daemon.c I had used "errno" as an
4353 argument name, which causes warnings on some systems. This was basically a
4354 typo, since it was named "eno" in the comments!
4356 4. The code that waits for the clock to tick (at a resolution of some fraction
4357 of a second) so as to ensure message-id uniqueness was always waiting for
4358 at least one whole tick, when it could have waited for less. [This is
4359 almost certainly not relevant at current processor speeds, where it is
4360 unlikely to ever wait at all. But we try to future-proof.]
4362 5. The function that sleeps for a time interval that includes fractions of a
4363 second contained a race. It did not block SIGALRM between setting the
4364 timer, and suspending (a couple of lines later). If the interval was short
4365 and the sigsuspend() was delayed until after it had expired, the suspension
4366 never ended. On busy systems this could lead to processes getting stuck for
4369 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4370 process, before it forks any delivery processes. The open lookup caching
4371 mechanism meant that the open file or database connection was passed into
4372 the delivery process. The problem was that delivery processes always tidy
4373 up cached lookup data. This could cause a problem for the next delivery
4374 process started by the queue runner, because the external queue runner
4375 process does not know about the closure. So the next delivery process
4376 still has data in the lookup cache. In the case of a file lookup, there was
4377 no problem because closing a file descriptor in a subprocess doesn't affect
4378 the parent. However, if the lookup was caching a connection to a database,
4379 the connection was closed, and the second delivery process was likely to
4380 see errors such as "PGSQL: query failed: server closed the connection
4381 unexpectedly". The problem has been fixed by closing all cached lookups
4382 in a queue runner before running a delivery process.
4384 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4385 seem to have the "const" qualifier which it has on other OS. I've
4388 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4389 *of the same type* (client or server) with the same public name that an
4390 error should be diagnosed.
4392 9. When Exim looked up a host name for an IP address, but failed to find the
4393 original IP address when looking up the host name (a safety check), it
4394 output the message "<ip address> does not match any IP for NULL", which was
4395 confusing, to say the least. The bug was that the host name should have
4396 appeared instead of "NULL".
4398 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4399 user that is built into the binary, and the -C or -D options is used, root
4400 privilege is dropped before the configuration file is read. In addition,
4401 logging is switched to stderr instead of the normal log files. If the
4402 configuration then re-defines the Exim user, the unprivileged environment
4403 is probably not what is expected, so Exim logs a panic warning message (but
4406 However, if deliver_drop_privilege is set, the unprivileged state may well
4407 be exactly what is intended, so the warning has been cut out in that case,
4408 and Exim is allowed to try to write to its normal log files.
4414 1. smtp_return_error_details was not giving details for temporary sender
4415 or receiver verification errors.
4417 2. Diagnose a configuration error if two authenticators have the same public
4420 3. Exim used not to create the message log file for a message until the first
4421 delivery attempt. This could be confusing when incoming messages were held
4422 for policy or load reasons. The message log file is now created at the time
4423 the message is received, and an initial "Received" line is written to it.
4425 4. The automatically generated man page for command line options had a minor
4426 bug that caused no ill effects; however, a more serious problem was that
4427 the procedure for building the man page automatically didn't always
4428 operate. Consequently, release 4.20 contains an out-of-date version. This
4429 shouldn't happen again.
4431 5. When building Exim with embedded Perl support, the script that builds the
4432 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4433 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4435 6. The freeze_tell option was not being used for messages that were frozen on
4436 arrival, either by an ACL or by local_scan().
4438 7. Added the smtp_incomplete_transaction log selector.
4440 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4441 was accepting AUTH without a new EHLO.
4443 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4444 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4447 10. Logging of TCP/IP connections (when configured) now happens in the main
4448 daemon process instead of the child process, so that the TCP/IP connection
4449 count is more accurate (but it can never be perfect).
4451 11. The use of "drop" in a nested ACL was not being handled correctly in the
4452 outer ACL. Now, if condition failure induced by the nested "drop" causes
4453 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4454 or "require"), the connection is dropped.
4456 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4457 that yield "discard" can now be used with an "accept" or a "discard" verb,
4458 but an error is generated for any others (because I can't see a useful way
4459 to define what should happen).
4461 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4462 are now processed in the same way as lines in the Exim configuration file.
4463 In particular, continuation lines are supported.
4465 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4467 15. Added -ti meaning -t -i.
4469 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4470 domains, and warn by logging if others are found.
4472 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4473 returned from GIFCONF when Exim is trying to find the list of interfaces on
4474 a host. The code in os.c has been modified to copy each ifreq to an aligned
4475 structure in all cases.
4477 Also, in some cases, the returned ifreq's were being copied to a 'struct
4478 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4479 means the last couple of bytes of an IPv6 address could be chopped if the
4480 ifreq contained only a normal sockaddr (14 bytes storage).
4482 18. Named domain lists were not supported in the hosts_treat_as_local option.
4483 An entry such as +xxxx was not recognized, and was treated as a literal
4486 19. Ensure that header lines added by a DATA ACL are included in the reject log
4487 if the ACL subsequently rejects the message.
4489 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4490 MD5 (which is deprecated).
4492 21. When testing a filter file using -bf, Exim was writing a message when it
4493 took the sender from a "From " line in the message, but it was not doing so
4494 when it took $return_path from a Return-Path: header line. It now does.
4496 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4497 with a valid header line field name (a series of printing characters
4498 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4500 23. Changed "disc" in the source to "disk" to conform to the documentation and
4501 the book and for uniformity.
4503 24. Ignore Sendmail's -Ooption=value command line item.
4505 25. When execve() failed while trying to run a command in a pipe transport,
4506 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4507 could be confused with a return value of 69 from the command itself. This
4508 has been changed to 127, the value the shell returns if it is asked to run
4509 a non-existent command. The wording for the related log line suggests a
4510 non-existent command as the problem.
4512 26. If received_header_text expands to an empty string, do not add a Received:
4513 header line to the message. (Well, it adds a token one on the spool, but
4514 marks it "old" so that it doesn't get used or transmitted.)
4516 27. Installed eximstats 1.28 (addition of -nt option).
4518 28. There was no check for failure on the call to getsockname() in the daemon
4519 code. This can fail if there is a shortage of resources on the system, with
4520 ENOMEM, for example. A temporary error is now given on failure.
4522 29. Contrary to the C standard, it seems that in some environments, the
4523 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4524 program. Exim now does this explicitly; it affects the formatting of
4525 timestamps using strftime().
4527 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4528 complaints. I've now initialized all the variables, to avoid this.
4530 32. Header lines added by a system filter were not being "seen" during
4531 transport-time rewrites.
4533 33. The info_callback() function passed to OpenSSL is set up with type void
4534 (*)(SSL *, int, int), as described somewhere. However, when calling the
4535 function (actually a macro) that sets it up, the type void(*)() is
4536 expected. I've put in a cast to prevent warnings from picky compilers.
4538 34. If a DNS black list lookup found a CNAME record, but there were no A
4539 records associated with the domain it pointed at, Exim crashed.
4541 35. If a DNS black list lookup returned more than one A record, Exim ignored
4542 all but the first. It now scans all returned addresses if a particular IP
4543 value is being sought. In this situation, the contents of the
4544 $dnslist_value variable are a list of all the addresses, separated by a
4547 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4548 to accept a host name and all its aliases if the forward lookup for any of
4549 them yielded the IP address of the incoming connection. Now it accepts only
4550 those names whose forward lookup yields the correct IP address. Any other
4551 names are discarded. This closes a loophole whereby a rogue DNS
4552 administrator could create reverse DNS records to break through a
4553 wildcarded host restriction in an ACL.
4555 37. If a user filter or a system filter that ran in a subprocess used any of
4556 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4557 the wrong values were passed to the pipe command ($thisaddress had the
4558 value of $0, $0 had the value of $1, etc). This bug was introduced by
4559 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4561 38. Improved the line breaking for long SMTP error messages from ACLs.
4562 Previously, if there was no break point between 40 and 75 characters, Exim
4563 left the rest of the message alone. Two changes have been made: (a) I've
4564 reduced the minimum length to 35 characters; (b) if it can't find a break
4565 point between 35 and 75 characters, it looks ahead and uses the first one
4566 that it finds. This may give the occasional overlong line, but at least the
4567 remaining text gets split now.
4569 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4570 file descriptors might be low, and that setting 1000 would always raise it.
4571 It turns out that in some environments, the limit is already over 1000 and
4572 that lowering it causes trouble. So now Exim takes care not to decrease it.
4574 40. When delivering a message, the value of $return_path is set to $sender_
4575 address at the start of routing (routers may change the value). By an
4576 oversight, this default was not being set up when an address was tested by
4577 -bt or -bv, which affected the outcome if any router or filter referred to
4580 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4581 log only when "message" or "log_message" are set. However, if one of the
4582 conditions was an address verification, or a call to a nested ACL, the
4583 messages generated by the underlying test were being passed through. This
4584 no longer happens. The underlying message is available in $acl_verify_
4585 message for both "message" and "log_message" expansions, so it can be
4586 passed through if needed.
4588 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4589 new expansion $bh_ to give the encoded byte string without charset
4590 translation. Translation happens only if iconv() is available; HAVE_ICONV
4591 indicates this at build time. HEADERS_CHARSET gives the charset to
4592 translate to; headers_charset can change it in the configuration, and
4593 "headers charset" can change it in an individual filter file.
4595 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4596 that creates RFC 2047 encoded "words" labels them as that charset instead
4597 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4598 expansion operator; (ii) when Exim creates a From: line for a local
4599 message; (iii) when a header line is rewritten to include a "phrase" part.
4601 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4602 buggy, causing it to skip the first lines of messages whose message ID
4603 ended in 'D'. This would not have bitten before Exim release 4.14, because
4604 message IDs were unlikely to end in 'D' before then. The effect was to have
4605 incorrect size information for certain domains.
4607 45. #include "config.h" was missing at the start of the crypt16.c module. This
4608 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4611 46. If there was a timeout during a "random" callout check, Exim treated it as
4612 a failure of the random address, and carried on sending RSET and the real
4613 address. If the delay was just some slowness somewhere, the response to the
4614 original RCPT would be taken as a response to RSET and so on, causing
4615 mayhem of various kinds.
4617 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4618 when I implemented it. It didn't allow for the fact that some option values
4619 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4620 the right test for positive values.
4622 48. Domain names in DNS records are case-independent. Exim always looks them up
4623 in lower case. Some resolvers return domain names in exactly the case they
4624 appear in the zone file, that is, they may contain uppercase letters. Not
4625 all resolvers do this - some return always lower case. Exim was treating a
4626 change of case by a resolver as a change of domain, similar to a widening
4627 of a domain abbreviation. This triggered its re-routing code and so it was
4628 trying to route what was effectively the same domain again. This normally
4629 caused routing to fail (because the router wouldn't handle the domain
4630 twice). Now Exim checks for this case specially, and just changes the
4631 casing of the domain that it ultimately uses when it transmits the message
4634 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4637 50. If a filter generated a file delivery with a non-absolute name (possible if
4638 no home directory exists for the router), the forbid_file option was not
4641 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4642 the existing equality matching.
4644 52. Exim was using ints instead of ino_t variables in some places where it was
4645 dealing with inode numbers.
4647 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4648 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4649 called TMPDIR, and if it finds it is different, it changes its value.
4651 54. The smtp_printf() function is now made available to local_scan() so
4652 additional output lines can be written before returning. There is also an
4653 smtp_fflush() function to enable the detection of a dropped connection.
4654 The variables smtp_input and smtp_batched_input are exported to
4657 55. Changed the default runtime configuration: the message "Unknown user"
4658 has been removed from the ACL, and instead placed on the localuser router,
4659 using the cannot_route_message feature. This means that any verification
4660 failures that generate their own messages won't get overridden. Similarly,
4661 the "Unrouteable address" message that was in the ACL for unverifiable
4662 relay addresses has also been removed.
4664 56. Added hosts_avoid_esmtp to the smtp transport.
4666 57. The exicyclog script was not checking for the esoteric option
4667 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4668 will work only if exicyclog is run under the appropriate euid.
4670 58. Following a discussion on the list, the rules by which Exim recognises line
4671 endings on incoming messages have been changed. The -dropcr and drop_cr
4672 options are now no-ops, retained only for backwards compatibility. The
4673 following line terminators are recognized: LF CRLF CR. However, special
4674 processing applies to CR:
4676 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4677 nor a local message in the state where . is a terminator.
4679 (ii) If a bare CR is encountered in a header line, an extra space is added
4680 after the line terminator so as not to end the header. The reasoning
4681 behind this is that bare CRs in header lines are most likely either
4682 to be mistakes, or people trying to play silly games.
4684 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4685 was being incorrectly given as 18 bytes larger than it should have been.
4686 This is a VOB (very old bug).
4688 60. This may never have affected anything current, but just in case it has:
4689 When the local host is found other than at the start of a list of hosts,
4690 the local host, those with the same MX, and any that follow, are discarded.
4691 When the list in question was part of a longer list of hosts, the following
4692 hosts (not currently being processed) were also being discarded. This no
4693 longer happens. I'm not sure if this situation could ever has previously
4696 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4699 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4700 header. This is something that is recommended in a new Internet Draft, and
4701 is something that is documented as being done by Sendmail. There are two
4702 possible values. For messages generated by the autoreply transport, Exim
4705 Auto-Submitted: auto-replied
4707 whereas for all other generated messages (e.g. bounces) it adds
4709 Auto-Submitted: auto-generated
4711 63. The "personal" condition in filters now includes a test for the
4712 Auto-Submitted: header. If it contains the string "auto-" the message it
4713 not considered personal.
4715 64. Added rcpt_include_affixes as a generic transport option.
4717 65. Added queue_only_override (default true).
4719 66. Added the syslog_duplication option.
4721 67. If what should have been the first header line of a message consisted of
4722 a space followed by a colon, Exim was mis-interpreting it as a header line.
4723 It isn't of course - it is syntactically invalid and should therefore be
4724 treated as the start of the message body. The misbehaviour could have
4725 caused a number of strange effects, including loss of data in subsequent
4726 header lines, and spool format errors.
4728 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4729 client host had authenticated. This control can now be exercised by an ACL
4730 for more flexibility.
4732 69. By default, callouts do not happen when testing with -bh. There is now a
4733 variant, -bhc, which does actually run the callout code, including
4734 consulting and updating the callout cache.
4736 70. Added support for saslauthd authentication, courtesy of Alexander
4739 71. If statvfs() failed on the spool or log directories while checking their
4740 size for availability, Exim confusingly gave the error "space shortage".
4741 Furthermore, in debugging mode it crashed with a floating point exception.
4742 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4743 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4744 serious problem, Exim now writes to the main and panic logs when this
4745 happens, with details of the failure. It then refuses to accept the
4746 incoming message, giving the message "spool directory problem" or "log
4747 directory problem" with a 421 code for SMTP messages.
4749 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4750 0, 1, and 2 exist, because some OS complain for execs without them (see
4751 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4752 descriptors. However, the code omitted to check that the open succeeded,
4753 causing mysterious errors if for some reason the permissions on /dev/null
4754 got screwed. Now Exim writes a message to the main and panic logs, and
4755 bombs out if it can't open /dev/null.
4757 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4758 interact so that it is all more flexible. It is supposed to remain
4759 backwards compatible. Also added extra_local_interfaces.
4761 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4762 to bomb out with an assertion failure - to the client this appears as a
4763 connection drop. This problem occurs in the part of the code that was taken
4764 from the Samba project. Fortunately, the assertion is in a very simple
4765 function, so I have fixed this by reproducing the function inline in the
4766 one place where it is called, and arranging for authentication to fail
4767 instead of killing the process with assert().
4769 75. The SPA client code was not working when the server requested OEM rather
4770 than Unicode encoding.
4772 76. Added code to make require_files with a specific uid setting more usable in
4773 the case where statting the file as root fails - usually a non-root-mounted
4774 NFS file system. When this happens and the failure is EACCES, Exim now
4775 forks a subprocess and does the per-uid checking as the relevant uid.
4777 77. Added process_log_path.
4779 78. If log_file_path was not explicitly set, a setting of check_log_space or
4780 check_log_inodes was ignored.
4782 79. If a space check for the spool or log partitions fails, the incident is now
4783 logged. Of course, in the latter case the data may get lost...
4785 80. Added the %p formatting code to string_format() so that it can be used to
4786 print addresses in debug_print(). Adjusted all the address printing in the
4787 debugging in store.c to use %p rather than %d.
4789 81. There was a concern that a line of code in smtp_in.c could overflow a
4790 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4791 initially expressed, the concern was not well-founded, because trailing
4792 spaces are removed early. However, if the trailing spaces were followed by
4793 a NULL, they did not get removed, so the overflow was possible. Two fixes
4796 (a) I re-wrote the offending code in a cleaner fashion.
4797 (b) If an incoming SMTP command contains a NULL character, it is rejected
4800 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4801 runs initgroups(), so that if the Exim user is in any additional groups,
4802 they will be used during message reception.
4808 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob