1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.490 2007/03/13 10:05:16 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
10 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
13 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
16 PH/02 In an ACL statement such as
18 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
20 if a client was not listed at all, or was listed with a value other than
21 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
22 the condition was not true (as it should be), so access was not denied.
23 The bug was that the ! inversion was incorrectly passed on to the second
24 item. This has been fixed.
26 PH/03 Added additional dnslists conditions == and =& which are different from
27 = and & when the dns lookup returns more than one IP address.
29 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
30 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
32 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
33 FSYNC, which compiles an option called disable_fsync that allows for
34 bypassing fsync(). The documentation is heavily laced with warnings.
36 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
38 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
39 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
40 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
41 including adding "make clean"; (3) Added -fPIC when compiling the test
42 dynamically loaded module, to get rid of a warning.
44 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
45 message fails, move_frozen_messages = true and ignore_bounce_errors_after
46 = 0s. The bug is otherwise harmless.
48 PH/07 There was a bug in the dovecot authenticator such that the value of
49 $auth1 could be overwritten, and so not correctly preserved, after a
50 successful authentication. This usually meant that the value preserved by
51 the server_setid option was incorrect.
53 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
55 PH/09 Installed PCRE release 7.0.
57 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
58 run for batched SMTP input. It is now run at the start of every message
59 in the batch. While fixing this I discovered that the process information
60 (output by running exiwhat) was not always getting set for -bs and -bS
61 input. This is fixed, and it now also says "batched" for BSMTP.
63 PH/11 Added control=no_pipelining.
65 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
66 patch, slightly modified), and move the expansion of helo_data till after
67 the connection is made in the smtp transport (so it can use these
70 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
72 PH/14 Added log_selector = +pid.
74 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
76 PH/16 Add ${if forany and ${if forall.
78 PH/17 Added dsn_from option to vary the From: line in DSNs.
80 PH/18 Flush SMTP output before performing a callout, unless control =
81 no_callout_flush is set.
83 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
84 was true (the default) a successful delivery failed to delete the retry
85 item, thus causing premature timeout of the address. The bug is now
88 PH/20 Added hosts_avoid_pipelining to the smtp transport.
90 PH/21 Long custom messages for fakedefer and fakereject are now split up
91 into multiline reponses in the same way that messages for "deny" and
92 other ACL rejections are.
94 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
95 with slight modification.
97 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
98 draft, changing the syntax and factoring some duplicate code".
100 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
101 for deliveries of the second and subsequent messages over the same SMTP
104 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
105 ${reduce, with only minor "tidies".
107 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
109 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
110 expansion side effects.
112 PH/27 When a message times out after an over-quota error from an Exim-imposed
113 quota, the bounce message says "mailbox is full". This message was not
114 being given when it was a system quota that was exceeded. It now should
117 MH/03 Made $recipients available in local_scan(). local_scan() already has
118 better access to the recipient list through recipients_list[], but
119 $recipients can be useful in postmaster-provided expansion strings.
121 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
122 in the case of a MAIL command with additional options following the
123 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
124 were accidentally chopped off.
126 PH/29 SMTP synchronization checks are implemented when a command is read -
127 there is a check that no more input is waiting when there shouldn't be
128 any. However, for some commands, a delay in an ACL can mean that it is
129 some time before the response is written. In this time, more input might
130 arrive, invalidly. So now there are extra checks after an ACL has run for
131 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
132 pipelining has not been advertised.
134 PH/30 MH's patch to allow iscntrl() characters to be list separators.
136 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
137 returned in the SMTP response when smtp_return_error_details was false.
140 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
141 instead of the C I/O that was originally supplied, because problems were
144 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
145 Exim which did not show up earlier: it was assuming that a call to
146 SSL_CTX_set_info_callback() might give an error value. In fact, there is
147 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
148 was a macro that became an assignment, so it seemed to work. This has
149 changed to a proper function call with a void return, hence the compile
150 error. Exim's code has been fixed.
152 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
155 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
161 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
162 fixed by 4.65/MH/01 (is this a record?) are fixed:
164 (i) An empty string was always treated as zero by the numeric comparison
165 operators. This behaviour has been restored.
167 (ii) It is documented that the numeric comparison operators always treat
168 their arguments as decimal numbers. This was broken in that numbers
169 starting with 0 were being interpreted as octal.
171 While fixing these problems I realized that there was another issue that
172 hadn't been noticed. Values of message_size_limit (both the global option
173 and the transport option) were treated as octal if they started with 0.
174 The documentation was vague. These values are now always treated as
175 decimal, and I will make that clear in the documentation.
181 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
182 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
185 MH/01 Don't check that the operands of numeric comparison operators are
186 integers when their expansion is in "skipping" mode (fixes bug
187 introduced by 4.64-PH/07).
189 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
190 child addresses, Exim now panics and dies. Previously, because the count
191 is held in a short int, deliveries were likely to be lost. As such a
192 large number of recipients for a single message is ridiculous
193 (performance will be very, very poor), I have chosen to impose a limit
194 rather than extend the field.
200 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
201 leftover -K file (the existence of which was triggered by #402).
202 While we were at it, introduced process PID as part of the -K
203 filename. This should rule out race conditions when creating
206 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
207 processing considerably. Previous code took too long for large mails,
208 triggering a timeout which in turn triggers #401.
210 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
211 in the DK code in transports.c. sendfile() is not really portable,
212 hence the _LINUX specificness.
214 TF/01 In the add_headers option to the mail command in an Exim filter,
215 there was a bug that Exim would claim a syntax error in any
216 header after the first one which had an odd number of characters
219 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
220 callout verification, Exim cached a "reject" for the entire domain. This
221 is correct for most verifications, but it is not correct for a recipient
222 verification with use_sender or use_postmaster set, because in that case
223 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
224 case of MAIL FROM:<> rejection from other early rejections (e.g.
225 rejection of HELO). When verifying a recipient using a non-null MAIL
226 address, the cache is ignored if it shows MAIL FROM:<> rejection.
227 Whatever the result of the callout, the value of the domain cache is
228 left unchanged (for any other kind of callout, getting as far as trying
229 RCPT means that the domain itself is ok).
231 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
234 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
235 manifest itself as EPIPE rather than ECONNECT. When tidying away a
236 session, the daemon ignores ECONNECT errors and logs others; it now
237 ignores EPIPE as well.
239 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
240 (quoted-printable decoding).
242 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
243 later the small subsequent patch to fix an introduced bug.
245 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
247 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
249 PH/08 An error is now given if message_size_limit is specified negative.
251 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
252 to be given (somewhat) arbitrary names.
254 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
257 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
258 miscellaneous code fixes
260 PH/10 Added the log_reject_target ACL modifier to specify where to log
263 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
264 hostname. This is wrong, because it relates to the incoming message (and
265 probably the interface on which it is arriving) and not to the outgoing
266 callout (which could be using a different interface). This has been
267 changed to use the value of the helo_data option from the smtp transport
268 instead - this is what is used when a message is actually being sent. If
269 there is no remote transport (possible with a router that sets up host
270 addresses), $smtp_active_hostname is used.
272 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
273 tweaks were necessary in order to get it to work (see also 21 below):
274 (a) The code assumed that strncpy() returns a negative number on buffer
275 overflow, which isn't the case. Replaced with Exim's string_format()
277 (b) There were several signed/unsigned issues. I just did the minimum
278 hacking in of casts. There is scope for a larger refactoring.
279 (c) The code used strcasecmp() which is not a standard C function.
280 Replaced with Exim's strcmpic() function.
281 (d) The code set only $1; it now sets $auth1 as well.
282 (e) A simple test gave the error "authentication client didn't specify
283 service in request". It would seem that Dovecot has changed its
284 interface. Fortunately there's a specification; I followed it and
285 changed what the client sends and it appears to be working now.
287 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
290 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
291 address A is aliased to B and C, where B exists and C does not. Without
292 -v the output is "A verified" because verification stops after a
293 successful redirection if more than one address is generated. However,
294 with -v the child addresses are also verified. Exim was outputting "A
295 failed to verify" and then showing the successful verification for C,
296 with its parentage. It now outputs "B failed to verify", showing B's
297 parentage before showing the successful verification of C.
299 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
300 look up a TXT record in a specific list after matching in a combined
303 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
304 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
305 they consult the DNS. I had assumed they would set it the way they
306 wanted; and indeed my experiments on Linux seem to show that in some
307 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
308 To be on the safe side, however, I have now made the interface to
309 host_find_byname() similar to host_find_bydns(), with an argument
310 containing the DNS resolver options. The host_find_byname() function now
311 sets these options at its start, just as host_find_bydns() does. The smtp
312 transport options dns_qualify_single and dns_search_parents are passed to
313 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
314 of host_find_byname() use the default settings of RES_DEFNAMES
315 (qualify_single) but not RES_DNSRCH (search_parents).
317 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
318 spool_read_header() do less string testing, by means of a preliminary
319 switch on the second character of optional "-foo" lines. (This is
320 overdue, caused by the large number of possibilities that now exist.
321 Originally there were few.) While I was there, I also converted the
322 str(n)cmp tests so they don't re-test the leading "-" and the first
323 character, in the hope this might squeeze out yet more improvement.
325 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
326 flag allowing group syntax was set by the header_syntax check but not
327 turned off, possible causing trouble later; (2) The flag was not being
328 set at all for the header_verify test, causing "group"-style headers to
329 be rejected. I have now set it in this case, and also caused header_
330 verify to ignore an empty address taken from a group. While doing this, I
331 came across some other cases where the code for allowing group syntax
332 while scanning a header line wasn't quite right (mostly, not resetting
333 the flag correctly in the right place). These bugs could have caused
334 trouble for malformed header lines. I hope it is now all correct.
336 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
337 with the "reply" argument non-NULL. The code, however (which originally
338 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
339 but it didn't always do it. This confused somebody who was copying the
340 code for some other use. I have removed all the tests.
342 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
343 feature that was used to support insecure browsers during the U.S. crypto
344 embargo. It requires special client support, and Exim is probably the
345 only MTA that supported it -- and would never use it because real RSA is
346 always available. This code has been removed, because it had the bad
347 effect of slowing Exim down by computing (never used) parameters for the
348 RSA_EXPORT functionality.
350 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
351 authenticator to fail if there's a tab character in the incoming data
352 (there should never be unless someone is messing about, as it's supposed
353 to be base64-encoded). Also added, on Timo's advice, the "secured" option
354 if the connection is using TLS or if the remote IP is the same as the
355 local IP, and the "valid-client-cert option" if a client certificate has
358 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
359 authenticators. This can be used for authorization after authentication
360 succeeds. (In the case of plaintext, it servers for both authentication
363 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
364 if any retry times were supplied.
366 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
367 connection, where there is no client IP address to check. In this
368 situation, the verify now always succeeds.
370 PH/25 Applied John Jetmore's -Mset patch.
372 PH/26 Added -bem to be like -Mset, but loading a message from a file.
374 PH/27 In a string expansion for a processed (not raw) header when multiple
375 headers of the same name were present, leading whitespace was being
376 removed from all of them, but trailing whitespace was being removed only
377 from the last one. Now trailing whitespace is removed from each header
378 before concatenation. Completely empty headers in a concatenation (as
381 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
382 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
384 PH/29 [Removed. This was a change that I later backed out, and forgot to
385 correct the ChangeLog entry (that I had efficiently created) before
386 committing the later change.]
388 PH/30 Exim was sometimes attempting to deliver messages that had suffered
389 address errors (4xx response to RCPT) over the same connection as other
390 messages routed to the same hosts. Such deliveries are always "forced",
391 so retry times are not inspected. This resulted in far too many retries
392 for the affected addresses. The effect occurred only when there were more
393 hosts than the hosts_max_try setting in the smtp transport when it had
394 the 4xx errors. Those hosts that it had tried were not added to the list
395 of hosts for which the message was waiting, so if all were tried, there
396 was no problem. Two fixes have been applied:
398 (i) If there are any address or message errors in an SMTP delivery, none
399 of the hosts (tried or untried) are now added to the list of hosts
400 for which the message is waiting, so the message should not be a
401 candidate for sending over the same connection that was used for a
402 successful delivery of some other message. This seems entirely
403 reasonable: after all the message is NOT "waiting for some host".
404 This is so "obvious" that I'm not sure why it wasn't done
405 previously. Hope I haven't missed anything, but it can't do any
406 harm, as the worst effect is to miss an optimization.
408 (ii) If, despite (i), such a delivery is accidentally attempted, the
409 routing retry time is respected, so at least it doesn't keep
410 hammering the server.
412 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
413 in ${readsocket because some servers need this prod.
415 PH/32 Added some extra debug output when updating a wait-xxx database.
417 PH/33 The hint "could be header name not terminated by colon", which has been
418 given for certain expansion errors for a long time, was not being given
419 for the ${if def:h_colon_omitted{... case.
421 PH/34 The spec says: "With one important exception, whenever a domain list is
422 being scanned, $domain contains the subject domain." There was at least
423 one case where this was not true.
425 PH/35 The error "getsockname() failed: connection reset by peer" was being
426 written to the panic log as well as the main log, but it isn't really
427 panic-worthy as it just means the connection died rather early on. I have
428 removed the panic log writing for the ECONNRESET error when getsockname()
431 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
432 runs only) independently of the message's sender address. This meant
433 that, if the 4xx error was in fact related to the sender, a different
434 message to the same recipient with a different sender could confuse
435 things. In particualar, this can happen when sending to a greylisting
436 server, but other circumstances could also provoke similar problems.
437 I have changed the default so that the retry time for these errors is now
438 based a combination of the sender and recipient addresses. This change
439 can be overridden by setting address_retry_include_sender=false in the
442 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
443 remote server are returned as part of bounce messages. This was not
444 happening for LMTP over a pipe (the lmtp transport), but now it is the
445 same for both kinds of LMTP.
447 PH/38 Despite being documented as not happening, Exim was rewriting addresses
448 in header lines that were in fact CNAMEs. This is no longer the case.
450 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
451 and queue runs started by the daemon processed all messages. This has
452 been fixed so that -R and -S can now usefully be given with -q<time>.
454 PH/40 Import PCRE release 6.7 (fixes some bugs).
456 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
458 PH/42 Give an error if -q is specified more than once.
460 PH/43 Renamed the variables $interface_address and $interface_port as
461 $received_ip_address and $received_port, to make it clear that these
462 values apply to message reception, and not to the outgoing interface when
463 a message is delivered. (The old names remain recognized, of course.)
465 PH/44 There was no timeout on the connect() call when using a Unix domain
466 socket in the ${readsocket expansion. There now is.
468 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
469 be meaningful with "accept".
471 SC/01 Eximstats V1.43
472 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
474 SC/02 Eximstats V1.44
475 Use a glob alias rather than an array ref in the generated
476 parser. This improves both readability and performance.
478 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
479 Collect SpamAssassin and rejection statistics.
480 Don't display local sender or destination tables unless
481 there is data to show.
482 Added average volumes into the top table text output.
484 SC/04 Eximstats V1.46
485 Collect data on the number of addresses (recipients)
486 as well as the number of messages.
488 SC/05 Eximstats V1.47
489 Added 'Message too big' to the list of mail rejection
490 reasons (thanks to Marco Gaiarin).
492 SC/06 Eximstats V1.48
493 Mainlog lines which have GMT offsets and are too short to
494 have a flag are now skipped.
496 SC/07 Eximstats V1.49 (Alain Williams)
497 Added the -emptyok flag.
499 SC/08 Eximstats V1.50
500 Fixes for obtaining the IP address from reject messages.
502 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
503 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
504 whitesspace changes from 4.64-PH/27
506 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
509 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
510 are found, allow negative numbers in numeric criteria)
512 JJ/06 exipick.20061117.2, added new $message_body_missing variable
514 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
515 to match changes made in 4.64-PH/43
517 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
519 PH/47 Put in an explicit test for a DNS lookup of an address record where the
520 "domain" is actually an IP address, and force a failure. This locks out
521 those revolvers/nameservers that support "A-for-A" lookups, in
522 contravention of the specifications.
524 PH/48 When a host name was looked up from an IP address, and the subsequent
525 forward lookup of the name timed out, the host name was left in
526 $sender_host_name, contrary to the specification.
528 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
529 restricted to single-key lookups, Exim was not diagnosing an error if
530 * or *@ was used with a query-style lookup.
532 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
534 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
535 long ago, but noone interested enough thought of it. Let's just say that
536 the "1.1" means that there are some new functions that weren't there at
537 some point in the past.
539 PH/51 Error processing for expansion failure of helo_data from an smtp
540 transport during callout processing was broken.
542 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
543 tested/used via the -bh/-bhc/-bs options.
545 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
546 bug, fixed in subsequent PCRE releases).
548 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
549 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
551 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
557 SC/01 Use a glob alias rather than an array ref in eximstats generated
558 parser. This improves both readability and performance.
560 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
561 Don't display local sender or destination tables in eximstats unless
562 there is data to show.
563 Added average volumes into the eximstats top table text output.
565 SC/03 Collect data on the number of addresses (recipients) as well
566 as the number of messages in eximstats.
568 TF/01 Correct an error in the documentation for the redirect router. Exim
569 does (usually) call initgroups() when daemonizing.
571 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
572 with consistent privilege compared to when running as a daemon.
574 TF/03 Note in the spec that $authenticated_id is not set for local
575 submissions from trusted users.
577 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
578 Thanks to Dean Brooks <dean@iglou.com> for the patch.
580 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
581 by adding some example configuration directives to the default
582 configuration file. A little bit of work is required to uncomment the
583 directives and define how usernames and passwords are checked, but
584 there is now a framework to start from.
586 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
587 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
588 without this. I don't know how relevant this is to other LDAP libraries.
590 PH/02 Add the verb name to the "unknown ACL verb" error.
592 PH/03 Magnus Holmgren's patch for filter_prepend_home.
594 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
596 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
597 directory not expanded when it should be if an expanded home directory
598 was set for the address (which is overridden by the transport).
600 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
603 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
604 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
605 because it is too late at that time, and has no effect.
607 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
608 security issue with \' (bugzilla #107). I could not use the
609 PQescapeStringConn() function, because it needs a PGconn value as one of
612 PH/08 When testing addresses using -bt, indicate those final addresses that
613 are duplicates that would not cause an additional delivery. At least one
614 person was confused, thinking that -bt output corresponded to deliveries.
615 (Suppressing duplicates isn't a good idea as you lose the information
616 about possibly different redirections that led to the duplicates.)
618 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
619 systems where poll() doesn't work, in particular OS X.
621 PH/10 Added more information to debugging output for retry time not reached.
623 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
624 operations in malware.c.
626 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
629 PH/13 If write_rejectlog was set false when logging was sent to syslog with
630 syslog_duplication set false, log lines that would normally be written
631 both the the main log and to the reject log were not written to syslog at
634 PH/14 In the default configuration, change the use of "message" in ACL warn
635 statements to "add_header".
637 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
638 not followed by a command (e.g. "seen endif").
640 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
641 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
644 PH/17 Added extra conditions to the default value of delay_warning_condition
648 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
649 { match{$h_precedence:}{(?i)bulk|list|junk} } \
650 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
653 The Auto-Submitted: and various List- headers are standardised, whereas I
654 don't think Precedence: ever was.
656 PH/18 Refactored debugging code in route_finduser() to show more information,
657 in particular, the error code if getpwnam() issues one.
659 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
660 This is apparently needed in addition to the PH/07 change above to avoid
661 any possible encoding problems.
663 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
664 but not after initializing Perl.
666 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
667 output them only if debugging. By default they are written stderr,
668 apparently, which is not desirable.
670 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
673 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
676 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
678 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
679 authenticated or an ident call has been made. Suppress the default
680 values for $authenticated_id and $authenticated_sender (but permit -oMai
681 and -oMas) when testing with -bh.
683 PH/24 Re-jigged the order of the tests in the default configuration so that the
684 tests for valid domains and recipients precede the DNS black list and CSA
685 tests, on the grounds that those ones are more expensive.
687 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
688 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
689 This bug exists in every version of Exim that I still have, right back to
692 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
693 However, an attempt to turn on case-sensitivity in a regex key by
694 including (?-i) didn't work because the subject string was already
695 lowercased, and the effects were non-intuitive. It turns out that a
696 one-line patch can be used to allow (?-i) to work as expected.
702 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
703 other effects) broke the use of negated acl sub-conditions.
705 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
708 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
709 "Deny" causes Exim to reject the incoming connection with a 554 error.
710 Unfortunately, if there is a major crisis, such as a disk failure,
711 tcp-wrappers gives "deny", whereas what one would like would be some
712 kind of temporary error. A kludge has been added to help with this.
713 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
714 554 error is used if errno is still zero or contains ENOENT (which occurs
715 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
718 PH/03 Add -lutil to the default FreeBSD LIBS setting.
720 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
721 errors. Otherwise a message that provokes a temporary error (when other
722 messages do not) can cause a whole host to time out.
724 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
725 addresses were routed directly to files or pipes from a redirect router.
726 File deliveries just didn't batch; pipe deliveries might have suffered
729 PH/06 A failure to get a lock for a hints database would erroneously always say
730 "Failed to get write lock", even when it was really a read lock.
732 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
733 of 0600. This has been changed to use the value of the lockfile_mode
734 option (which defaults to 0600).
736 PH/08 Applied small patch from the Sieve maintainer.
738 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
739 folder from quota calculations, a direct delivery into this folder messed
740 up the contents of the maildirsize file. This was because the regex was
741 used only to exclude .Trash (or whatever) when the size of the mailbox
742 was calculated. There was no check that a delivery was happening into an
743 excluded directory. This bug has been fixed by ignoring all quota
744 processing for deliveries into excluded directories.
746 PH/10 Added the maildirfolder_create_regex option to appendfile.
752 PH/01 The code for finding all the local interface addresses on a FreeBSD
753 system running IPv6 was broken. This may well have applied to all BSD
754 systems, as well as to others that have similar system calls. The broken
755 code found IPv4 interfaces correctly, but gave incorrect values for the
756 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
757 that it would not match correctly against @[] and not recognize the IPv6
760 PH/02 The ipliteral router was not recognizing addresses of the form user@
761 [ipv6:....] because it didn't know about the "ipv6:" prefix.
763 PH/03 Added disable_ipv6.
765 PH/04 Changed $reply_address to use the raw form of the headers instead of the
766 decoded form, because it is most often used to construct To: headers
767 lines in autoreplies, and the decoded form may well be syntactically
768 invalid. However, $reply_address has leading white space removed, and all
769 newlines turned into spaces so that the autoreply transport does not
772 PH/05 If group was specified without a user on a router, and no group or user
773 was specified on a transport, the group from the router was ignored.
775 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
776 for visible compile-time settings that can be used to change these
777 numbers, for those that want even more. Backwards compatibility with old
778 spool files has been maintained. However, going back to a previous Exim
779 release will lost any variables that are in spool files.
781 PH/07 Two small changes when running in the test harness: increase delay when
782 passing a TCP/IP connection to a new process, in case the original
783 process has to generate a bounce, and remove special handling of
784 127.0.0.2 (sic), which is no longer necessary.
786 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
787 be the same on different OS.
789 PH/09 Moved a debug statement in filter processing to avoid a race problem when
792 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
793 whether --show-vars was specified or not
795 JJ/02 exipick: Added support for new ACL variable spool format introduced
798 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
799 syntactically invalid From: or Reply-to: line, and a filter used this to
800 generate an autoreply, and therefore failed to obtain an address for the
801 autoreply, Exim could try to deliver to a non-existent relative file
802 name, causing unrelated and misleading errors. What now happens is that
803 it logs this as a hard delivery error, but does not attempt to create a
806 PH/11 The exinext utility has a -C option for testing purposes, but although
807 the given file was scanned by exinext itself; it wasn't being passed on
808 when Exim was called.
810 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
811 an end-of-file indication when reading a command response.
813 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
814 compiled. In many other places in Exim, IPv6 addresses are always
815 recognized, so I have changed this. It also means that IPv4 domain
816 literals of the form [IPV4:n.n.n.n] are now always recognized.
818 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
819 used if the router is not running as root, for example, when verifying at
820 ACL time, or when using -bh. The debugging output from this situation was
821 non-existent - all you got was a failure to exec. I have made two
824 (a) Failures to set uid/gid, the current directory, or a process leader
825 in a subprocess such as that created by queryprogram now generate
826 suitable debugging ouput when -d is set.
828 (b) The queryprogram router detects when it is not running as root,
829 outputs suitable debugging information if -d is set, and then runs
830 the subprocess without attempting to change uid/gid.
832 PH/15 Minor change to Makefile for building test_host (undocumented testing
835 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
836 additional section of a DNS packet that returns MX or SRV records.
837 Instead, it always explicitly searches for A/AAAA records. This avoids
838 major problems that occur when a DNS server includes only records of one
839 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
840 fixed another bug: if SRV records were looked up and the corresponding
841 address records were *not* found in the additional section, the port
842 values from the SRV records were lost.
844 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
845 using the correct key (the original address) when searching the retry
846 rules in order to find which one to use for generating the retry hint.
848 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
849 adding the default one. Similarly, if it contains a Reply-To: header, the
850 errors_reply_to option, if set, is not used.
852 PH/19 When calculating a retry time, Exim used to measure the "time since
853 failure" by looking at the "first failed" field in the retry record. Now
854 it does not use this if it is later than than the arrival time of the
855 message. Instead it uses the arrival time. This makes for better
856 behaviour in cases where some deliveries succeed, thus re-setting the
857 "first failed" field. An example is a quota failure for a huge message
858 when small messages continue to be delivered. Without this change, the
859 "time since failure" will always be short, possible causing more frequent
860 delivery attempts for the huge message than are intended.
861 [Note: This change was subsequently modified - see PH/04 for 4.62.]
863 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
864 $1, $2, $3) because the numerical variables can be reset during some
865 expansion items (e.g. "match"), thereby losing the authentication data.
867 PH/21 Make -bV show the size of off_t variables so that the test suite can
868 decide whether to run tests for quotas > 2G.
870 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
871 mailbox_size, and mailbox_filecount in the appendfile transport. If a
872 filecount value is greater than 2G or if a quota value is greater than 2G
873 on a system where the size of off_t is not greater than 4, a panic error
876 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
877 never match. The debug and -bh output now contains an explicit error
878 message indicating a malformed IPv4 address or mask.
880 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
881 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
884 PH/25 Do not write to syslog when running in the test harness. The only
885 occasion when this arises is a failure to open the main or panic logs
886 (for which there is an explicit test).
888 PH/26 Added the /no_tell option to "control=freeze".
890 PH/27 If a host name lookup failed very early in a connection, for example, if
891 the IP address matched host_lookup and the reverse lookup yielded a name
892 that did not have a forward lookup, an error message of the form "no IP
893 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
894 could be logged. Now it outputs the IP address instead of "NULL".
896 PH/28 An enabling patch from MH: add new function child_open_exim2() which
897 allows the sender and the authenticated sender to be set when
898 submitting a message from within Exim. Since child_open_exim() is
899 documented for local_scan(), the new function should be too.
901 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
902 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
903 results in an empty string is now treated as unset.
905 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
907 PH/31 Added sender_verify_fail logging option.
909 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
910 needed by RFC 822 but not by RFC 2822 was commented out. I have now
911 tidied the source and removed it altogether.
913 PH/33 When a queue run was abandoned because the load average was too high, a
914 log line was always written; now it is written only if the queue_run log
915 selector is set. In addition, the log line for abandonment now contains
916 information about the queue run such as the pid. This is always present
917 in "start" and "stop" lines but was omitted from the "abandon" line.
919 PH/34 Omit spaces between a header name and the colon in the error message that
920 is given when verify = headers_syntax fails (if there are lots of them,
921 the message gets confusing).
923 PH/35 Change the default for dns_check_names_pattern to allow slashes within
924 names, as there are now some PTR records that contain slashes. This check
925 is only to protect against broken name servers that fall over on strange
926 characters, so the fact that it applies to all lookups doesn't matter.
928 PH/36 Now that the new test suite is complete, we can remove some of the
929 special code in Exim that was needed for the old test suite. For example,
930 sorting DNS records because real resolvers return them in an arbitrary
931 order. The new test suite's fake resolver always returns records in the
934 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
935 bounces) except when queue_only is set, to avoid logging races between
936 the different processes.
938 PH/38 Panic-die if .include specifies a non-absolute path.
940 PH/39 A tweak to the "H" retry rule from its user.
942 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
943 a label. They prevented compilation on older perls.
945 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
946 a warning to be raised on newish perls.
948 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
949 on queue. Changes to match documented behaviour of showing count of
950 messages matching specified criteria.
952 PH/40 Changed the default ident timeout from 30s to 5s.
954 PH/41 Added support for the use of login_cap features, on those BSD systems
955 that have them, for controlling the resources used by pipe deliveries.
957 PH/42 The content-scanning code uses fopen() to create files in which to put
958 message data. Previously it was not paying any attention to the mode of
959 the files. Exim runs with umask(0) because the rest of the code creates
960 files with open(), and sets the required mode explicitly. Thus, these
961 files were ending up world-writeable. This was not a big issue, because,
962 being within the spool directory, they were not world-accessible. I have
963 created a function called modefopen, which takes an additional mode
964 argument. It sets umask(777), creates the file, chmods it to the required
965 mode, then resets the umask. All the relevant calls to fopen() in the
966 content scanning code have been changed to use this function.
968 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
969 to 24 hours. This avoids potential overflow problems when processing G
970 and H retry rules. I suspect nobody ever tinkers with this value.
972 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
974 PH/45 When the plaintext authenticator is running as a client, the server's
975 challenges are checked to ensure they are valid base64 strings. By
976 default, the authentication attempt is cancelled if an invalid string is
977 received. Setting client_ignore_invalid_base64 true ignores these errors.
978 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
979 they are received. Thus, the responses can be made to depend on the
980 challenges. If an invalid string is ignored, an empty string is placed in
983 PH/46 Messages that are created by the autoreply transport now contains a
984 References: header, in accordance with RFCs 2822 and 3834.
986 PH/47 Added authenticated_sender_force to the smtp transport.
988 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
990 PH/49 Installed latest patch from the Sieve maintainer.
992 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
993 was also set, the appendfile transport was unnecessarily scanning a
994 directory of message files (e.g. for maildir delivery) to find the count
995 of files (along with the size), even though it did not need this
996 information. It now does the scan only if it needs to find either the
997 size of the count of files.
999 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1001 PH/52 Two bugs concerned with error handling when the smtp transport is
1004 (i) Exim was not creating retry information for temporary errors given
1005 for individual recipients after the DATA command when the smtp transport
1006 was used in LMTP mode. This meant that they could be retried too
1007 frequently, and not timed out correctly.
1009 (ii) Exim was setting the flag that allows error details to be returned
1010 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1011 recipients that were returned after the DATA command.
1013 PH/53 This is related to PH/52, but is more general: for any failing address,
1014 when detailed error information was permitted to be returned to the
1015 sender, but the error was temporary, then after the final timeout, only
1016 "retry timeout exceeded" was returned. Now it returns the full error as
1017 well as "retry timeout exceeded".
1019 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1020 do this, and (what is worse) MTAs that accept it.
1022 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1023 will now be deprecated.
1025 PH/56 New os.c-cygwin from the Cygwin maintainer.
1027 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1028 formats (previously only available in exim formats via -bpr, -bpru,
1029 and -bpra. Now also available in native and exiqgrep formats)
1031 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1032 with very large, slow to parse queues
1034 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1036 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1038 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1039 responses to authentication challenges, though it was showing the
1040 challenges; (ii) I've removed the CR characters from the debug output for
1043 PH/58 Allow for the insertion of a newline as well as a space when a string
1044 is turned into more than one encoded-word during RFC 2047 encoding. The
1045 Sieve code now uses this.
1047 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1048 data_4xx, lost_connection, tls_required.
1050 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1051 message was being sent as an SMTP response.
1053 PH/61 Add -l and -k options to exicyclog.
1055 PH/62 When verifying, if an address was redirected to one new address, so that
1056 verification continued, and the new address failed or deferred after
1057 having set something in $address_data, the value of $address_data was not
1058 passed back to the ACL. This was different to the case when no
1059 redirection occurred. The value is now passed back in both cases.
1061 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1062 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1063 use login_cap.h, so on its own it isn't the distinguishing feature. The
1064 new name refers directly to the setclassresources() function.
1066 PH/65 Added configuration files for NetBSD3.
1068 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1070 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1071 is preferred over IPv4.
1073 PH/68 The bounce_return_message and bounce_return_body options were not being
1074 honoured for bounces generated during the reception of non-SMTP messages.
1075 In particular, this applied to messages rejected by the ACL. This bug has
1076 been fixed. However, if bounce_return_message is true and bounce_return_
1077 body is false, the headers that are returned for a non-SMTP message
1078 include only those that have been read before the error was detected.
1079 (In the case of an ACL rejection, they have all been read.)
1081 PH/69 The HTML version of the specification is now built in a directory called
1082 spec_html instead of spec.html, because the latter looks like a path with
1083 a MIME-type, and this confuses some software.
1085 PH/70 Catch two compiler warnings in sieve.c.
1087 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1088 function verify_get_ident() calls ip_connect() to connect a socket, but
1089 if the "connect()" function timed out, ip_connect() used to close the
1090 socket. However, verify_get_ident() also closes the socket later, and in
1091 between Exim writes to the log, which may get opened at this point. When
1092 the socket was closed in ip_connect(), the log could get the same file
1093 descriptor number as the socket. This naturally causes chaos. The fix is
1094 not to close the socket in ip_connect(); the socket should be closed by
1095 the function that creates it. There was only one place in the code where
1096 this was missing, in the iplookup router, which I don't think anybody now
1097 uses, but I've fixed it anyway.
1099 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1100 well as to direct DNS lookups. Otherwise the handling of names in host
1101 lists is inconsistent and therefore confusing.
1107 PH/01 Two changes to the default runtime configuration:
1109 (1) Move the checks for relay_from_hosts and authenticated clients from
1110 after to before the (commented out) DNS black list checks.
1112 (2) Add control=submission to the relay_from_hosts and authenticated
1113 clients checks, on the grounds that messages accepted by these
1114 statements are most likely to be submissions.
1116 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1118 (1) Generate an error if the third argument for the ${prvs expansion is
1121 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1124 (3) Reset the variables that are obtained from the first argument of
1125 ${prvscheck and used in the second argument before leaving the code,
1126 because their memory is reclaimed, so using them afterwards may do
1129 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1130 one (it's much easier than Tom thought :-).
1132 (5) Because of (4), we can now allow for the use of $prvscheck_result
1133 inside the third argument.
1135 PH/03 For some reason, the default setting of PATH when running a command from
1136 a pipe transport was just "/usr/bin". I have changed it to
1139 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1140 anything to be listed in the output from -bV.
1142 PH/05 When a filter generated an autoreply, the entire To: header line was
1143 quoted in the delivery log line, like this:
1145 => >A.N.Other <ano@some.domain> <original@ddress> ...
1147 This has been changed so that it extracts the operative address. There
1148 may be more than one such address. If so, they are comma-separated, like
1151 => >ano@some.domain,ona@other.domain <original@ddress> ...
1153 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1154 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1155 not being looked up in the rDNS to get a host name, Exim was showing the
1156 IP address twice in Received: lines, even though the IP addresses were
1157 identical. For example:
1159 Received: from [1.2.3.4] (helo=[1.2.3.4])
1161 However, if the real host name was known, it was omitting the HELO data
1162 if it matched the actual IP address. This has been tidied up so that it
1163 doesn't show the same IP address twice.
1165 PH/07 When both +timestamp and +memory debugging was on, the value given by
1166 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1167 called by the debug printing, thereby overwriting the timestamp buffer.
1168 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1170 PH/08 When the original message was included in an autoreply transport, it
1171 always said "this is a copy of the message, including all the headers",
1172 even if body_only or headers_only was set. It now gives an appropriate
1175 PH/09 Applied a patch from the Sieve maintainer which:
1177 o fixes some comments
1178 o adds the (disabled) notify extension core
1179 o adds some debug output for the result of if/elsif tests
1180 o points to the current vacation draft in the documentation
1181 and documents the missing references header update
1185 o fixes a bug in processing the envelope test (when testing
1186 multiple envelope elements, the last element determinted the
1189 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1190 Electronic Mail") by including:
1192 Auto-submitted: auto-generated
1194 in the messages that it generates (bounce messages and others, such as
1195 warnings). In the case of bounce messages for non-SMTP mesages, there was
1196 also a typo: it was using "Auto_submitted" (underscore instead of
1197 hyphen). Since every message generated by Exim is necessarily in response
1198 to another message, thes have all been changed to:
1200 Auto-Submitted: auto-replied
1202 in accordance with these statements in the RFC:
1204 The auto-replied keyword:
1206 - SHOULD be used on messages sent in direct response to another
1207 message by an automatic process,
1209 - MUST NOT be used on manually-generated messages,
1211 - MAY be used on Delivery Status Notifications (DSNs) and Message
1212 Disposition Notifications (MDNs),
1214 - MUST NOT be used on messages generated by automatic or periodic
1215 processes, except for messages which are automatic responses to
1218 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1219 to the default Received: header definition.
1221 PH/12 Added log selector acl_warn_skipped (default on).
1223 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1224 variables because (a) they are in the wrong storage pool and (b) even if
1225 they were copied, it wouldn't work properly because of the caching.
1227 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1228 checking when decoding. Apparently there are clients that generate
1229 overlong encoded strings. Why am I not surprised?
1231 PH/15 If the first argument of "${if match_address" was not empty, but did not
1232 contain an "@" character, Exim crashed. Now it writes a panic log message
1233 and treats the condition as false.
1235 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1237 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1238 extension "envelope-auth". The code is finished and in agreement with
1239 other implementations, but there is no documentation so far and in fact,
1240 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1241 not changing the active code.
1243 Print executed "if" and "elsif" statements when debugging is used. This
1244 helps a great deal to understand what a filter does.
1246 Document more things not specified clearly in RFC3028. I had all this
1247 sorted out, when out of a sudden new issues came to my mind. Oops."
1249 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1252 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1253 canonical form (as documented). However, after a host name lookup from
1254 the IP address, check_host() was doing a simple string comparison with
1255 addresses acquired from the DNS when checking that the found name did
1256 have the original IP as one of its addresses. Since any found IPv6
1257 addresses are likely to be in abbreviated form, the comparison could
1258 fail. Luckily, there already exists a function for doing the comparison
1259 by converting both addresses to binary, so now that is used instead of
1260 the text comparison.
1262 PH/20 There was another similar case to PH/19, when a complete host name was
1263 given in a host list; looking up its IP address could give an abbreviated
1264 form, whereas the current host's name might or might not be abbreviated.
1265 The same fix has been applied.
1271 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1272 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1275 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1276 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1278 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1280 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1281 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1282 Also, for "Auto-Submitted", treat anything other than "no" as
1283 non-personal, in accordance with RFC 3834. (Previously it treated
1284 anything starting "auto-" as non-personal.)
1286 TF/01 The control=submission/name=... option had a problem with syntax
1287 errors if the name included a slash character. The /name= option
1288 now slurps the rest of the string, so it can include any characters
1289 but it must come last in the list of options (after /sender_retain
1292 PH/05 Some modifications to the interface to the fake nameserver for the new
1300 TK/01 Added the "success_on_redirect" address verification option. See
1301 NewStuff for rationale and an example.
1303 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1305 PH/02 Patch to exigrep to allow it to work on syslog lines.
1307 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1308 fread() to skip over the body file's header line, because in Cygwin the
1309 header line is locked and is inaccessible.
1311 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1312 co-exist for some time) to make it clear that it is the Exim ID that is
1313 referenced, not the Message-ID: header line.
1315 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1316 string_format() function, because snprintf() does not exist on all
1319 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1320 ${stat: expansion item.
1322 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1323 protocol synchronization error", to keep the pedants happy.
1325 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1326 well as for IRIX systems, when gcc is being used. See the host.c source
1329 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1331 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1334 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1335 transport and to the smtp transport in LMTP mode.
1337 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1339 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1340 run a filter in a subprocess. This could lead to confusion in subsequent
1341 lookups in the parent process. There should also be a search_tidyup() at
1342 the end of the subprocess.
1344 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1345 only if the host matched helo_try_verify_hosts, which caused the
1346 verification to occur when the EHLO/HELO command was issued. The ACL just
1347 tested the remembered result. Now, if a previous verification attempt has
1348 not happened, "verify = helo" does it there and then.
1350 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1352 TK/03 Fix log output including CR from clamd.
1354 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1355 exist provoked a memory error which could cause a segfault.
1357 PH/15 Installed PCRE 6.2
1359 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1361 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1362 of the problem. Specifically, suggested +O2 rather than +O1 for the
1365 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1367 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1368 record, the port was not correctly logged when the outgoing_port log
1369 selector was set (it logged the transort's default port).
1371 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1372 fallback_hosts, and "hosts" in the smtp transport.
1374 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1375 host errors such as "Connection refused".
1377 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1378 authentication with radiusclient 0.4.9:
1380 - Error returned from rc_read_config was caught wrongly
1381 - Username/password not passed on to radius server due to wrong length.
1383 The presumption is that some radiusclient API changes for 4.51/PH/17
1384 were not taken care of correctly. The code is still untested by me (my
1385 Linux distribution still has 0.3.2 of radiusclient), but it was
1386 contributed by a Radius user.
1388 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1389 expanding the "port" option of the smtp transport.
1391 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1392 while reading a MIME header. Thanks to Tom Hughes for a patch.
1394 PH/24 Include config.h inside local_scan.h so that configuration settings are
1397 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1398 that in an ACL for RCPT (for example), you can examine exactly what was
1401 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1402 commands, but it was not correctly comparing the address with the actual
1403 client host address. Thus, it would show the EHLO address in Received:
1404 header lines when this was not necessary.
1406 PH/27 Added the % operator to ${eval:}.
1408 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1409 it should be ignoring failures (because with -C, for example, it has lost
1410 privilege). It wasn't ignoring creation failures other than "already
1413 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1416 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1417 because an input line was too long, either on its own, or by virtue of
1418 too many continuations, the temporary file was not being removed, and the
1419 return code was incorrect.
1421 PH/31 Missing "BOOL" in function definition in filtertest.c.
1423 PH/32 Applied Sieve patches from the maintainer.
1425 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1427 PH/33 Added "verify = not_blind".
1429 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1430 Local/Makefile (with some defaults set). These are used in built scripts
1431 such as exicyclog, but they have never been used in the exim_install
1432 script (though there are many overriding facilities there). I have
1433 arranged that the exim_install script now takes note of these two
1436 PH/35 Installed configuration files for Dragonfly.
1438 PH/36 When a locally submitted message by a trusted user did not contain a
1439 From: header, and the sender address was obtained from -f or from an SMTP
1440 MAIL command, and the trusted user did not use -F to supply a sender
1441 name, $originator_name was incorrectly used when constructing a From:
1442 header. Furthermore, $originator_name was used for submission mode
1443 messages from external hosts without From: headers in a similar way,
1444 which is clearly wrong.
1446 PH/37 Added control=suppress_local_fixups.
1448 PH/38 When log_selector = +received_sender was set, and the addition of the
1449 sender made the log line's construction buffer exactly full, or one byte
1450 less than full, an overflow happened when the terminating "\n" was
1453 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1454 when the result of a list match is failure because a DNS lookup failed.
1456 PH/40 RM_COMMAND is now used in the building process.
1458 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1459 the "build-* directories that it finds.
1461 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1462 address in a domain literal was a prefix of an interface address.
1464 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1465 when verifying a sender address, unless rewrite_headers is false.
1467 PH/44 Wrote a long comment about why errors_to addresses are verified as
1468 recipients, not senders.
1470 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1471 the ratelimit ACL was added.
1473 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1475 PH/46 Added extra information about PostgreSQL errors to the error string.
1477 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1478 avoiding the need to install special zones in a real server. This is
1479 backwards compatible; if it can't find the fake resolver, it drops back.
1480 Thus, both old and new test suites can be run.
1482 TF/02 Added util/ratelimit.pl
1484 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1485 clock is set back in time.
1487 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1488 Candler <B.Candler@pobox.com>.
1490 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1491 OK for addresses that are the result of redirections.
1493 PH/48 A number of further additions for the benefit of the new test suite,
1494 including a fake gethostbyname() that interfaces to the fake DNS resolver
1497 TF/06 The fix for widen_domains has also been applied to qualify_single and
1498 search_parents which are the other dnslookup options that can cause
1501 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1504 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1506 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1507 service" if it encountered an MX record with an empty target hostname.
1508 The message is now "an MX or SRV record indicated no SMTP service".
1510 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1511 if the DNS of the sending site is misconfigured. This is quite a
1512 common situation. This change restores the behaviour of treating a
1513 helo verification defer as a failure.
1515 PH/51 If self=fail was set on a router, the bounce message did not include the
1516 actual error message.
1522 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1524 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1525 command itself ended in error, the underlying message about the transport
1526 filter timeout was being overwritten with the pipe command error. Now the
1527 underlying error message should be appended to the second error message.
1529 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1531 PH/02 Reduce the amount of output that "make" produces by default. Full output
1532 can still be requested.
1534 PH/03 The warning log line about a condition test deferring for a "warn" verb
1535 was being output only once per connection, rather than after each
1536 occurrence (because it was using the same function as for successful
1537 "warn" verbs). This seems wrong, so I have changed it.
1539 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1540 it should not have, which might have caused a crash in the right
1541 circumstances, but probably never did.
1543 PH/04 Installed a modified version of Tony Finch's patch to make submission
1544 mode fix the return path as well as the Sender: header line, and to
1545 add a /name= option so that you can make the user's friendly name appear
1548 TF/03 Added the control = fakedefer ACL modifier.
1550 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1551 Mark Lowes for thorough testing.
1553 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1555 TK/03 Merged latest SRS patch from Miles Wilton.
1557 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1558 with the definition in sysexits.h (which is #included earlier).
1559 Fortunately, Exim does not actually use EX_OK. The code used to try to
1560 preserve the sysexits.h value, by assumimg that macro definitions were
1561 scanned for macro replacements. I have been disabused of this notion,
1562 so now the code just undefines EX_OK before #including unistd.h.
1564 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1565 in the smtp transport. When a block could not be written in a single
1566 write() function, the timeout was being re-applied to each part-write.
1567 This seems wrong - if the receiver was accepting one byte at a time it
1568 would take for ever. The timeout is now adjusted when this happens. It
1569 doesn't have to be particularly precise.
1571 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1572 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1574 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1575 without a domain if the check to <postmaster@domain> fails.
1577 SC/01 Eximstats: added -xls and the ability to specify output files
1578 (patch written by Frank Heydlauf).
1580 SC/02 Eximstats: use FileHandles for outputing results.
1582 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1584 SC/04 Eximstats: fixed display of large numbers with -nvr option
1586 SC/05 Eximstats: fixed merging of reports with empty tables.
1588 SC/06 Eximstats: added the -include_original_destination flag
1590 SC/07 Eximstats: removed tabs and trailing whitespace.
1592 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1594 TK/06 MBOX spool code: Add real "From " MBOX separator line
1595 so the .eml file is really in mbox format (even though
1596 most programs do not really care). Patch from Alex Miller.
1598 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1599 The latter is generated from $received_to and is only set if the
1600 message has one envelope recipient. SA can use these headers,
1601 obviously out-of-the-box. Patch from Alex Miller.
1603 PH/08 The ${def test on a variable was returning false if the variable's
1604 value was "0", contrary to what the specification has always said!
1605 The result should be true unless the variable is empty.
1607 PH/09 The syntax error of a character other than { following "${if
1608 def:variable_name" (after optional whitespace) was not being diagnosed.
1609 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1610 accidental colon was present, for example, could give incorrect results.
1612 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1613 result is used (not including appendfile, where other changes are about
1616 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1617 This involved changing a lot of size variables from int to off_t. It
1618 should work with maildirs and everything.
1620 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1621 spamd dying while we are connected to it.
1623 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1626 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1627 function for BDB 4.3.
1629 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1631 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1632 the sources for PCRE have been re-arranged and more files are now
1635 PH/15 The code I had for printing potentially long long variables in PH/11
1636 above was not the best (it lost precision). The length of off_t variables
1637 is now inspected at build time, and an appropriate printing format (%ld
1638 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1639 to be "long long int" or "long int". This is needed for the internal
1640 formatting function string_vformat().
1642 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1643 the configuration file to be ":syslog", then the script "guesses" where
1644 the logs files are, rather than using the compiled in default. In our
1645 case the guess is not the same as the compiled default, so the script
1646 suddenly stopped working when I started to use syslog. The patch checks
1647 to see if log_file_path is "". If so, it attempts to read it from exim
1648 with no configuration file to get the compiled in version, before it
1649 falls back to the previous guessing code."
1651 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1652 implementing BATV in an Exim configuration. See NewStuff for the gory
1655 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1656 Makefile that are specific to HP-UX.
1658 PH/18 If the "use_postmaster" option was set for a recipient callout together
1659 with the "random" option, the postmaster address was used as the MAIL
1660 FROM address for the random test, but not for the subsequent recipient
1661 test. It is now used for both.
1663 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1664 patch removes a few documentation additions to RFC 3028, because the
1665 latest draft now contains them. It adds the new en;ascii-case comparator
1666 and a new error check for 8bit text in MIME parts. Comparator and
1667 require names are now matched exactly. I enabled the subaddress
1668 extension, but it is not well tested yet (read: it works for me)."
1670 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1671 rework some of the code of TK/09 above to avoid the hardwired use of
1672 "%lld" and "long long". Replaced the call to snprintf() with a call to
1675 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1676 records point to non-existent hosts", "retry timeout exceeded", and
1677 "retry time not reached for any host after a long failure period".
1679 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1680 experimental DomainKeys support:
1682 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1683 (2) On an error such as an illegally used "control", the wrong name for
1684 the control was given.
1686 These problems did NOT occur unless DomainKeys support was compiled.
1688 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1690 PH/24 Added ${if match_ip condition.
1692 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1693 databases so that it will be absolutely obvious if a crash occurs in the
1694 DB library. This is a regular occurrence (often caused by mis-matched
1697 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1698 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1699 picked up on a user's system that detects such things. There doesn't seem
1700 to be a gcc warning option for this - only an attribute that has to be
1701 put on the function's prototype. It seems that in Fedora Core 4 they have
1702 set this on a number of new functions. No doubt there will be more in due
1705 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1706 specify a transport. However, if an address that was verified by such a
1707 router was the subject of a callout, Exim crashed because it tried to
1708 read the rcpt_include_affixes from the non-existent transport. Now it
1709 just assumes that the setting of that option is false. This bug was
1710 introduced by 4.51/PH/31.
1712 PH/28 Changed -d+all to exclude +memory, because that information is very
1713 rarely of interest, but it makes the output a lot bigger. People tend to
1714 do -d+all out of habit.
1716 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1717 code in os-type was giving problems when libc.so lives in lib64, like on
1720 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1721 aren't the modern standard, and it seems that some systems' include files
1722 don't always have them. Exim was already checking for some of the newer
1723 ones like T_AAAA, and defining it itself. I've added checks for all the
1724 record types that Exim uses.
1726 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1727 not automatically generating a new one, as it is supposed to. This
1728 prevented TLS from working. If the file did exist, but contained invalid
1729 data, a new version was generated, as expected. It was only the case of a
1730 non-existent file that was broken.
1732 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1733 with a change in libdomainkeys > 0.64.
1735 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1736 from DNS. If the selector record carries the flag, it now has
1737 precedence over the domain-wide flag.
1739 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1741 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1742 the use of a port name that isn't defined in /etc/services) occurred, the
1743 message was deferred as in a normal delivery, and thus remained on the
1744 spool, instead of being failed because of the mua_wrapper setting. This
1745 is now fixed, and I tidied up some of the mua_wrapper messages at the
1748 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1749 the messages in a hash of arrays rather than using individual hashes.
1750 This is a bit cleaner and results in dramatic memory savings, albeit
1751 at a slight CPU cost.
1753 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1754 as requested by Marc Sherman.
1756 SC/10 Eximstats: added histograms for user specified patterns as requested
1759 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1761 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1762 fopen() in the content-scanning modules that did not already have it.
1768 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1769 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1771 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1773 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1775 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1776 to test Sieve filters that use "vacation".
1778 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1779 that changes the way the GnuTLS parameters are stored in the cache file.
1780 The new format can be generated externally. For backward compatibility,
1781 if the data in the cache doesn't make sense, Exim assumes it has read an
1782 old-format file, and it generates new data and writes a new file. This
1783 means that you can't go back to an older release without removing the
1786 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1787 work if there are any delivery delays because "one_time" forces the
1788 parent to be marked "delivered", so its unseen clone is never tried
1789 again. For this reason, Exim now forbids the simultaneous setting of
1792 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1793 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1794 entry if you want to know the details. The fix, however, neglected to
1795 consider the case when local delivery batching is involved. The test for
1796 "previously delivered" was not happening when checking to see if an
1797 address could be batched with a previous (undelivered) one; under
1798 certain circumstances this could lead to multiple deliveries to the same
1801 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1802 in its include files, and this causes problems building Exim.
1804 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1805 header_syntax) but Exim was just ignoring anything given after a slash.
1806 In particular, this caused confusion with an attempt to use "verify =
1807 reverse_host_lookup/defer_ok". An error is now given when options are
1808 supplied for verify items that do not have them. (Maybe reverse_host_
1809 lookup should have a defer_ok option, but that's a different point.)
1811 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1812 defined by RFC 821) to 2048, because there were problems with some AUTH
1813 commands, and RFC 1869 says the size should be increased for extended
1814 SMTP commands that take arguments.
1816 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1819 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1820 "unknown" error; now it says that the functionality isn't in the binary.
1822 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1823 an address' error message when a string expansion fails (syntax or
1824 whatever). Otherwise the password may appear in the log. Following change
1825 PH/42 below, there is no longer a chance of it appearing in a bounce
1828 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1830 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1831 of its addresses was ever tried. (Bugzilla bug #2.)
1833 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1834 the result incorrectly in the debug output. (It correctly added a newline
1835 to what was transported.)
1837 TF/01 Added $received_time.
1839 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1840 commented out examples of how to interface to a virus scanner and to
1841 SpamAssassin. Also added commented examples of av_scanner and
1842 spamd_address settings.
1844 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1845 and controls are allowed in which ACLs. There were a couple of minor
1846 errors. Some of the entries in the conditions table (which is a table of
1847 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1848 negation of where the condition IS allowed.
1850 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1852 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1853 header file does not have a version number, so I've had to invent a new
1854 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1855 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1856 radiusclient), but it was contributed by a Radius user.
1858 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1859 files or directories, for OpenSSL.
1861 PH/19 When an Exim process that is running as root has to create an Exim log
1862 file, it does so in a subprocess that runs as exim:exim so as to get the
1863 ownership right at creation (otherwise, other Exim processes might see
1864 the file with the wrong ownership). There was no test for failure of this
1865 fork() call, which would lead to the process getting stuck as it waited
1866 for a non-existent subprocess. Forks do occasionally fail when resources
1867 run out. I reviewed all the other calls to fork(); they all seem to check
1870 PH/20 When checking for unexpected SMTP input at connect time (before writing
1871 the banner), Exim was not dealing correctly with a non-positive return
1872 from the read() function. If the client had disconnected by this time,
1873 the result was a log entry for a synchronization error with an empty
1874 string after "input=" when read() returned zero. If read() returned -1
1875 (an event I could not check), uninitialized data bytes were printed.
1876 There were reports of junk text (parts of files, etc) appearing after
1879 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1881 PH/22 Added support for macro redefinition, and (re)definition in between
1882 driver and ACL definitions.
1884 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1885 forgetting to use the resulting value; it was using the unexpanded value.
1887 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1888 hadn't been configured. The fix is from Juergen Kreileder, who
1889 understands it better than I do:
1891 "Here's what I see happening with three configured cyrus_sasl
1892 authenticators configured (plain, login, cram-md5):
1894 On startup auth_cyrus_sasl_init() gets called for each of these.
1895 This means three calls to sasl_listmech() without a specified mech_list.
1896 => SASL tests which mechs of all available mechs actually work
1897 => three warnings about OTP not working
1898 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1900 With the patch, sasl_listmech() also gets called three times. But now
1901 SASL's mech_list option is set to the server_mech specified in the the
1902 authenticator. Or in other words, the answer from sasl_listmech()
1903 gets limited to just the mech you're testing for (which is different
1905 => the return list contains just 'plain' or 'login', 'cram-md5' or
1906 nothing depending on the value of ob->server_mech.
1908 I've just tested the patch: Authentication still works fine,
1909 unavailable mechs specified in the exim configuration are still
1910 caught, and the auth.log warnings about OTP are gone."
1912 PH/25 When debugging is enabled, the contents of the command line are added
1913 to the debugging output, even when log_selector=+arguments is not
1916 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1917 answer is "GNU", and only if the return is "GNU/something" is the answer
1920 PH/27 $acl_verify_message is now set immediately after the failure of a
1921 verification in an ACL, and so is available in subsequent modifiers. In
1922 particular, the message can be preserved by coding like this:
1924 warn !verify = sender
1925 set acl_m0 = $acl_verify_message
1927 Previously, $acl_verify_message was set only while expanding "message"
1928 and "log_message" when a very denied access.
1930 PH/28 Modified OS/os.c-Linux with
1932 -#ifndef OS_LOAD_AVERAGE
1933 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1935 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1936 nomenclature these days.)
1938 PH/29 Installed patch from the Sieve maintainer that adds the options
1939 sieve_useraddress and sieve_subaddress to the redirect router.
1941 PH/30 In these circumstances:
1942 . Two addresses routed to the same list of hosts;
1943 . First host does not offer TLS;
1944 . First host accepts first address;
1945 . First host gives temporary error to second address;
1946 . Second host offers TLS and a TLS session is established;
1947 . Second host accepts second address.
1948 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1949 and peerdn, if requested) that were in fact used only for the second
1952 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1953 attention to any local part prefix or suffix that was matched by the
1954 router that accepted the address. It now behaves in the same way as it
1955 does for delivery: the affixes are removed from the local part unless
1956 rcpt_include_affixes is set on the transport.
1958 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1959 timeout during the DATA phase of an incoming message.
1961 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1962 applied a patch sanctioned by the Sieve maintainer.
1964 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1965 the uid or gid is negative. A case of a negative gid caused this to be
1966 noticed. The fix allows for either to be negative.
1968 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1969 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1972 PH/36 The change PH/12 above was broken. Fixed it.
1974 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1975 the grounds that routing the same address twice would always produce the
1976 same answer. This might have been true once, but it is certainly no
1977 longer true now. Routing a child address may depend on the previous
1978 routing that produced that child. Some complicated redirection strategies
1979 went wrong when messages had multiple recipients, and made Exim's
1980 behaviour dependent on the order in which the addresses were given.
1982 I have moved the duplicate checking until after the routing is complete.
1983 Exim scans the addresses that are assigned to local and remote
1984 transports, and removes any duplicates. This means that more work will be
1985 done, as duplicates will always all be routed, but duplicates are
1986 presumably rare, so I don't expect this is of any significance.
1988 For deliveries to pipes, files, and autoreplies, the duplicate checking
1989 still happens during the routing process, since they are not going to be
1992 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1993 It corrects a timeout issue with spamd. This is Ian's comment: "The
1994 background is that sometimes spamd either never reads data from a
1995 connection it has accepted, or it never writes response data. The exiscan
1996 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1997 blindly assumes that writes won't block so it may never time out."
1999 PH/39 Allow G after quota size as well as K and M.
2001 PH/40 The value set for $authenticated_id in an authenticator may not contain
2002 binary zeroes or newlines because the value is written to log lines and
2003 to spool files. There was no check on this. Now the value is run through
2004 the string_printing() function so that such characters are converted to
2005 printable escape sequences.
2007 PH/41 $message_linecount is a new variable that contains the total number of
2008 lines in the message. Compare $body_linecount, which is the count for the
2011 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2012 bounce and delay warning messages, except in certain special cases, which
2015 (a) An SMTP error message from a remote host;
2016 (b) A message specified in a :fail: redirection;
2017 (c) A message specified in a "fail" command in a system filter;
2018 (d) A message specified in a FAIL return from the queryprogram router;
2019 (e) A message specified by the cannot_route_message router option.
2021 In these cases only, Exim does include the error details in bounce and
2022 warning messages. There are also a few cases where bland messages such
2023 as "unrouteable address" or "local delivery error" are given.
2025 PH/43 $value is now also set for the "else" part of a ${run expansion.
2027 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2028 being worked on, but at least Exim now implements the latest version to
2031 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2032 process to complete was treated as a delivery failure, a timeout while
2033 writing the message to the pipe was logged, but erroneously treated as a
2034 successful delivery. Such timeouts include transport filter timeouts. For
2035 consistency with the overall process timeout, these timeouts are now
2036 treated as errors, giving rise to delivery failures by default. However,
2037 there is now a new Boolean option for the pipe transport called
2038 timeout_defer, which, if set TRUE, converts the failures into defers for
2039 both kinds of timeout. A transport filter timeout is now identified in
2042 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2043 systems where "make" and "gmake" are different, calling "gmake" at top
2044 level broke things. I've arranged for the value of $(MAKE) to be passed
2045 from the Makefile to this script so that it can call the same version of
2049 A note about Exim versions 4.44 and 4.50
2050 ----------------------------------------
2052 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2053 changes of various kinds. As a consequence, a big documentation update was
2054 needed. This delayed the release for rather longer than seemed good, especially
2055 in the light of a couple of (minor) security issues. Therefore, the changes
2056 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2057 release. So 4.44 and 4.50 are in effect two different branches that both start
2060 I have left the 4.50 change log unchanged; it contains all the changes since
2061 4.43. The change log for 4.44 is below; many of its items are identical to
2062 those for 4.50. This seems to be the most sensible way to preserve the
2063 historical information.
2069 1. Minor wording change to the doc/README.SIEVE file.
2071 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2072 computation of the current number of files was incorrect.
2074 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2075 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2078 4. Give more explanation in the error message when the command for a transport
2079 filter fails to execute.
2081 5. There are several places where Exim runs a non-Exim command in a
2082 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2083 was being done only for the command run by the queryprogram router. It is
2084 now done for all such subprocesses. The other cases are: ${run, transport
2085 filters, and the commands run by the lmtp and pipe transports.
2087 6. Added CONFIGURE_GROUP build-time option.
2089 7. Some older OS have a limit of 256 on the maximum number of file
2090 descriptors. Exim was using setrlimit() to set 1000 as a large value
2091 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2092 systems. I've change it so that if it can't get 1000, it tries for 256.
2094 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2095 was an oversight, and furthermore, ever since the addition of extra
2096 controls (e.g. 4.43/32), the checks on when to allow different forms of
2097 "control" were broken. There should now be diagnostics for all cases when a
2098 control that does not make sense is encountered.
2100 9. Added the /retain_sender option to "control=submission".
2102 10. $recipients is now available in the predata ACL (oversight).
2104 11. Tidy the search cache before the fork to do a delivery from a message
2105 received from the command line. Otherwise the child will trigger a lookup
2106 failure and thereby defer the delivery if it tries to use (for example) a
2107 cached ldap connection that the parent has called unbind on.
2109 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2110 of $address_data from the recipient verification was clobbered by the
2111 sender verification.
2113 13. The value of address_data from a sender verification is now available in
2114 $sender_address_data in subsequent conditions in the ACL statement.
2116 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2118 15. Added a new option "connect=<time>" to callout options, to set a different
2121 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2122 was its contents. (It was OK if the option was not defined at all.)
2124 17. A "Completed" log line is now written for messages that are removed from
2125 the spool by the -Mrm option.
2127 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2128 information about exactly what failed.
2130 19. Added -dd to debug only the daemon process.
2132 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2133 handles timeouts, both on the server side and network timeouts. Renamed the
2134 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2136 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2137 It is now set to "smtps".
2139 22. $host_address is now set to the target address during the checking of
2140 ignore_target_hosts.
2142 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2143 being passed; this would have caused $sender_host_name to have been used if
2144 matching the list had actually called for a host name (not very likely,
2145 since this list is usually IP addresses). A host name is now passed as
2148 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2149 code that specifies a non-restarting handler (typically sigaction() in
2150 modern systems) in an attempt to fix a rare and obscure crash bug.
2152 25. Narrowed the window for a race in the daemon that could cause it to ignore
2153 SIGCHLD signals. This is not a major problem, because they are used only to
2154 wake it up if nothing else does.
2156 26. A malformed maildirsize file could cause Exim to calculate negative values
2157 for the mailbox size or file count. Odd effects could occur as a result.
2158 The maildirsize information is now recalculated if the size or filecount
2161 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2162 support for a long time. Removed HAVE_SYS_VFS_H.
2164 28. Installed the latest version of exipick from John Jetmore.
2166 29. In an address list, if the pattern was not a regular expression, an empty
2167 subject address (from a bounce message) matched only if the pattern was an
2168 empty string. Non-empty patterns were not even tested. This was the wrong
2169 because it is perfectly reasonable to use an empty address as part of a
2170 database query. An empty address is now tested by patterns that are
2171 lookups. However, all the other forms of pattern expect the subject to
2172 contain a local part and a domain, and therefore, for them, an empty
2173 address still always fails if the pattern is not itself empty.
2175 30. Exim went into a mad DNS loop when attempting to do a callout where the
2176 host was specified on an smtp transport, and looking it up yielded more
2177 than one IP address.
2179 31. Re-factored the code for checking spool and log partition space into a
2180 function that finds that data and another that does the check. The former
2181 is then used to implement four new variables: $spool_space, $log_space,
2182 $spool_inodes, and $log_inodes.
2184 32. The RFC2047 encoding function was originally intended for short strings
2185 such as real names; it was not keeping to the 75-character limit for
2186 encoded words that the RFC imposes. It now respects the limit, and
2187 generates multiple encoded words if necessary. To be on the safe side, I
2188 have increased the buffer size for the ${rfc2047: expansion operator from
2191 33. It is now permitted to omit both strings after an "if" condition; if the
2192 condition is true, the result is "true". As before, when the second string
2193 is omitted, a false condition yields an empty string. This makes it less
2194 cumbersome to write custom ACL and router conditions.
2196 34. Failure to deliver a bounce message always caused it to be frozen, even if
2197 there was an errors_to setting on the router. The errors_to setting is now
2200 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2201 canonical form (fully expanded) before being placed in
2202 $sender_host_address.
2204 36. The table in the code that translates DNS record types into text (T_A to
2205 "A" for instance) was missing entries for NS and CNAME. It is just possible
2206 that this could have caused confusion if both these types were looked up
2207 for the same domain, because the text type is used as part of Exim's
2208 per-process caching. But the chance of anyone hitting this buglet seems
2211 37. The dnsdb lookup has been extended in a number of ways.
2213 (1) There is a new type, "zns", which walks up the domain tree until it
2214 finds some nameserver records. It should be used with care.
2216 (2) There is a new type, "mxh", which is like "mx" except that it returns
2217 just the host names, not the priorities.
2219 (3) It is now possible to give a list of domains (or IP addresses) to be
2220 looked up. The behaviour when one of the lookups defers can be
2221 controlled by a keyword.
2223 (4) It is now possible to specify the separator character for use when
2224 multiple records are returned.
2226 38. The dnslists ACL condition has been extended: it is now possible to supply
2227 a list of IP addresses and/or domains to be looked up in a particular DNS
2230 39. Added log_selector=+queue_time_overall.
2232 40. When running the queue in the test harness, wait just a tad after forking a
2233 delivery process, to get repeatability of debugging output.
2235 41. Include certificate and key file names in error message when GnuTLS fails
2236 to set them up, because the GnuTLS error message doesn't include the name
2237 of the failing file when there is a problem reading it.
2239 42. Allow both -bf and -bF in the same test run.
2241 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2243 44. The "Exiscan patch" is now merged into the mainline Exim source.
2245 45. Sometimes the final signoff response after QUIT could fail to get
2246 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2247 < 0 before doing a fflush(). This bug looks as though it goes back to the
2248 introduction of TLS in release 3.20, but "sometimes" must have been rare
2249 because the tests only now provoked it.
2251 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2252 (this can affect the format of dates).
2254 47. exim_tidydb, when checking for the continued existence of a message for
2255 which it has found a message-specific retry record, was not finding
2256 messages that were in split spool directories. Consequently, it was
2257 deleting retry records that should have stayed in existence.
2259 48. Steve fixed some bugs in eximstats.
2261 49. The SPA authentication driver was not abandoning authentication and moving
2262 on to the next authenticator when an expansion was forced to fail,
2263 contradicting the general specification for all authenticators. Instead it
2264 was generating a temporary error. It now behaves as specified.
2266 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2267 (the order specifies the preference for clients). The order is now AES256,
2268 AES128, 3DES, ARCFOUR128.
2270 51. Small patch to Sieve code - explicitly set From: when generating an
2273 52. Exim crashed if a remote delivery caused a very long error message to be
2274 recorded - for instance if somebody sent an entire SpamAssassin report back
2275 as a large number of 550 error lines. This bug was coincidentally fixed by
2276 increasing the size of one of Exim's internal buffers (big_buffer) that
2277 happened as part of the Exiscan merge. However, to be on the safe side, I
2278 have made the code more robust (and fixed the comments that describe what
2281 53. Now that there can be additional text after "Completed" in log lines (if
2282 the queue_time_overall log selector is set), a one-byte patch to exigrep
2283 was needed to allow it to recognize "Completed" as not the last thing in
2286 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2287 patch that reportedly fixes this has been added. I am not expert enough to
2288 create a test for it. This is what the patch creator wrote:
2290 "I found a little strange behaviour of ldap code when working with
2291 Windows 2003 AD Domain, where users was placed in more than one
2292 Organization Units. When I tried to give exim partial DN, the exit code
2293 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2294 But simultaneously result of request was absolutely normal ldap result,
2295 so I produce this patch..."
2297 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2298 REFERENCE, so I have modified the code to exclude the patch when that macro
2301 55. Some experimental protocols are using DNS PTR records for new purposes. The
2302 keys for these records are domain names, not reversed IP addresses. The
2303 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2304 leaves it alone. Component reversal etc. now happens only for IP addresses.
2307 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2309 57. Double the size of the debug message buffer (to 2048) so that more of very
2310 long debug lines gets shown.
2312 58. The exicyclog utility now does better if the number of log files to keep
2313 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2315 59. Two changes related to the smtp_active_hostname option:
2317 (1) $smtp_active_hostname is now available as a variable.
2318 (2) The default for smtp_banner uses $smtp_active_hostname instead
2319 of $primary_hostname.
2321 60. The host_aton() function is supposed to be passed a string that is known
2322 to be a valid IP address. However, in the case of IPv6 addresses, it was
2323 not checking this. This is a hostage to fortune. Exim now panics and dies
2324 if the condition is not met. A case was found where this could be provoked
2325 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2326 components; fortuitously, this particular loophole had already been fixed
2327 by change 4.50/55 above.
2329 If there are any other similar loopholes, the new check in host_aton()
2330 itself should stop them being exploited. The report I received stated that
2331 data on the command line could provoke the exploit when Exim was running as
2332 exim, but did not say which command line option was involved. All I could
2333 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2334 running as the user.
2337 61. There was a buffer overflow vulnerability in the SPA authentication code
2338 (which came originally from the Samba project). I have added a test to the
2339 spa_base64_to_bits() function which I hope fixes it.
2342 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2343 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2345 63. The daemon start-up calls getloadavg() while still root for those OS that
2346 need the first call to be done as root, but it missed one case: when
2347 deliver_queue_load_max is set with deliver_drop_privilege. This is
2348 necessary for the benefit of the queue runner, because there is no re-exec
2349 when deliver_drop_privilege is set.
2351 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2352 This has been fixed.
2354 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2355 was in use, was not putting the data itself into the right store pool;
2356 consequently, it could be overwritten for a subsequent message in the same
2357 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2360 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2362 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2363 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2364 the function were treating the return as a boolean value, which happened to
2365 work because 0=false and not-0=true, but is not correct code.
2367 68. The host_aton() function was not handling scoped IPv6 addresses (those
2368 with, for example, "%eth0" on the end) correctly.
2370 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2371 negated items (that is, ~something) in unsigned ints. Some compilers
2372 apparently mutter when there is no cast.
2374 70. If an address verification called from an ACL failed, and did not produce a
2375 user-specific message (i.e. there was only a "system" message), nothing was
2376 put in $acl_verify_message. In this situation, it now puts the system
2379 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2380 session; change 4.31/43 added the unwanted input to the log line - except
2381 that it did not do this in the start of session case. It now does.
2383 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2384 This is wrong and can cause the other end to generate a synchronization
2385 error if it is another Exim or anything else that does the synchronization
2386 check. A QUIT command is no longer sent after a timeout.
2388 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2389 during host lookups.
2391 74. The defer_ok option of callout verification was not working if it was used
2392 when verifying addresses in header lines, that is, for this case:
2394 verify = header_sender/callout=defer_ok
2396 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2397 those file descriptors could be used for SMTP connections. If anything
2398 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2399 could be sent to the SMTP client, causing chaos. The daemon now opens
2400 stdin, stdout, and stderr to /dev/null when it puts itself into the
2403 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2404 log by default. The user can override this with suitable Perl magic.
2406 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2407 the log message when discard triggers, was not working for the DATA ACL or
2408 for the non-SMTP ACL.
2410 78. Error message wording change in sieve.c.
2412 79. If smtp_accept_max_per_host was set, the number of connections could be
2413 restricted to fewer than expected, because the daemon was trying to set up
2414 a new connection before checking whether the processes handling previous
2415 connections had finished. The check for completed processes is now done
2416 earlier. On busy systems, this bug wouldn't be noticed because something
2417 else would have woken the daemon, and it would have reaped the completed
2420 80. If a message was submitted locally by a user whose login name contained one
2421 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2422 It caused a spool format error. I have fixed the spool reading code. A
2423 related problem was that the "from" clause in the Received: line became
2424 illegal because of the space(s). It is now covered by ${quote_local_part.
2426 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2429 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2431 83. Patch from Sieve maintainer to fix unterminated string problem in
2432 "vacation" handling.
2434 84. Some minor changes to the Linux configuration files to help with other
2435 OS variants using glibc.
2437 85. One more patch for Sieve to update vacation handling to latest spec.
2440 ----------------------------------------------------
2441 See the note above about the 4.44 and 4.50 releases.
2442 ----------------------------------------------------
2448 1. Change 4.43/35 introduced a bug that caused file counts to be
2449 incorrectly computed when quota_filecount was set in an appendfile
2452 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2453 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2456 3. Give more explanation in the error message when the command for a transport
2457 filter fails to execute.
2459 4. There are several places where Exim runs a non-Exim command in a
2460 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2461 was being done only for the command run by the queryprogram router. It is
2462 now done for all such subprocesses. The other cases are: ${run, transport
2463 filters, and the commands run by the lmtp and pipe transports.
2465 5. Some older OS have a limit of 256 on the maximum number of file
2466 descriptors. Exim was using setrlimit() to set 1000 as a large value
2467 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2468 systems. I've change it so that if it can't get 1000, it tries for 256.
2470 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2471 was an oversight, and furthermore, ever since the addition of extra
2472 controls (e.g. 4.43/32), the checks on when to allow different forms of
2473 "control" were broken. There should now be diagnostics for all cases when a
2474 control that does not make sense is encountered.
2476 7. $recipients is now available in the predata ACL (oversight).
2478 8. Tidy the search cache before the fork to do a delivery from a message
2479 received from the command line. Otherwise the child will trigger a lookup
2480 failure and thereby defer the delivery if it tries to use (for example) a
2481 cached ldap connection that the parent has called unbind on.
2483 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2484 of $address_data from the recipient verification was clobbered by the
2485 sender verification.
2487 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2488 was its contents. (It was OK if the option was not defined at all.)
2490 11. A "Completed" log line is now written for messages that are removed from
2491 the spool by the -Mrm option.
2493 12. $host_address is now set to the target address during the checking of
2494 ignore_target_hosts.
2496 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2497 being passed; this would have caused $sender_host_name to have been used if
2498 matching the list had actually called for a host name (not very likely,
2499 since this list is usually IP addresses). A host name is now passed as
2502 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2503 code that specifies a non-restarting handler (typically sigaction() in
2504 modern systems) in an attempt to fix a rare and obscure crash bug.
2506 15. Narrowed the window for a race in the daemon that could cause it to ignore
2507 SIGCHLD signals. This is not a major problem, because they are used only to
2508 wake it up if nothing else does.
2510 16. A malformed maildirsize file could cause Exim to calculate negative values
2511 for the mailbox size or file count. Odd effects could occur as a result.
2512 The maildirsize information is now recalculated if the size or filecount
2515 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2516 support for a long time. Removed HAVE_SYS_VFS_H.
2518 18. Updated exipick to current release from John Jetmore.
2520 19. Allow an empty sender to be matched against a lookup in an address list.
2521 Previously the only cases considered were a regular expression, or an
2524 20. Exim went into a mad DNS lookup loop when doing a callout where the
2525 host was specified on the transport, if the DNS lookup yielded more than
2528 21. The RFC2047 encoding function was originally intended for short strings
2529 such as real names; it was not keeping to the 75-character limit for
2530 encoded words that the RFC imposes. It now respects the limit, and
2531 generates multiple encoded words if necessary. To be on the safe side, I
2532 have increased the buffer size for the ${rfc2047: expansion operator from
2535 22. Failure to deliver a bounce message always caused it to be frozen, even if
2536 there was an errors_to setting on the router. The errors_to setting is now
2539 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2540 canonical form (fully expanded) before being placed in
2541 $sender_host_address.
2543 24. Updated eximstats to version 1.33
2545 25. Include certificate and key file names in error message when GnuTLS fails
2546 to set them up, because the GnuTLS error message doesn't include the name
2547 of the failing file when there is a problem reading it.
2549 26. Expand error message when OpenSSL has problems setting up cert/key files.
2552 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2553 (this can affect the format of dates).
2555 28. exim_tidydb, when checking for the continued existence of a message for
2556 which it has found a message-specific retry record, was not finding
2557 messages that were in split spool directories. Consequently, it was
2558 deleting retry records that should have stayed in existence.
2560 29. eximstats updated to version 1.35
2561 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2562 1.35 - bugfix such that pie charts by volume are generated correctly
2564 30. The SPA authentication driver was not abandoning authentication and moving
2565 on to the next authenticator when an expansion was forced to fail,
2566 contradicting the general specification for all authenticators. Instead it
2567 was generating a temporary error. It now behaves as specified.
2569 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2570 (the order specifies the preference for clients). The order is now AES256,
2571 AES128, 3DES, ARCFOUR128.
2573 31. Small patch to Sieve code - explicitly set From: when generating an
2576 32. Exim crashed if a remote delivery caused a very long error message to be
2577 recorded - for instance if somebody sent an entire SpamAssassin report back
2578 as a large number of 550 error lines. This bug was coincidentally fixed by
2579 increasing the size of one of Exim's internal buffers (big_buffer) that
2580 happened as part of the Exiscan merge. However, to be on the safe side, I
2581 have made the code more robust (and fixed the comments that describe what
2584 33. Some experimental protocols are using DNS PTR records for new purposes. The
2585 keys for these records are domain names, not reversed IP addresses. The
2586 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2587 leaves it alone. Component reversal etc. now happens only for IP addresses.
2590 34. The host_aton() function is supposed to be passed a string that is known
2591 to be a valid IP address. However, in the case of IPv6 addresses, it was
2592 not checking this. This is a hostage to fortune. Exim now panics and dies
2593 if the condition is not met. A case was found where this could be provoked
2594 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2595 components; fortuitously, this particular loophole had already been fixed
2596 by change 4.50/55 or 4.44/33 above.
2598 If there are any other similar loopholes, the new check in host_aton()
2599 itself should stop them being exploited. The report I received stated that
2600 data on the command line could provoke the exploit when Exim was running as
2601 exim, but did not say which command line option was involved. All I could
2602 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2603 running as the user.
2606 35. There was a buffer overflow vulnerability in the SPA authentication code
2607 (which came originally from the Samba project). I have added a test to the
2608 spa_base64_to_bits() function which I hope fixes it.
2611 36. The daemon start-up calls getloadavg() while still root for those OS that
2612 need the first call to be done as root, but it missed one case: when
2613 deliver_queue_load_max is set with deliver_drop_privilege. This is
2614 necessary for the benefit of the queue runner, because there is no re-exec
2615 when deliver_drop_privilege is set.
2617 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2618 was in use, was not putting the data itself into the right store pool;
2619 consequently, it could be overwritten for a subsequent message in the same
2620 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2623 38. Sometimes the final signoff response after QUIT could fail to get
2624 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2625 < 0 before doing a fflush(). This bug looks as though it goes back to the
2626 introduction of TLS in release 3.20, but "sometimes" must have been rare
2627 because the tests only now provoked it.
2633 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2634 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2635 Now it returns an integer. A number of calls were still expecting a T/F
2636 return. Fortuitously, in all cases, the tests worked in OK situations,
2637 which is the norm. However, things would have gone wrong on any write
2638 failures on the smtp file descriptor. This function is used when sending
2639 messages over SMTP and also when doing verify callouts.
2641 2. When Exim is called to do synchronous delivery of a locally submitted
2642 message (the -odf or -odi options), it no longer closes stderr before doing
2645 3. Implemented the mua_wrapper option.
2647 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2649 5. Implemented the functions header_remove(), header_testname(),
2650 header_add_at_position(), and receive_remove_recipient(), and exported them
2653 6. If an ACL "warn" statement specified the addition of headers, Exim already
2654 inserted X-ACL-Warn: at the start if there was no header name. However, it
2655 was not making this test for the second and subsequent header lines if
2656 there were newlines in the string. This meant that an invalid header could
2657 be inserted if Exim was badly configured.
2659 7. Allow an ACL "warn" statement to add header lines at the start or after all
2660 the Received: headers, as well as at the end.
2662 8. Added the rcpt_4xx retry error code.
2664 9. Added postmaster_mailfrom=xxx to callout verification option.
2666 10. Added mailfrom=xxxx to the callout verification option, for verify=
2669 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2670 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2672 12. Inserted some casts to stop certain compilers warning when using pointer
2673 differences as field lengths or precisions in printf-type calls (mostly
2674 affecting debugging statements).
2676 13. Added optional readline() support for -be (dynamically loaded).
2678 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2679 same clock tick as a message's arrival, so that its received time was the
2680 same as the "first fail" time on the retry record, and that message
2681 remained on the queue past the ultimate address timeout, every queue runner
2682 would try a delivery (because it was past the ultimate address timeout) but
2683 after another failure, the ultimate address timeout, which should have then
2684 bounced the address, did not kick in. This was a "< instead of <=" error;
2685 in most cases the first failure would have been in the next clock tick
2686 after the received time, and all would be well.
2688 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2689 being recognized when the domain list was tested by the match_domain
2690 condition in an expansion string.
2692 16. Added the ${str2b64: operator.
2694 17. Exim was always calling setrlimit() to set a large limit for the number of
2695 processes, without checking whether the existing limit was already
2696 adequate. (It did check for the limit on file descriptors.) Furthermore,
2697 errors from getrlimit() and setrlimit() were being ignored. Now they are
2698 logged to the main and panic logs, but Exim does carry on, to try to do its
2699 job under whatever limits there are.
2701 18. Imported PCRE 5.0.
2703 19. Trivial typo in log message " temporarily refused connection" (the leading
2706 20. If the log selector return_path_on_delivery was set and an address was
2707 redirected to /dev/null, the delivery process crashed because it assumed
2708 that a return path would always be set for a "successful" delivery. In this
2709 case, the whole delivery is bypassed as an optimization, and therefore no
2712 21. Internal re-arrangement: the function for sending a challenge and reading
2713 a response while authentication was assuming a zero-terminated challenge
2714 string. It's now changed to take a pointer and a length, to allow for
2715 binary data in such strings.
2717 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2719 23. Exim was not respecting finduser_retries when seeking the login of the
2720 uid under which it was called; it was always trying 10 times. (The default
2721 setting of finduser_retries is zero.) Also, it was sleeping after the final
2722 failure, which is pointless.
2724 24. Implemented tls_on_connect_ports.
2726 25. Implemented acl_smtp_predata.
2728 26. If the domain in control=submission is set empty, Exim assumes that the
2729 authenticated id is a complete email address when it generates From: or
2730 Sender: header lines.
2732 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2733 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2734 chown and chgrp in /bin and hostname in /usr/bin.
2736 28. Exim was keeping the "process log" file open after each use, just as it
2737 does for the main log. This opens the possibility of it remaining open for
2738 long periods when the USR1 signal hits a daemon. Occasional processlog
2739 errors were reported, that could have been caused by this. Anyway, it seems
2740 much more sensible not to leave this file open at all, so that is what now
2743 29. The long-running daemon process does not normally write to the log once it
2744 has entered its main loop, and it closes the log before doing so. This is
2745 so that log files can straightforwardly be renamed and moved. However,
2746 there are a couple of unusual error situations where the daemon does write
2747 log entries, and I had neglected to close the log afterwards.
2749 30. The text of an SMTP error response that was received during a remote
2750 delivery was being truncated at 512 bytes. This is too short for some of
2751 the long messages that one sometimes sees. I've increased the limit to
2754 31. It is now possible to make retry rules that apply only when a message has a
2755 specific sender, in particular, an empty sender.
2757 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2758 it possible to be selective about when SMTP synchronization is enforced.
2760 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2762 32. Implemented hosts_connection_nolog.
2764 33. Added an ACL for QUIT.
2766 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2769 35. Added mailbox_size and mailbox_filecount to appendfile.
2771 36. Added control = no_multiline_responses to ACLs.
2773 37. There was a bug in the logic of the code that waits for the clock to tick
2774 in the case where the clock went backwards by a substantial amount such
2775 that the microsecond fraction of "now" was more than the microsecond
2776 fraction of "then" (but the whole seconds number was less).
2778 38. Added support for the libradius Radius client library this is found on
2779 FreeBSD (previously only the radiusclient library was supported).
2785 1. When certain lookups returned multiple values in the form name=value, the
2786 quoting of the values was not always being done properly. Specifically:
2787 (a) If the value started with a double quote, but contained no whitespace,
2789 (b) If the value contained whitespace other than a space character (i.e.
2790 tabs or newlines or carriage returns) it was not quoted.
2791 This fix has been applied to the mysql and pgsql lookups by writing a
2792 separate quoting function and calling it from the lookup code. The fix
2793 should probably also be applied to nisplus, ibase and oracle lookups, but
2794 since I cannot test any of those, I have not disturbed their existing code.
2796 2. A hit in the callout cache for a specific address caused a log line with no
2797 reason for rejecting RCPT. Now it says "Previous (cached) callout
2798 verification failure".
2800 3. There was an off-by-one bug in the queryprogram router. An over-long
2801 return line was truncated at 256 instead of 255 characters, thereby
2802 overflowing its buffer with the terminating zero. As well as fixing this, I
2803 have increased the buffer size to 1024 (and made a note to document this).
2805 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2806 when Exim is waiting for an SMTP response from a remote server, Exim
2807 restarts its select() call on the socket, thereby resetting its timeout.
2808 This is not a problem when such interrupts are rare. Somebody set up a cron
2809 job to run exiwhat every 2 minutes, which is less than the normal select()
2810 timeout (5 or 10 minutes). This meant that the select() timeout never
2811 kicked in because it was always reset. I have fixed this by comparing the
2812 time when an interrupt arrives with the time at the start of the first call
2813 to select(). If more time than the timeout has elapsed, the interrupt is
2814 treated as a timeout.
2816 5. Some internal re-factoring in preparation for the addition of Sieve
2817 extensions (by MH). In particular, the "personal" test is moved to a
2818 separate function, and given an option for scanning Cc: and Bcc: (which is
2819 not set for Exim filters).
2821 6. When Exim created an email address using the login of the caller as the
2822 local part (e.g. when creating a From: or Sender: header line), it was not
2823 quoting the local part when it contained special characters such as @.
2825 7. Installed new OpenBSD configuration files.
2827 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2828 try to make them clearer.
2830 9. Callout options, other than the timeout value, were being ignored when
2831 verifying sender addresses in header lines. For example, when using
2833 verify = header_sender/callout=no_cache
2835 the cache was (incorrectly) being used.
2837 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2838 only the Cygwin environment.
2840 11. When return_path_on_delivery was set as a log selector, if different remote
2841 addresses in the same message used different return paths and parallel
2842 remote delivery occurred, the wrong values would sometimes be logged.
2843 (Whenever a remote delivery process finished, the return path value from
2844 the most recently started remote delivery process was logged.)
2846 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2847 lines when AUTH and/or TLS are in use. This is the "received protocol"
2848 field. Exim used to use "asmtp" for authenticated SMTP, without any
2849 indication (in the protocol name) for TLS use. Now it follows the RFC and
2850 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2851 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2852 names appear in log lines as well as in Received: header lines.
2854 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2855 extensions, and comparison tests, and to fix some bugs.
2857 14. Changes to the "personal" filter test:
2859 (1) The test was buggy in that it was just doing the equivalent of
2860 "contains" tests on header lines. For example, if a user's address was
2861 anne@some.where, the "personal" test would incorrectly be true for
2863 To: susanne@some.where
2865 This test is now done by extracting each address from the header in turn,
2866 and checking the entire address. Other tests that are part of "personal"
2867 are now done using regular expressions (for example, to check local parts
2868 of addresses in From: header lines).
2870 (2) The list of non-personal local parts in From: addresses has been
2871 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2872 taken from the Sieve specification recommendations.
2874 (3) If the message contains any header line starting with "List-" it is
2875 treated as non-personal.
2877 (4) The test for "circular" in the Subject: header line has been removed
2878 because it now seems ill-conceived.
2880 15. Minor typos in src/EDITME comments corrected.
2882 16. Installed latest exipick from John Jetmore.
2884 17. If headers_add on a router specified a text string that was too long for
2885 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2886 of string_sprintf() is now avoided.
2888 18. $message_body_size was not set (it was always zero) when running the DATA
2889 ACL and the local_scan() function.
2891 19. For the "mail" command in an Exim filter, no default was being set for
2892 the once_repeat time, causing a random time value to be used if "once" was
2893 specified. (If the value happened to be <= 0, no repeat happened.) The
2894 default is now 0s, meaning "never repeat". The "vacation" command was OK
2895 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2896 (I found it when inspecting the code).
2898 20. There is now an overall timeout for performing a callout verification. It
2899 defaults to 4 times the callout timeout, which applies to individual SMTP
2900 commands during the callout. The overall timeout applies when there is more
2901 than one host that can be tried. The timeout is checked before trying the
2902 next host. This prevents very long delays if there are a large number of
2903 hosts and all are timing out (e.g. when the network connections are timing
2904 out). The value of the overall timeout can be changed by specifying an
2905 additional sub-option for "callout", called "maxwait". For example:
2907 verify = sender/callout=5s,maxwait=20s
2909 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2910 to the end before writing, but this should make it even safer).
2912 22. Exim was forgetting that it had advertised PIPELINING for the second and
2913 subsequent messages on an SMTP connection. It was also not resetting its
2914 memory on STARTTLS and an internal HELO.
2916 23. When Exim logs an SMTP synchronization error within a session, it now
2917 records whether PIPELINING has been advertised or not.
2919 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2920 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2921 rather than long int.
2923 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2925 26. Added the never_mail option to autoreply.
2931 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2932 crash if the getsockname() call failed; this can happen if a connection is
2933 closed very soon after it is established. The problem was simply in the
2934 order in which certain operations were done, causing Exim to try to write
2935 to the SMTP stream before it had set up the file descriptor. The bug has
2936 been fixed by making things happen in the correct order.
2942 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2943 before the connection was closed, thus losing the rejection response.
2945 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2946 some early Solaris releases, but causes trouble in current releases where
2947 socklen_t is defined.
2949 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2952 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2955 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2956 that releasing the top of it at the end releases what was used for sub-
2957 expansions (unless the block got too big). However, discard this block if
2958 the first thing is a variable or header, so that we can use its block when
2959 it is dynamic (useful for very large $message_headers, for example).
2961 6. Lookups now cache *every* query, not just the most recent. A new, separate
2962 store pool is used for this. It can be recovered when all lookup caches are
2963 flushed. Lookups now release memory at the end of their result strings.
2964 This has involved some general refactoring of the lookup sources.
2966 7. Some code has been added to the store_xxx() functions to reduce the amount
2967 of flapping under certain conditions.
2969 8. log_incoming_interface used to affect only the <= reception log lines. Now
2970 it causes the local interface and port to be added to several more SMTP log
2971 lines, for example "SMTP connection from", and rejection lines.
2973 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2975 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2977 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2978 could be overwritten at the end of the current message (or the start of a
2979 new message if it was set in a HELO ACL). The value is now preserved for
2980 the duration of the SMTP connection.
2982 12. If a transport had a headers_rewrite setting, and a matching header line
2983 contained an unqualified address, that address was qualified, even if it
2984 did not match any rewriting rules. The underlying bug was that the values
2985 of the flags that permit the existence of unqualified sender and recipient
2986 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2987 non-local messages, and by -bnq for local messages) were not being
2988 preserved with the message after it was received.
2990 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2991 "next input=" as part of the text comprising the host identity instead of
2992 the correct text. The code was using the same buffer for two different
2993 strings. However, depending on which order the printing function evaluated
2994 its arguments, the bug did not always show up. Under Linux, for example, my
2995 test suite worked just fine.
2997 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2998 doesn't work with some older versions of Perl. It has been changed to "my",
2999 which in any case is probably the better facility to use.
3001 15. A really picky compiler found some instances of statements for creating
3002 error messages that either had too many or two few arguments for the format
3005 16. The size of the buffer for calls to the DNS resolver has been increased
3006 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3007 for addresses that have a lot of PTR records. This alleviates a problem; it
3008 does not fully solve it.
3010 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3011 buffer now truncates the list and logs the incident, which is the same
3012 action as happens when Exim is looking up a host name and its aliases.
3013 Previously in this situation something unpredictable would happen;
3014 sometimes it was "internal error: store_reset failed".
3016 18. If a server dropped the connection unexpectedly when an Exim client was
3017 using GnuTLS and trying to read a response, the client delivery process
3018 crashed while trying to generate an error log message.
3020 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3021 string, for example:
3023 warn message = H1: something\nH2: something
3025 the text was added as a single header line from Exim's point of view
3026 though it ended up OK in the delivered message. However, searching for the
3027 second and subsequent header lines using $h_h2: did not work. This has been
3028 fixed. Similarly, if a system filter added multiple headers in this way,
3029 the routers could not see them.
3031 20. Expanded the error message when iplsearch is called with an invalid key to
3032 suggest using net-iplsearch in a host list.
3034 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3035 ACLs are no longer actually imposed (and a message to that effect is
3038 22. If a "gecos" field in a passwd entry contained escaped characters, in
3039 particular, if it contained a \" sequence, Exim got it wrong when building
3040 a From: or a Sender: header from that name. A second bug also caused
3041 incorrect handling when an unquoted " was present following a character
3042 that needed quoting.
3044 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3045 was not being matched caselessly.
3047 24. Arranged for all hyphens in the exim.8 source to be escaped with
3050 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3051 properly. Recipient callout cache records were still being keyed to include
3052 the sender, even when use_sender was set false. This led to far more
3053 callouts that were necessary. The sender is no longer included in the key
3054 when use_sender is false.
3056 26. Added "control = submission" modifier to ACLs.
3058 27. Added the ${base62d: operator to decode base 62 numbers.
3060 28. dnsdb lookups can now access SRV records.
3062 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3063 the configuration file.
3065 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3066 (-v) mode. This makes the output for a verbose queue run more intelligible.
3068 31. Added a use_postmaster feature to recipient callouts.
3070 32. Added the $body_zerocount variable, containing the number of binary zero
3071 bytes in the message body.
3073 33. The time of last modification of the "new" subdirectory is now used as the
3074 "mailbox time last read" when there is a quota error for a maildir
3077 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3079 35. Added +ignore_unknown as a special item in host lists.
3081 36. Code for decoding IPv6 addresses in host lists is now included, even if
3082 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3083 address was recognized as an IP address, but was then not correctly decoded
3084 into binary, causing unexpected and incorrect effects when compared with
3091 1. Very minor rewording of debugging text in manualroute to say "list of
3092 hosts" instead of "hostlist".
3094 2. If verify=header_syntax was set, and a header line with an unqualified
3095 address (no domain) and a large number of spaces between the end of the
3096 name and the colon was received, the reception process suffered a buffer
3097 overflow, and (when I tested it) crashed. This was caused by some obsolete
3098 code that should have been removed. The fix is to remove it!
3100 3. When running in the test harness, delay a bit after writing a bounce
3101 message to get a bit more predictability in the log output.
3103 4. Added a call to search_tidyup() just before forking a reception process. In
3104 theory, someone could use a lookup in the expansion of smtp_accept_max_
3105 per_host which, without the tidyup, could leave open a database connection.
3107 5. Added the variables $recipient_data and $sender_data which get set from a
3108 lookup success in an ACL "recipients" or "senders" condition, or a router
3109 "senders" option, similar to $domain_data and $local_part_data.
3111 6. Moved the writing of debug_print from before to after the "senders" test
3114 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3115 problems for message scanning, either using a data ACL, or using
3116 local_scan() because the Received: header was not generated till after they
3117 were called (in order to set the time as the time of reception completion).
3118 I have revised the way this works. The header is now generated after the
3119 body is received, but before the ACL or local_scan() are called. After they
3120 are run, the timestamp in the header is updated.
3126 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3127 before starting a queue runner without re-exec. This happened only when
3128 deliver_drop_privilege was set or when the Exim user was set to root. The
3129 effect of the bug was that timeouts during subsequent deliveries caused
3130 crashes instead of being properly handled. The handler is now left at its
3131 default (and expected) setting.
3133 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3134 message, again when deliver_drop_privilege is set or Exim is run as root.
3135 The bug described in (1) was not present in this case, but the tidying up
3136 of the other signals was missing. I have made the two cases consistent.
3138 3. The ignore_target_hosts setting on a manualroute router was being ignored
3139 for hosts that were looked up using the /MX notation.
3141 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3144 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3145 operated on the sender address. After changing the $sender_address to <>
3146 for the sender address verify, Exim was re-instated it as the original
3147 (before rewriting) address, but remembering that it had rewritten it, so it
3148 wasn't rewriting it again. This bug also had the effect of breaking the
3149 sender address verification caching when the sender address was rewritten.
3151 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3152 This has been changed so that if the ip literal address matches
3153 ignore_target_hosts, the router declines.
3155 7. Added expansion conditions match_domain, match_address, and match_local_
3156 part (NOT match_host).
3158 8. The placeholder for the Received: header didn't have a length field set.
3160 9. Added code to Exim itself and to exim_lock to test for a specific race
3161 condition that could lead to file corruption when using MBX delivery. The
3162 issue is with the lockfile that is created in /tmp. If this file is removed
3163 after a process has opened it but before that process has acquired a lock,
3164 there is the potential for a second process to recreate the file and also
3165 acquire a lock. This could lead to two Exim processes writing to the file
3166 at the same time. The added code performs the same test as UW imapd; it
3167 checks after acquiring the lock that its file descriptor still refers to
3168 the same named file.
3170 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3171 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3174 11. Added the smtp_active_hostname option. If used, this will typically be made
3175 to depend on the incoming interface address. Because $interface_address is
3176 not set up until the daemon has forked a reception process, error responses
3177 that can happen earlier (such as "too many connections") no longer contain
3180 12. If an expansion in a condition on a "warn" statement fails because a lookup
3181 defers, the "warn" statement is abandoned, and the next ACL statement is
3182 processed. Previously this caused the whole ACL to be aborted.
3184 13. Added the iplsearch lookup type.
3186 14. Added ident_timeout as a log selector.
3188 15. Added tls_certificate_verified as a log selector.
3190 16. Added a global option tls_require_ciphers (compare the smtp transport
3191 option of the same name). This controls incoming TLS connections.
3193 17. I finally figured out how to make tls_require_ciphers do a similar thing
3194 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3195 before starting the TLS session.
3197 18. Tabs are now shown as \t in -bP output.
3199 19. If the log selector return_path_on_delivery was set, Exim crashed when
3200 bouncing a message because it had too many Received: header lines.
3202 20. If two routers both had headers_remove settings, and the first one included
3203 a superfluous trailing colon, the final name in the first list and the
3204 first name in the second list were incorrectly joined into one item (with a
3205 colon in the middle).
3211 1. Added -C and -D options to the exinext utility, mainly to make it easier
3212 to include in the automated testing, but these could be helpful when
3213 multiple configurations are in use.
3215 2. The exinext utility was not formatting the output nicely when there was
3216 an alternate port involved in the retry record key, nor when there was a
3217 message id as well (for retries that were specific to a specific message
3218 and a specific host). It was also confused by IPv6 addresses, because of
3219 the additional colons they contain. I have fixed the IPv4 problem, and
3220 patched it up to do a reasonable job for IPv6.
3222 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3223 delivery, the log line now contains "pipelined" if PIPELINING was used.
3225 4. An SMTP transport process used to panic and die if the bind() call to set
3226 an explicit outgoing interface failed. This has been changed; it is now
3227 treated in the same way as a connect() failure.
3229 5. A reference to $sender_host_name in the part of a conditional expansion
3230 that was being skipped was still causing a DNS lookup. This no longer
3233 6. The def: expansion condition was not recognizing references to header lines
3234 that used bh_ and bheader_.
3236 7. Added the _cache feature to named lists.
3238 8. The code for checking quota_filecount in the appendfile transport was
3239 allowing one more file than it should have been.
3241 9. For compatibility with Sendmail, the command line option
3249 and sets the incoming protocol and host name (for trusted callers). The
3250 host name and its colon can be omitted when only the protocol is to be set.
3251 Note the Exim already has two private options, -pd and -ps, that refer to
3252 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3253 "s", but I don't think that's a major issue.
3255 10. A number of refactoring changes to the code, none of which should affect
3258 (a) The number of logging options was getting close to filling up the
3259 32-bit word that was used as a bit map. I have split them into two classes:
3260 those that are passed in the argument to log_write(), and those that are
3261 only ever tested independently outside of that function. These are now in
3262 separate 32-bit words, so there is plenty of room for expansion again.
3263 There is no change in the user interface or the logging behaviour.
3265 (b) When building, for example, log lines, the code previously used a
3266 macro that called string_cat() twice, in order to add two strings. This is
3267 not really sufficiently general. Furthermore, there was one instance where
3268 it was actually wrong because one of the argument was used twice, and in
3269 one call a function was used. (As it happened, calling the function twice
3270 did not affect the overall behaviour.) The macro has been replaced by a
3271 function that can join an arbitrary number of extra strings onto a growing
3274 (c) The code for expansion conditions now uses a table and a binary chop
3275 instead of a serial search (which was left over from when there were very
3276 few conditions). Also, it now recognizes conditions like "pam" even when
3277 the relevant support is not compiled in: a suitably worded error message is
3278 given if an attempt is made to use such a condition.
3280 11. Added ${time_interval:xxxxx}.
3282 12. A bug was causing one of the ddress fields not to be passed back correctly
3283 from remote delivery subprocesses. The field in question was not being
3284 subsequently used, so this caused to problems in practice.
3286 13. Added new log selectors queue_time and deliver_time.
3288 14. Might have fixed a bug in maildirsizefile handling that threw up
3289 "unexpected character" debug warnings, and recalculated the data
3290 unnecessarily. In any case, I expanded the warning message to give more
3293 15. Added the message "Restricted characters in address" to the statements in
3294 the default ACL that block characters like @ and % in local parts.
3296 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3297 Three changes have been made:
3299 (a) There was a serious bug; a negative response to MAIL caused the whole
3300 recipient domain to be cached as invalid, thereby blocking all messages
3301 to all local parts at the same domain, from all senders. This bug has
3302 been fixed. The domain is no longer cached after a negative response to
3303 MAIL if the sender used is not empty.
3305 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3308 (c) A new callout option, "use_sender" has been added for people who want
3309 the modified behaviour.
3315 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3318 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3319 indeed breaks things for older releases.
3321 3. Added additional logging to the case where there is a problem reading data
3322 from a filter that is running in a subprocess using a pipe, in order to
3323 try to track down a specific problem.
3325 4. Testing facility fudge: when running in the test harness and attempting
3326 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3327 getting "No route to host". Convert this to a timeout.
3329 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3332 6. Some OS don't have socklen_t but use size_t instead. This affects the
3333 fifth argument of getsockopt() amongst other things. This is now
3334 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3335 can be set for individual OS. I have set it for SunOS5, OSF1, and
3336 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3337 some earlier ones do not.
3339 7. Change 4.30/15 was not doing the test caselessly.
3341 8. The standard form for an IPv6 address literal was being rejected by address
3342 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3343 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3344 this, as well as the form without the "IPv6" on the front (but only when
3345 address literals are enabled, of course).
3347 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3349 10. Exim crashed if a message with an empty sender address specified by -f
3350 encountered a router with an errors_to setting. This could be provoked only
3351 by a command such as
3355 where an empty string was supplied; "<>" did not hit this bug.
3357 11. Installed PCRE release 4.5.
3359 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3360 remained set. It is now erased.
3362 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3363 times from message ids (which are base 36 rather than the normal 62).
3365 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3366 were being counted as actual protocol errors, and logged if the log
3367 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3368 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3369 and DATA following a set of rejected RCPTs do not count as protocol errors.
3370 In other words, Exim assumes they were pipelined, though this may not
3371 actually be the case. Of course, in all cases the client gets an
3372 appropriate error code.
3374 15. If a lookup fails in an ACL condition, a message about the failure may
3375 be available; it is used if testing the ACL cannot continue, because most
3376 such messages specify what the cause of the deferral is. However, some
3377 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3378 that caused an old message to be retained and used if a later statement
3379 caused a defer, replacing the real cause of the deferral.
3381 16. If an IP address had so many PTR records that the DNS lookup buffer
3382 was not large enough to hold them, Exim could crash while trying to process
3383 the truncated data. It now detects and logs this case.
3385 17. Further to 4.21/58, another change has been made: if (and only if) the
3386 first line of a message (the first header line) ends with CRLF, a bare LF
3387 in a subsequent header line has a space inserted after it, so as not to
3388 terminate the header.
3390 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3391 unnecessarily, used atoi() instead of strtol(), and didn't check the
3392 termination when getting file sizes from file names by regex.
3394 19. Completely re-implemented the support for maildirsize files, in the light
3395 of a number of problems with the previous contributed implementation
3396 (4.30/29). In particular:
3398 . If the quota is zero, the maildirsize file is maintained, but no quota is
3401 . If the maildir directory does not exist, it is created before any attempt
3402 to write a maildirsize file.
3404 . The quota value in the file is just a cache; if the quota is changed in
3405 the transport, the new value overrides.
3407 . A regular expression is available for excluding directories from the
3410 20. The autoreply transport checks the characters in options that define the
3411 message's headers; it allows continued headers, but it was checking with
3412 isspace() after an embedded newline instead of explicitly looking for a
3415 21. If all the "regular" hosts to which an address was routed had passed their
3416 expiry times, and had not reached their retry times, the address was
3417 bounced, even if fallback hosts were defined. Now Exim should go on to try
3420 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3421 equivalent code in the SMTP transport. Some hosts send humungous responses
3422 to HELO/EHLO, more than 1024 it seems.
3424 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3425 gives compiler warnings in some environments. I've now done it "properly",
3428 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3429 (because of problems with the built-in one) was declared to return uschar *
3430 instead of char *, causing compiler failure.
3432 25. Fixed a file descriptor leak when processing alias/forward files.
3434 26. Fixed a minor format string issue in dbfn.c.
3436 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3438 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3439 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3442 29. When a sender address is verified, it is cached, to save repeating the test
3443 when there is more than one recipient in a message. However, when the
3444 verification involves a callout, it is possible for different callout
3445 options to be set for different recipients. It is too complicated to keep
3446 track of this in the cache, so now Exim always runs a verification when a
3447 callout is required, relying on the callout cache for the optimization.
3448 The overhead is duplication of the address routing, but this should not be
3451 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3452 to be verified with callout=postmaster, and the main callout worked but the
3453 postmaster check failed, the verification correctly failed. However, if a
3454 subsequent RCPT command asked for sender verification *without* the
3455 postmaster check, incorrect caching caused this verification also to fail,
3458 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3459 it was not caching the DNS options (qualify_single, search_parents) that
3460 were used when the lookup failed. A subsequent lookup with different
3461 options therefore always gave the same answer, though there were cases
3462 where it should not have. (Example: a "domains = !$mx_any" option on a
3463 dnslookup router: the "domains" option is always processed without any
3464 widening, but the router might have qualify_single set.) Now Exim uses the
3465 cached value only when the same options are set.
3467 32. Added John Jetmore's "exipick" utility to the distribution.
3469 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3470 than a timeout (e.g. a certificate is required, and is not provided), an
3471 Exim server now closes the connection immediately. Previously it waited for
3472 the client to close - but if the client is SSL, it seems that they each
3473 wait for each other, leading to a delay before one of them times out.
3475 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3476 maintained 0.8.x compatibility because I don't think many are using it, and
3477 it is clearly obsolete.
3479 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3482 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3483 client certificate was expired. A simple patch fixes this, though I don't
3484 understand the full logic of why the verify callback is called multiple
3487 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3488 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3489 which causes problems with some clients (such as the Certicom SSL Plus
3490 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3491 disables the coutermeasure allowing Eudora to connect."
3493 38. Exim was not checking that a write() to a log file succeeded. This could
3494 lead to Bad Things if a log got too big, in particular if it hit a file
3495 size limit. Exim now panics and dies if it cannot write to a log file, just
3496 as it does if it cannot open a log file.
3498 39. Modified OS/Makefile-Linux so that it now contains
3500 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3502 The two -D definitions ensure that Exim is compiled with large file
3503 support, which makes it possible to handle log files that are bigger than
3506 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3507 instance) a domain was checked against a named list that involved a lookup,
3508 causing $domain_data to be set, then another domain was checked against the
3509 same list, then the first domain was re-checked, the value of $domain_data
3510 after the final check could be wrong. In particular, if the second check
3511 failed, it could be set empty. This bug probably also applied to
3514 41. The strip_trailing_dot option was not being applied to the address given
3515 with the -f command-line option.
3517 42. The code for reading a message's header from the spool was incrementing
3518 $received_count, but never initializing it. This meant that the value was
3519 incorrect (doubled) while delivering a message in the same process in which
3520 it was received. In the most common configuration of Exim, this never
3521 happens - a fresh exec is done - but it can happen when
3522 deliver_drop_privilege is set.
3524 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3525 it now includes up to 150 characters of the unexpected data in the log
3528 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3529 and building data strings. The size of both of these buffers was 10 000
3530 bytes - far larger than anybody would *ever* want, thought I. Needless to
3531 say, somebody hit the limit. I have increased the maximum line length to
3532 20 000 and the maximum data length of concatenated lines to 100 000. I have
3533 also fixed two bugs, because there was no checking on these buffers. Tsk,
3534 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3535 buffer is too small.
3537 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3538 lsearch lookups. Now it does.
3540 46. When parsing a route_list item in a manualroute router, a fixed-length
3541 buffer was used for the list of hosts. I made this 1024 bytes long,
3542 thinking that nobody would ever have a list of hosts that long. Wrong.
3543 Somebody had a whole pile of complicated expansion conditions, and the
3544 string was silently truncated, leading to an expansion error. It turns out
3545 that it is easier to change to an unlimited length (owing to other changes
3546 that have happened since this code was originally written) than to build
3547 structure for giving a limitation error. The length of the item that
3548 expands into the list of hosts is now unlimited.
3550 47. The lsearch lookup could not handle data where the length of text line was
3551 more than 4095 characters. Such lines were truncated, leading to shortened
3552 data being returned. It should now handle lines of any length.
3554 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3555 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3558 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3559 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3560 no longer happens in comment lines. A list of replacements is now placed
3561 at the head of all of the source files, except those whose only change is
3562 to replace PERL_COMMAND in the very first #! line.
3564 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3565 messages on the queue, with a bottom-up merge sort, using code contributed
3566 by Michael Haardt. This should make operations like -bp somewhat faster on
3567 large queues. It won't affect queue runners, except when queue_run_in_order
3570 51. Installed eximstats 1.31 in the distribution.
3572 52. Added support for SRV lookups to the dnslookup router.
3574 53. If an ACL referred to $message_body or $message_body_end, the value was not
3575 reset for any messages that followed in the same SMTP session.
3577 54. The store-handling optimization for building very long strings was not
3578 differentiating between the different store pools. I don't think this
3579 actually made any difference in practice, but I've tidied it.
3581 55. While running the routers to verify a sender address, $sender_address
3582 was still set to the sender address. This is wrong, because when routing to
3583 send a bounce to the sender, it would be empty. Therefore, I have changed
3584 it so that, while verifying a sender address, $sender_address is set to <>.
3585 (There is no change to what happens when verifying a recipient address.)
3587 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3588 target A or AAAA records (if not already returned) without resetting the
3589 qualify_single or search_parents options of the DNS resolver. These are
3590 inappropriate in this case because the targets of MX and SRV records must
3591 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3592 target that, when qualified, matched something in the local domain. These
3593 two options are now turned off when doing these lookups.
3595 57. It seems that at least some releases of Reiserfs (which does not have the
3596 concept of a fixed number of inodes) returns zero and not -1 for the
3597 number of available inodes. This interacted badly with check_spool_inodes,
3598 which assumed that -1 was the "no such thing" setting. What I have done is
3599 to check that the total number of inodes is greater than zero before doing
3600 the test of how many are available.
3602 58. When a "warn" ACL statement has a log_message modifier, the message is
3603 remembered, and not repeated. This is to avoid a lot of repetition when a
3604 message has many recipients that cause the same warning to be written.
3605 Howewer, Exim was preserving the list of already written lines for an
3606 entire SMTP session, which doesn't seem right. The memory is now reset if a
3607 new message is started.
3609 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3610 debugging output unless log_rewrite was also set.
3612 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3613 of (int)(handle) when we know that handle contains (void *)(-1).
3615 61. The Exim daemon panic-logs an error return when it closes the incoming
3616 connection. However "connection reset by peer" seems to be common, and
3617 isn't really an error worthy of noting specially, so that particular error
3620 62. When Exim is trying to find all the local interfaces, it used to panic and
3621 die if the ioctl to get the interface flags failed. However, it seems that
3622 on at least one OS (Solaris 9) it is possible to have an interface that is
3623 included in the list of interfaces, but for which you get a failure error
3624 for this call. This happens when the interface is not "plumbed" into a
3625 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3626 failure of the "get flags" call assumes that the interface is down.
3628 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3629 makes life easier for people who are doing arithmetic on fields extracted
3630 from dates, where you often get leading zeros that should not be
3631 interpreted as octal.
3633 64. Added qualify_domain to the redirect router, to override the global
3636 65. If a pathologically long header line contained very many addresses (the
3637 report of this problem mentioned 10 000) and each of them was rewritten,
3638 Exim could use up a very large amount of memory. (It kept on making new
3639 copies of the header line as it rewrote, and never released the old ones.)
3640 At the expense of a bit more processing, the header rewriting function has
3641 been changed so that it no longer eats memory in this way.
3643 66. The generation of the Received: header has been moved from the time that a
3644 message starts to be received, to the time that it finishes. The timestamp
3645 in the Received: header should now be very close to that of the <= log
3646 line. There are two side-effects of this change:
3648 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3649 logged header lines no longer include the local Received: line, because
3650 it has not yet been created. The same applies to a copy of the message
3651 that is returned to a non-SMTP sender when a message is rejected.
3653 (b) When a filter file is tested using -bf, no additional Received: header
3654 is added to the test message. After some thought, I decided that this
3657 This change does not affect the value of $received_for. It is still set
3658 after address rewriting, but before local_scan() is called.
3660 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3662 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3663 gave an unhelpful panic error message, and a defer error. I have managed to
3664 change this behaviour so that it now rejects any supplied certificate,
3665 which seems right, as the list of acceptable certificates is empty.
3667 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3668 gave an unhelpful defer error. I have not managed to make this reject any
3669 supplied certificates, but the error message it gives is "no certificate
3670 supplied", which is not helpful.
3672 70. exigrep's output now also includes lines that are not associated with any
3673 message, but which match the given pattern. Implemented by a patch from
3674 Martin Sluka, which also tidied up the Perl a bit.
3676 71. Recipient callout verification, like sender verification, was using <> in
3677 the MAIL FROM command. This isn't really the right thing, since the actual
3678 sender may affect whether the remote host accepts the recipient or not. I
3679 have changed it to use the actual sender in the callout; this means that
3680 the cache record is now keyed on a recipient/sender pair, not just the
3681 recipient address. There doesn't seem to be a real danger of callout loops,
3682 since a callout by the remote host to check the sender would use <>.
3683 [SEE ABOVE: changed after hitting problems.]
3685 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3686 temporary errors. However, in the case of such a code being given after
3687 the end of a data transmission (i.e. after ".") Exim was failing to write
3688 a retry record for the message. (Yes, there was some broken host that was
3689 actually sending 8xx at this point.)
3691 73. An unknown lookup type in a host list could cause Exim to panic-die when
3692 the list was checked. (An example that provoked this was putting <; in the
3693 middle of a list instead of at the start.) If this happened during a DATA
3694 ACL check, a -D file could be left lying around. This kind of configuration
3695 error no longer causes Exim to die; instead it causes a defer errror. The
3696 incident is still logged to the main and panic logs.
3698 74. Buglet left over from Exim 3 conversion. The message "too many messages
3699 in one connection" was written to the rejectlog but not the mainlog, except
3700 when address rewriting (yes!) was being logged.
3702 75. Added write_rejectlog option.
3704 76. When a system filter was run not as root (that is, when system_filter_user
3705 was set), the values of the $n variables were not being returned to the
3706 main process; thus, they were not subsequently available in the $sn
3709 77. Added +return_path_on_delivery log selector.
3711 78. A connection timeout was being treated differently from recipients deferred
3712 when testing hosts_max_try with a message that was older than the host's
3713 retry timeout. (The host should not be counted, thus allowing all hosts to
3714 be tried at least once before bouncing.) This may have been the cause of an
3715 occasionally reported bug whereby a message would remain on the queue
3716 longer than the retry timeout, but would be bounced if a delivery was
3717 forced. I say "may" because I never totally pinned down the problem;
3718 setting up timeout/retry tests is difficult. See also the next item.
3720 79. The ultimate address timeout was not being applied to errors that involved
3721 a combination of host plus message (for example, a timeout on a MAIL
3722 command). When an address resolved to a number of possible hosts, and they
3723 were not all tried for each delivery (e.g. because of hosts_max_try), a
3724 message could remain on the queue longer than the retry timeout.
3726 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3729 81. Fixed an obscure SMTP outgoing bug which required at least the following
3730 conditions: (a) there was another message waiting for the same server;
3731 (b) the server returned 5xx to all RCPT commands in the first message so
3732 that the message was not completed; (c) the server dropped the connection
3733 or gave a negative response to the RSET that Exim sends to abort the
3734 transaction. The observed case was a dropped connection after DATA that had
3735 been sent in pipelining mode. That is, the server had advertised PIPELINING
3736 but was not implementing it correctly. The effect of the bug was incorrect
3737 behaviour, such as trying another host, and this could lead to a crash.
3743 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3744 and daemon.c were passed as pointers to ints; they should have been
3745 pointers to socklen_t variables (which are typically unsigned ints).
3747 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3750 3. Fixed a really odd bug that affected only the testing scheme; patching a
3751 certain fixed string in the binary changed the value of another string that
3752 happened to be identical to the end of the original first string.
3754 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3755 name", it returns that address as the IP address. On some operating
3756 systems (e.g. Solaris), it also passes back the IP address string as the
3757 "host name". However, on others (e.g. Linux), it passes back an empty
3758 string. Exim wasn't checking for this, and was changing the host name to an
3759 empty string, assuming it had been canonicized.
3761 5. Although rare, it is permitted to have more than one PTR record for a given
3762 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3763 all the names associated with an address, because they do in Solaris.
3764 However, it seems that they do not in Linux for data that comes from the
3765 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3766 I found this out when I moved to a new Linux workstation and tried to run
3767 the Exim test suite.
3769 To get round this problem I have changed the code so that it now does its
3770 own call to the DNS to look up PTR records when searching for a host name.
3771 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3772 addresses that are only in /etc/hosts are still found.
3774 This behaviour is, however, controlled by an option called host_lookup_
3775 order, which defaults to "bydns:byaddr". If people want to use the other
3776 order, or indeed, just use one or the other means of lookup, they can
3777 specify it in this variable.
3779 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3780 some operating systems, this comes back from gethostbyaddr() as an empty
3781 string, and this is what Exim used to test for. However, it seems that in
3782 other systems, "." is yielded. Exim now tests for this case too.
3784 7. The values of check_spool_space and check_log_space are now held internally
3785 as a number of kilobytes instead of an absolute number of bytes. If a
3786 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3787 kilobyte. This means that much larger values can be stored.
3789 8. Exim monitor: an attempt to get the action menu when not actually pointing
3790 at a message produces an empty menu entitled "No message selected". This
3791 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3792 no entries in it ("Shell widget menu has zero width and/or height"). So I
3793 have added a single, blank menu entry in this case.
3795 9. Added ${quote_local_part.
3797 10. MIME decoding is now applied to the contents of Subject: header lines when
3800 11. Now that a reference to $sender_host_address automatically causes a reverse
3801 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3802 host lookup before query-style lookups in lists that might use this
3803 variable. This has therefore been abolished, and the "net-" prefix is no
3804 longer necessary for query-style lookups.
3806 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3807 have been a typo for LFLAGS, so it has been changed.
3809 13. The install script calls Exim with "-C /dev/null" in order to find the
3810 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3811 to be output. Howeve, since Exim outputs its version number before the
3812 error, it didn't break the script. It just looked ugly. I fixed this by
3813 always allowing "-C /dev/null" if the caller is root.
3815 14. Ignore overlarge ACL variable number when reading spool file - insurance
3816 against a later release with more variables having written the file.
3818 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3819 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3820 this, as well as the form without the "IPv6" on the front.
3822 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3823 OS/Makefile-Darwin file.
3825 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3826 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3828 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3829 variable. However, some OS use ints for inodes. I've added cast to long int
3830 to get rid of the compiler warning.
3832 19. I had forgotten to lock out "/../" in configuration file names when
3833 ALT_CONFIG_PREFIX was set.
3835 20. Routers used for verification do not need to specify transports. However,
3836 if such a router generated a host list, and callout was configured, Exim
3837 crashed, because it could not find a port number from the (non-existent)
3838 transport. It now assumes port 25 in this circumstance.
3840 21. Added the -t option to exigrep.
3842 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3843 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3844 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3845 them is set, LOOKUP_LSEARCH is forced.
3847 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3849 24. Added sender and host information to the "rejected by local_scan()" log
3850 line; previously there was no indication of these.
3852 25. Added .include_if_exists.
3854 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3855 Linux. It turns out that not all file systems support this. Apparently some
3856 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3857 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3858 not supported on the file descriptor, is now ignored when Exim is trying to
3859 sync a directory. This applies only to Linux.
3861 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3863 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3864 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3867 29. Added support for maildirsize files from supplied patch (modified a bit).
3869 30. The use of :fail: followed by an empty string could lead Exim to respond to
3870 sender verification failures with (e.g.):
3872 550 Verification failed for <xxx>
3873 550 Sender verify failed
3875 where the first response line was missing the '-' that indicates it is not
3876 the final line of the response.
3878 31. The loop for finding the name of the user that called Exim had a hardwired
3879 limit of 10; it now uses the value of finduser_retries, which is used for
3880 all other user lookups.
3882 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3885 33. Exim was neglecting to zero errno before one call of strtol() when
3886 expanding a string and expecting an integer value. On some systems this
3887 resulted in spurious "integer overflow" errors. Also, it was casting the
3888 result into an int without checking.
3890 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3891 did not work. The code looks as if it has *never* worked, though it appears
3892 to have been documented since at least releast 1.62. I have made it work.
3894 35. The "timeout_DNS" error in retry rules, also documented since at least
3895 1.62, also never worked. As it isn't clear exactly what this means, and
3896 clearly it isn't a major issue, I have abolished the feature by treating it
3897 as "timeout", and writing a warning to the main and panic logs.
3899 36. The display of retry rules for -brt wasn't always showing the error code
3902 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3903 timeout_connect_A, timeout_connect_MX.
3905 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3906 to the empty sender.
3908 39. The daemon was not analysing the content of -oX till after it had closed
3909 stderr and disconnected from the controlling terminal. This meant that any
3910 syntax errors were only noted on the panic log, and the return code from
3911 the command was 0. By re-arranging the code a little, I've made the
3912 decoding happen first, so such errors now appear on stderr, and the return
3913 code is 1. However, the actual setting up of the sockets still happens in
3914 the disconnected process, so errors there are still only recorded on the
3917 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3918 connections (as happens on some IP stacks) was logged at start up time as
3919 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3920 it from "IPv6 and IPv4", which means that two separate sockets are being
3923 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3924 says whether AF_INET or AF_INET6 was passed as an argument.
3926 42. Exiwhat output was messed up when time zones were included in log
3929 43. Exiwhat now gives more information about the daemon's listening ports,
3930 and whether -tls-on-connect was used.
3932 44. The "port" option of the smtp transport is now expanded.
3934 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3935 silently ignored. Now an error message is written to the main and panic
3938 46. There's a new ACL modifier called "logwrite" which writes to a log file
3939 as soon as it is encountered.
3941 47. Added $local_user_uid and $local_user_gid at routing time.
3943 48. Exim crashed when trying to verify a sender address that was being
3946 49. Exim was recognizing only a space character after ".include". It now also
3947 recognizes a tab character.
3949 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3950 extracting the relevant information from the specification. The man page no
3951 longer contains scrambled data for the -d option, and I've added a section
3952 at the front about calling Exim under different names.
3954 51. Added "extra_headers" argument to the "mail" command in filter files.
3956 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3959 53. Installed eximstats 1.29.
3961 54. Added transport_filter_timeout as a generic transport option.
3963 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3964 Cc: header lines. This was required by RFC 822, but it not required by RFC
3967 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3968 incoming messages that did not have them. Now it does so only if the
3969 message originates locally, that is, if there is no associated remote host
3970 address. When Resent- header lines are present, this applies to the Resent-
3971 lines rather than the non-Resent- lines.
3973 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3974 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3976 58. Messages for configuration errors now include the name of the main
3977 configuration file - useful now that there may be more than one file in a
3978 list (.included file names were always shown).
3980 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3981 for those rare installations that do not start the daemon as root or run it
3982 setuid root. I've cut out the call to initgroups() if the daemon is not
3985 60. The Exim user and group can now be bound into the binary as text strings
3986 that are looked up at the start of Exim's processing.
3988 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3990 62. Added $mailstore_basename variable.
3992 63. Installed patch to sieve.c from Michael Haardt.
3994 64. When Exim failed to open the panic log after failing to open the main log,
3995 the original message it was trying to log was written to stderr and debug
3996 output, but if they were not available (the usual case in production), it
3997 was lost. Now it is written to syslog before the two lines that record the
3998 failures to open the logs.
4000 65. Users' Exim filters run in subprocesses under the user's uid. It is
4001 possible for a "deliver" command or an alias in a "personal" command to
4002 provoke an address rewrite. If logging of address rewriting is configured,
4003 this fails because the process is not running as root or exim. There may be
4004 a better way of dealing with this, but for the moment (because 4.30 needs
4005 to be released), I have disabled address rewrite logging when running a
4006 filter in a non-root, non-exim process.
4012 1. The buildconfig auxiliary program wasn't quoting the value set for
4013 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4014 not defined. This bug was masked in 4.22 by the effect that was fixed in
4017 2. Some messages that were rejected after a message id was allocated were
4018 shown as "incomplete" by exigrep. It no longer does this for messages that
4019 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4021 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4022 have allow_domain_literals set, the ID did not get logged in the <= line.
4023 Domain literals are now always recognized in Message-ID: header lines.
4025 4. The first argument for a ${extract expansion item is the key name or field
4026 number. Leading and trailing spaces in this item were not being ignored,
4027 causing some misleading effects.
4029 5. When deliver_drop_privilege was set, single queue runner processes started
4030 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4031 same command in the process it spins off) were not dropping privilege.
4033 6. When the daemon running as "exim" started a queue runner, it always
4034 re-executed Exim in the spun-off process. This is a waste of effort when
4035 deliver_drop_privilege is set. The new process now just calls the
4036 queue-runner function directly.
4042 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4045 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4046 ignored. Though the use of -oP was forcing the writing of a pid file, it
4047 was always written to the default place.
4049 3. If the message "no IP address found for host xxxx" is generated during
4050 incoming verification, it is now followed by identification of the incoming
4051 connection (so you can more easily find what provoked it).
4053 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4055 5. Added some features to "harden" Exim a bit more against certain attacks:
4057 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4058 be put in Local/Makefile. This is like the never_users runtime option,
4059 but it cannot be overridden. The default setting is "root".
4061 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4062 prefix string with which any file named in a -C command line option
4065 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4066 is retained for -C and -D only if the caller of Exim is root. Without
4067 it, the exim user may also use -C and -D and retain privilege.
4069 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4070 command line option is disabled.
4072 6. Macro names set by the -D option must start with an upper case letter, just
4073 like macro names defined in the configuration file.
4075 7. Added "dereference=" facility to LDAP.
4077 8. Two instances of the typo "uknown" in the source files are fixed.
4079 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4080 the Configure-Makefile script screwed up while processing it.
4082 10. Incorporated PCRE 4.4.
4084 11. The SMTP synchronization check was not operating right at the start of an
4085 SMTP session. For example, it could not catch a HELO sent before the client
4086 waited for the greeting. There is now a check for outstanding input at the
4087 point when the greeting is written. Because of the duplex, asynchronous
4088 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4089 way, but not yet received, when the check is performed.
4091 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4092 on TCP/IP sockets, because this apparently causes some broken clients to
4095 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4096 unchanged) from the Cygwin maintainer.
4098 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4099 was supported instead of testing for mbx. Effectively a typo.
4101 15. The spa authenticator server code was not checking that the input it
4102 received was valid base64.
4104 16. The debug output line for the "set" modifier in ACLs was not showing the
4105 name of the variable that was being set.
4107 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4109 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4110 lookup on its own. Something else was needed to trigger the lookup. For
4111 example, a match in host_lookup or the need for a host name in a host list.
4112 Now, if $sender_host_name is referenced and the host name has not yet been
4113 looked up, a lookup is performed. If the lookup fails, the variable remains
4114 empty, and $host_lookup_failed is set to "1".
4116 19. Added "eqi" as a case-independent comparison operator.
4118 20. The saslauthd authentication condition could segfault if neither service
4119 nor realm was specified.
4121 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4122 error message that was logged was misleading, and incoming SMTP
4123 connections were dropped. The message is now more accurate, and temporary
4124 errors are given to SMTP connections.
4126 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4127 (except RSET) with a 421 error, until QUIT is received. However, it was
4128 failing to send a response to QUIT.
4130 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4131 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4132 ACL is now run after the helo_try_verify_hosts code.
4134 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4135 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4136 case-independent, so other case variants are also recognized). Apparently
4137 some systems use these upper case variants.
4139 25. If more than two messages were waiting for the same host, and a transport
4140 filter was specified for the transport, Exim sent two messages over the
4141 same TCP/IP connection, and then failed with "socket operation on non-
4142 socket" when it tried to send the third.
4144 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4146 27. The extern definition of crypt16() in expand.c was not being excluded when
4147 the OS had its own crypt16() function.
4149 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4150 as a preferred synonym for return_size_limit, both as an option and as an
4153 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4155 30. Changed the default configuration ACL to relax the local part checking rule
4156 for addresses that are not in any local domains. For these addresses,
4157 slashes and pipe symbols are allowed within local parts, but the sequence
4158 /../ is explicitly forbidden.
4160 31. SPA server authentication was not clearing the challenge buffer before
4163 32. log_message in a "warn" ACL statement was writing to the reject log as
4164 well as to the main log, which contradicts the documentation and doesn't
4165 seem right (because no rejection is happening). So I have stopped it.
4167 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4168 However, I am unable to do any testing of this.
4170 34. Fixed an infelicity in the appendfile transport. When checking directories
4171 for a mailbox, to see if any needed to be created, it was accidentally
4172 using path names with one or more superfluous leading slashes; tracing
4173 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4175 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4176 contents are added to the log line that is written for every discarded
4177 recipient. (Previously a log_message setting was ignored.)
4179 36. The ${quote: operator now quotes the string if it is empty.
4181 37. The install script runs exim in order to find its version number. If for
4182 some reason other than non-existence or emptiness, which it checks, it
4183 could not run './exim', it was installing it with an empty version number,
4184 i.e. as "exim-". This error state is now caught, and the installation is
4187 38. An argument was missing from the function that creates an error message
4188 when Exim fails to connect to the socket for saslauthd authentication.
4189 This could cause Exim to crash, or give a corrupted message.
4191 39. Added isip, isip4, and isip6 to ${if conditions.
4193 40. The ACL variables $acl_xx are now saved with the message, and can be
4194 accessed later in routers, transports, and filters.
4196 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4197 strings in the file are not string-expanded.
4199 42. If a MAIL command specified a SIZE value that was too large to fit into an
4200 int variable, the check against message_size_limit failed. Such values are
4201 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4202 day this will have to be increased, but I don't think I want to be around
4203 when emails are that large.
4210 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4211 iconv() is not standard in FreeBSD.
4213 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4214 IPv6 enabled. The observed symptom was a segmentation fault on return from
4215 the function os_common_find_running_interfaces() in src/os.c.
4217 3. In the check_special_case() function in daemon.c I had used "errno" as an
4218 argument name, which causes warnings on some systems. This was basically a
4219 typo, since it was named "eno" in the comments!
4221 4. The code that waits for the clock to tick (at a resolution of some fraction
4222 of a second) so as to ensure message-id uniqueness was always waiting for
4223 at least one whole tick, when it could have waited for less. [This is
4224 almost certainly not relevant at current processor speeds, where it is
4225 unlikely to ever wait at all. But we try to future-proof.]
4227 5. The function that sleeps for a time interval that includes fractions of a
4228 second contained a race. It did not block SIGALRM between setting the
4229 timer, and suspending (a couple of lines later). If the interval was short
4230 and the sigsuspend() was delayed until after it had expired, the suspension
4231 never ended. On busy systems this could lead to processes getting stuck for
4234 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4235 process, before it forks any delivery processes. The open lookup caching
4236 mechanism meant that the open file or database connection was passed into
4237 the delivery process. The problem was that delivery processes always tidy
4238 up cached lookup data. This could cause a problem for the next delivery
4239 process started by the queue runner, because the external queue runner
4240 process does not know about the closure. So the next delivery process
4241 still has data in the lookup cache. In the case of a file lookup, there was
4242 no problem because closing a file descriptor in a subprocess doesn't affect
4243 the parent. However, if the lookup was caching a connection to a database,
4244 the connection was closed, and the second delivery process was likely to
4245 see errors such as "PGSQL: query failed: server closed the connection
4246 unexpectedly". The problem has been fixed by closing all cached lookups
4247 in a queue runner before running a delivery process.
4249 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4250 seem to have the "const" qualifier which it has on other OS. I've
4253 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4254 *of the same type* (client or server) with the same public name that an
4255 error should be diagnosed.
4257 9. When Exim looked up a host name for an IP address, but failed to find the
4258 original IP address when looking up the host name (a safety check), it
4259 output the message "<ip address> does not match any IP for NULL", which was
4260 confusing, to say the least. The bug was that the host name should have
4261 appeared instead of "NULL".
4263 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4264 user that is built into the binary, and the -C or -D options is used, root
4265 privilege is dropped before the configuration file is read. In addition,
4266 logging is switched to stderr instead of the normal log files. If the
4267 configuration then re-defines the Exim user, the unprivileged environment
4268 is probably not what is expected, so Exim logs a panic warning message (but
4271 However, if deliver_drop_privilege is set, the unprivileged state may well
4272 be exactly what is intended, so the warning has been cut out in that case,
4273 and Exim is allowed to try to write to its normal log files.
4279 1. smtp_return_error_details was not giving details for temporary sender
4280 or receiver verification errors.
4282 2. Diagnose a configuration error if two authenticators have the same public
4285 3. Exim used not to create the message log file for a message until the first
4286 delivery attempt. This could be confusing when incoming messages were held
4287 for policy or load reasons. The message log file is now created at the time
4288 the message is received, and an initial "Received" line is written to it.
4290 4. The automatically generated man page for command line options had a minor
4291 bug that caused no ill effects; however, a more serious problem was that
4292 the procedure for building the man page automatically didn't always
4293 operate. Consequently, release 4.20 contains an out-of-date version. This
4294 shouldn't happen again.
4296 5. When building Exim with embedded Perl support, the script that builds the
4297 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4298 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4300 6. The freeze_tell option was not being used for messages that were frozen on
4301 arrival, either by an ACL or by local_scan().
4303 7. Added the smtp_incomplete_transaction log selector.
4305 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4306 was accepting AUTH without a new EHLO.
4308 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4309 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4312 10. Logging of TCP/IP connections (when configured) now happens in the main
4313 daemon process instead of the child process, so that the TCP/IP connection
4314 count is more accurate (but it can never be perfect).
4316 11. The use of "drop" in a nested ACL was not being handled correctly in the
4317 outer ACL. Now, if condition failure induced by the nested "drop" causes
4318 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4319 or "require"), the connection is dropped.
4321 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4322 that yield "discard" can now be used with an "accept" or a "discard" verb,
4323 but an error is generated for any others (because I can't see a useful way
4324 to define what should happen).
4326 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4327 are now processed in the same way as lines in the Exim configuration file.
4328 In particular, continuation lines are supported.
4330 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4332 15. Added -ti meaning -t -i.
4334 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4335 domains, and warn by logging if others are found.
4337 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4338 returned from GIFCONF when Exim is trying to find the list of interfaces on
4339 a host. The code in os.c has been modified to copy each ifreq to an aligned
4340 structure in all cases.
4342 Also, in some cases, the returned ifreq's were being copied to a 'struct
4343 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4344 means the last couple of bytes of an IPv6 address could be chopped if the
4345 ifreq contained only a normal sockaddr (14 bytes storage).
4347 18. Named domain lists were not supported in the hosts_treat_as_local option.
4348 An entry such as +xxxx was not recognized, and was treated as a literal
4351 19. Ensure that header lines added by a DATA ACL are included in the reject log
4352 if the ACL subsequently rejects the message.
4354 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4355 MD5 (which is deprecated).
4357 21. When testing a filter file using -bf, Exim was writing a message when it
4358 took the sender from a "From " line in the message, but it was not doing so
4359 when it took $return_path from a Return-Path: header line. It now does.
4361 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4362 with a valid header line field name (a series of printing characters
4363 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4365 23. Changed "disc" in the source to "disk" to conform to the documentation and
4366 the book and for uniformity.
4368 24. Ignore Sendmail's -Ooption=value command line item.
4370 25. When execve() failed while trying to run a command in a pipe transport,
4371 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4372 could be confused with a return value of 69 from the command itself. This
4373 has been changed to 127, the value the shell returns if it is asked to run
4374 a non-existent command. The wording for the related log line suggests a
4375 non-existent command as the problem.
4377 26. If received_header_text expands to an empty string, do not add a Received:
4378 header line to the message. (Well, it adds a token one on the spool, but
4379 marks it "old" so that it doesn't get used or transmitted.)
4381 27. Installed eximstats 1.28 (addition of -nt option).
4383 28. There was no check for failure on the call to getsockname() in the daemon
4384 code. This can fail if there is a shortage of resources on the system, with
4385 ENOMEM, for example. A temporary error is now given on failure.
4387 29. Contrary to the C standard, it seems that in some environments, the
4388 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4389 program. Exim now does this explicitly; it affects the formatting of
4390 timestamps using strftime().
4392 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4393 complaints. I've now initialized all the variables, to avoid this.
4395 32. Header lines added by a system filter were not being "seen" during
4396 transport-time rewrites.
4398 33. The info_callback() function passed to OpenSSL is set up with type void
4399 (*)(SSL *, int, int), as described somewhere. However, when calling the
4400 function (actually a macro) that sets it up, the type void(*)() is
4401 expected. I've put in a cast to prevent warnings from picky compilers.
4403 34. If a DNS black list lookup found a CNAME record, but there were no A
4404 records associated with the domain it pointed at, Exim crashed.
4406 35. If a DNS black list lookup returned more than one A record, Exim ignored
4407 all but the first. It now scans all returned addresses if a particular IP
4408 value is being sought. In this situation, the contents of the
4409 $dnslist_value variable are a list of all the addresses, separated by a
4412 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4413 to accept a host name and all its aliases if the forward lookup for any of
4414 them yielded the IP address of the incoming connection. Now it accepts only
4415 those names whose forward lookup yields the correct IP address. Any other
4416 names are discarded. This closes a loophole whereby a rogue DNS
4417 administrator could create reverse DNS records to break through a
4418 wildcarded host restriction in an ACL.
4420 37. If a user filter or a system filter that ran in a subprocess used any of
4421 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4422 the wrong values were passed to the pipe command ($thisaddress had the
4423 value of $0, $0 had the value of $1, etc). This bug was introduced by
4424 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4426 38. Improved the line breaking for long SMTP error messages from ACLs.
4427 Previously, if there was no break point between 40 and 75 characters, Exim
4428 left the rest of the message alone. Two changes have been made: (a) I've
4429 reduced the minimum length to 35 characters; (b) if it can't find a break
4430 point between 35 and 75 characters, it looks ahead and uses the first one
4431 that it finds. This may give the occasional overlong line, but at least the
4432 remaining text gets split now.
4434 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4435 file descriptors might be low, and that setting 1000 would always raise it.
4436 It turns out that in some environments, the limit is already over 1000 and
4437 that lowering it causes trouble. So now Exim takes care not to decrease it.
4439 40. When delivering a message, the value of $return_path is set to $sender_
4440 address at the start of routing (routers may change the value). By an
4441 oversight, this default was not being set up when an address was tested by
4442 -bt or -bv, which affected the outcome if any router or filter referred to
4445 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4446 log only when "message" or "log_message" are set. However, if one of the
4447 conditions was an address verification, or a call to a nested ACL, the
4448 messages generated by the underlying test were being passed through. This
4449 no longer happens. The underlying message is available in $acl_verify_
4450 message for both "message" and "log_message" expansions, so it can be
4451 passed through if needed.
4453 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4454 new expansion $bh_ to give the encoded byte string without charset
4455 translation. Translation happens only if iconv() is available; HAVE_ICONV
4456 indicates this at build time. HEADERS_CHARSET gives the charset to
4457 translate to; headers_charset can change it in the configuration, and
4458 "headers charset" can change it in an individual filter file.
4460 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4461 that creates RFC 2047 encoded "words" labels them as that charset instead
4462 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4463 expansion operator; (ii) when Exim creates a From: line for a local
4464 message; (iii) when a header line is rewritten to include a "phrase" part.
4466 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4467 buggy, causing it to skip the first lines of messages whose message ID
4468 ended in 'D'. This would not have bitten before Exim release 4.14, because
4469 message IDs were unlikely to end in 'D' before then. The effect was to have
4470 incorrect size information for certain domains.
4472 45. #include "config.h" was missing at the start of the crypt16.c module. This
4473 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4476 46. If there was a timeout during a "random" callout check, Exim treated it as
4477 a failure of the random address, and carried on sending RSET and the real
4478 address. If the delay was just some slowness somewhere, the response to the
4479 original RCPT would be taken as a response to RSET and so on, causing
4480 mayhem of various kinds.
4482 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4483 when I implemented it. It didn't allow for the fact that some option values
4484 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4485 the right test for positive values.
4487 48. Domain names in DNS records are case-independent. Exim always looks them up
4488 in lower case. Some resolvers return domain names in exactly the case they
4489 appear in the zone file, that is, they may contain uppercase letters. Not
4490 all resolvers do this - some return always lower case. Exim was treating a
4491 change of case by a resolver as a change of domain, similar to a widening
4492 of a domain abbreviation. This triggered its re-routing code and so it was
4493 trying to route what was effectively the same domain again. This normally
4494 caused routing to fail (because the router wouldn't handle the domain
4495 twice). Now Exim checks for this case specially, and just changes the
4496 casing of the domain that it ultimately uses when it transmits the message
4499 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4502 50. If a filter generated a file delivery with a non-absolute name (possible if
4503 no home directory exists for the router), the forbid_file option was not
4506 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4507 the existing equality matching.
4509 52. Exim was using ints instead of ino_t variables in some places where it was
4510 dealing with inode numbers.
4512 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4513 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4514 called TMPDIR, and if it finds it is different, it changes its value.
4516 54. The smtp_printf() function is now made available to local_scan() so
4517 additional output lines can be written before returning. There is also an
4518 smtp_fflush() function to enable the detection of a dropped connection.
4519 The variables smtp_input and smtp_batched_input are exported to
4522 55. Changed the default runtime configuration: the message "Unknown user"
4523 has been removed from the ACL, and instead placed on the localuser router,
4524 using the cannot_route_message feature. This means that any verification
4525 failures that generate their own messages won't get overridden. Similarly,
4526 the "Unrouteable address" message that was in the ACL for unverifiable
4527 relay addresses has also been removed.
4529 56. Added hosts_avoid_esmtp to the smtp transport.
4531 57. The exicyclog script was not checking for the esoteric option
4532 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4533 will work only if exicyclog is run under the appropriate euid.
4535 58. Following a discussion on the list, the rules by which Exim recognises line
4536 endings on incoming messages have been changed. The -dropcr and drop_cr
4537 options are now no-ops, retained only for backwards compatibility. The
4538 following line terminators are recognized: LF CRLF CR. However, special
4539 processing applies to CR:
4541 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4542 nor a local message in the state where . is a terminator.
4544 (ii) If a bare CR is encountered in a header line, an extra space is added
4545 after the line terminator so as not to end the header. The reasoning
4546 behind this is that bare CRs in header lines are most likely either
4547 to be mistakes, or people trying to play silly games.
4549 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4550 was being incorrectly given as 18 bytes larger than it should have been.
4551 This is a VOB (very old bug).
4553 60. This may never have affected anything current, but just in case it has:
4554 When the local host is found other than at the start of a list of hosts,
4555 the local host, those with the same MX, and any that follow, are discarded.
4556 When the list in question was part of a longer list of hosts, the following
4557 hosts (not currently being processed) were also being discarded. This no
4558 longer happens. I'm not sure if this situation could ever has previously
4561 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4564 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4565 header. This is something that is recommended in a new Internet Draft, and
4566 is something that is documented as being done by Sendmail. There are two
4567 possible values. For messages generated by the autoreply transport, Exim
4570 Auto-Submitted: auto-replied
4572 whereas for all other generated messages (e.g. bounces) it adds
4574 Auto-Submitted: auto-generated
4576 63. The "personal" condition in filters now includes a test for the
4577 Auto-Submitted: header. If it contains the string "auto-" the message it
4578 not considered personal.
4580 64. Added rcpt_include_affixes as a generic transport option.
4582 65. Added queue_only_override (default true).
4584 66. Added the syslog_duplication option.
4586 67. If what should have been the first header line of a message consisted of
4587 a space followed by a colon, Exim was mis-interpreting it as a header line.
4588 It isn't of course - it is syntactically invalid and should therefore be
4589 treated as the start of the message body. The misbehaviour could have
4590 caused a number of strange effects, including loss of data in subsequent
4591 header lines, and spool format errors.
4593 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4594 client host had authenticated. This control can now be exercised by an ACL
4595 for more flexibility.
4597 69. By default, callouts do not happen when testing with -bh. There is now a
4598 variant, -bhc, which does actually run the callout code, including
4599 consulting and updating the callout cache.
4601 70. Added support for saslauthd authentication, courtesy of Alexander
4604 71. If statvfs() failed on the spool or log directories while checking their
4605 size for availability, Exim confusingly gave the error "space shortage".
4606 Furthermore, in debugging mode it crashed with a floating point exception.
4607 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4608 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4609 serious problem, Exim now writes to the main and panic logs when this
4610 happens, with details of the failure. It then refuses to accept the
4611 incoming message, giving the message "spool directory problem" or "log
4612 directory problem" with a 421 code for SMTP messages.
4614 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4615 0, 1, and 2 exist, because some OS complain for execs without them (see
4616 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4617 descriptors. However, the code omitted to check that the open succeeded,
4618 causing mysterious errors if for some reason the permissions on /dev/null
4619 got screwed. Now Exim writes a message to the main and panic logs, and
4620 bombs out if it can't open /dev/null.
4622 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4623 interact so that it is all more flexible. It is supposed to remain
4624 backwards compatible. Also added extra_local_interfaces.
4626 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4627 to bomb out with an assertion failure - to the client this appears as a
4628 connection drop. This problem occurs in the part of the code that was taken
4629 from the Samba project. Fortunately, the assertion is in a very simple
4630 function, so I have fixed this by reproducing the function inline in the
4631 one place where it is called, and arranging for authentication to fail
4632 instead of killing the process with assert().
4634 75. The SPA client code was not working when the server requested OEM rather
4635 than Unicode encoding.
4637 76. Added code to make require_files with a specific uid setting more usable in
4638 the case where statting the file as root fails - usually a non-root-mounted
4639 NFS file system. When this happens and the failure is EACCES, Exim now
4640 forks a subprocess and does the per-uid checking as the relevant uid.
4642 77. Added process_log_path.
4644 78. If log_file_path was not explicitly set, a setting of check_log_space or
4645 check_log_inodes was ignored.
4647 79. If a space check for the spool or log partitions fails, the incident is now
4648 logged. Of course, in the latter case the data may get lost...
4650 80. Added the %p formatting code to string_format() so that it can be used to
4651 print addresses in debug_print(). Adjusted all the address printing in the
4652 debugging in store.c to use %p rather than %d.
4654 81. There was a concern that a line of code in smtp_in.c could overflow a
4655 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4656 initially expressed, the concern was not well-founded, because trailing
4657 spaces are removed early. However, if the trailing spaces were followed by
4658 a NULL, they did not get removed, so the overflow was possible. Two fixes
4661 (a) I re-wrote the offending code in a cleaner fashion.
4662 (b) If an incoming SMTP command contains a NULL character, it is rejected
4665 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4666 runs initgroups(), so that if the Exim user is in any additional groups,
4667 they will be used during message reception.
4673 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob