1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.489 2007/03/13 09:59:07 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
159 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
160 fixed by 4.65/MH/01 (is this a record?) are fixed:
162 (i) An empty string was always treated as zero by the numeric comparison
163 operators. This behaviour has been restored.
165 (ii) It is documented that the numeric comparison operators always treat
166 their arguments as decimal numbers. This was broken in that numbers
167 starting with 0 were being interpreted as octal.
169 While fixing these problems I realized that there was another issue that
170 hadn't been noticed. Values of message_size_limit (both the global option
171 and the transport option) were treated as octal if they started with 0.
172 The documentation was vague. These values are now always treated as
173 decimal, and I will make that clear in the documentation.
179 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
180 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
183 MH/01 Don't check that the operands of numeric comparison operators are
184 integers when their expansion is in "skipping" mode (fixes bug
185 introduced by 4.64-PH/07).
187 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
188 child addresses, Exim now panics and dies. Previously, because the count
189 is held in a short int, deliveries were likely to be lost. As such a
190 large number of recipients for a single message is ridiculous
191 (performance will be very, very poor), I have chosen to impose a limit
192 rather than extend the field.
198 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
199 leftover -K file (the existence of which was triggered by #402).
200 While we were at it, introduced process PID as part of the -K
201 filename. This should rule out race conditions when creating
204 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
205 processing considerably. Previous code took too long for large mails,
206 triggering a timeout which in turn triggers #401.
208 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
209 in the DK code in transports.c. sendfile() is not really portable,
210 hence the _LINUX specificness.
212 TF/01 In the add_headers option to the mail command in an Exim filter,
213 there was a bug that Exim would claim a syntax error in any
214 header after the first one which had an odd number of characters
217 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
218 callout verification, Exim cached a "reject" for the entire domain. This
219 is correct for most verifications, but it is not correct for a recipient
220 verification with use_sender or use_postmaster set, because in that case
221 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
222 case of MAIL FROM:<> rejection from other early rejections (e.g.
223 rejection of HELO). When verifying a recipient using a non-null MAIL
224 address, the cache is ignored if it shows MAIL FROM:<> rejection.
225 Whatever the result of the callout, the value of the domain cache is
226 left unchanged (for any other kind of callout, getting as far as trying
227 RCPT means that the domain itself is ok).
229 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
232 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
233 manifest itself as EPIPE rather than ECONNECT. When tidying away a
234 session, the daemon ignores ECONNECT errors and logs others; it now
235 ignores EPIPE as well.
237 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
238 (quoted-printable decoding).
240 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
241 later the small subsequent patch to fix an introduced bug.
243 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
245 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
247 PH/08 An error is now given if message_size_limit is specified negative.
249 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
250 to be given (somewhat) arbitrary names.
252 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
255 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
256 miscellaneous code fixes
258 PH/10 Added the log_reject_target ACL modifier to specify where to log
261 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
262 hostname. This is wrong, because it relates to the incoming message (and
263 probably the interface on which it is arriving) and not to the outgoing
264 callout (which could be using a different interface). This has been
265 changed to use the value of the helo_data option from the smtp transport
266 instead - this is what is used when a message is actually being sent. If
267 there is no remote transport (possible with a router that sets up host
268 addresses), $smtp_active_hostname is used.
270 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
271 tweaks were necessary in order to get it to work (see also 21 below):
272 (a) The code assumed that strncpy() returns a negative number on buffer
273 overflow, which isn't the case. Replaced with Exim's string_format()
275 (b) There were several signed/unsigned issues. I just did the minimum
276 hacking in of casts. There is scope for a larger refactoring.
277 (c) The code used strcasecmp() which is not a standard C function.
278 Replaced with Exim's strcmpic() function.
279 (d) The code set only $1; it now sets $auth1 as well.
280 (e) A simple test gave the error "authentication client didn't specify
281 service in request". It would seem that Dovecot has changed its
282 interface. Fortunately there's a specification; I followed it and
283 changed what the client sends and it appears to be working now.
285 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
288 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
289 address A is aliased to B and C, where B exists and C does not. Without
290 -v the output is "A verified" because verification stops after a
291 successful redirection if more than one address is generated. However,
292 with -v the child addresses are also verified. Exim was outputting "A
293 failed to verify" and then showing the successful verification for C,
294 with its parentage. It now outputs "B failed to verify", showing B's
295 parentage before showing the successful verification of C.
297 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
298 look up a TXT record in a specific list after matching in a combined
301 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
302 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
303 they consult the DNS. I had assumed they would set it the way they
304 wanted; and indeed my experiments on Linux seem to show that in some
305 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
306 To be on the safe side, however, I have now made the interface to
307 host_find_byname() similar to host_find_bydns(), with an argument
308 containing the DNS resolver options. The host_find_byname() function now
309 sets these options at its start, just as host_find_bydns() does. The smtp
310 transport options dns_qualify_single and dns_search_parents are passed to
311 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
312 of host_find_byname() use the default settings of RES_DEFNAMES
313 (qualify_single) but not RES_DNSRCH (search_parents).
315 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
316 spool_read_header() do less string testing, by means of a preliminary
317 switch on the second character of optional "-foo" lines. (This is
318 overdue, caused by the large number of possibilities that now exist.
319 Originally there were few.) While I was there, I also converted the
320 str(n)cmp tests so they don't re-test the leading "-" and the first
321 character, in the hope this might squeeze out yet more improvement.
323 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
324 flag allowing group syntax was set by the header_syntax check but not
325 turned off, possible causing trouble later; (2) The flag was not being
326 set at all for the header_verify test, causing "group"-style headers to
327 be rejected. I have now set it in this case, and also caused header_
328 verify to ignore an empty address taken from a group. While doing this, I
329 came across some other cases where the code for allowing group syntax
330 while scanning a header line wasn't quite right (mostly, not resetting
331 the flag correctly in the right place). These bugs could have caused
332 trouble for malformed header lines. I hope it is now all correct.
334 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
335 with the "reply" argument non-NULL. The code, however (which originally
336 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
337 but it didn't always do it. This confused somebody who was copying the
338 code for some other use. I have removed all the tests.
340 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
341 feature that was used to support insecure browsers during the U.S. crypto
342 embargo. It requires special client support, and Exim is probably the
343 only MTA that supported it -- and would never use it because real RSA is
344 always available. This code has been removed, because it had the bad
345 effect of slowing Exim down by computing (never used) parameters for the
346 RSA_EXPORT functionality.
348 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
349 authenticator to fail if there's a tab character in the incoming data
350 (there should never be unless someone is messing about, as it's supposed
351 to be base64-encoded). Also added, on Timo's advice, the "secured" option
352 if the connection is using TLS or if the remote IP is the same as the
353 local IP, and the "valid-client-cert option" if a client certificate has
356 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
357 authenticators. This can be used for authorization after authentication
358 succeeds. (In the case of plaintext, it servers for both authentication
361 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
362 if any retry times were supplied.
364 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
365 connection, where there is no client IP address to check. In this
366 situation, the verify now always succeeds.
368 PH/25 Applied John Jetmore's -Mset patch.
370 PH/26 Added -bem to be like -Mset, but loading a message from a file.
372 PH/27 In a string expansion for a processed (not raw) header when multiple
373 headers of the same name were present, leading whitespace was being
374 removed from all of them, but trailing whitespace was being removed only
375 from the last one. Now trailing whitespace is removed from each header
376 before concatenation. Completely empty headers in a concatenation (as
379 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
380 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
382 PH/29 [Removed. This was a change that I later backed out, and forgot to
383 correct the ChangeLog entry (that I had efficiently created) before
384 committing the later change.]
386 PH/30 Exim was sometimes attempting to deliver messages that had suffered
387 address errors (4xx response to RCPT) over the same connection as other
388 messages routed to the same hosts. Such deliveries are always "forced",
389 so retry times are not inspected. This resulted in far too many retries
390 for the affected addresses. The effect occurred only when there were more
391 hosts than the hosts_max_try setting in the smtp transport when it had
392 the 4xx errors. Those hosts that it had tried were not added to the list
393 of hosts for which the message was waiting, so if all were tried, there
394 was no problem. Two fixes have been applied:
396 (i) If there are any address or message errors in an SMTP delivery, none
397 of the hosts (tried or untried) are now added to the list of hosts
398 for which the message is waiting, so the message should not be a
399 candidate for sending over the same connection that was used for a
400 successful delivery of some other message. This seems entirely
401 reasonable: after all the message is NOT "waiting for some host".
402 This is so "obvious" that I'm not sure why it wasn't done
403 previously. Hope I haven't missed anything, but it can't do any
404 harm, as the worst effect is to miss an optimization.
406 (ii) If, despite (i), such a delivery is accidentally attempted, the
407 routing retry time is respected, so at least it doesn't keep
408 hammering the server.
410 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
411 in ${readsocket because some servers need this prod.
413 PH/32 Added some extra debug output when updating a wait-xxx database.
415 PH/33 The hint "could be header name not terminated by colon", which has been
416 given for certain expansion errors for a long time, was not being given
417 for the ${if def:h_colon_omitted{... case.
419 PH/34 The spec says: "With one important exception, whenever a domain list is
420 being scanned, $domain contains the subject domain." There was at least
421 one case where this was not true.
423 PH/35 The error "getsockname() failed: connection reset by peer" was being
424 written to the panic log as well as the main log, but it isn't really
425 panic-worthy as it just means the connection died rather early on. I have
426 removed the panic log writing for the ECONNRESET error when getsockname()
429 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
430 runs only) independently of the message's sender address. This meant
431 that, if the 4xx error was in fact related to the sender, a different
432 message to the same recipient with a different sender could confuse
433 things. In particualar, this can happen when sending to a greylisting
434 server, but other circumstances could also provoke similar problems.
435 I have changed the default so that the retry time for these errors is now
436 based a combination of the sender and recipient addresses. This change
437 can be overridden by setting address_retry_include_sender=false in the
440 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
441 remote server are returned as part of bounce messages. This was not
442 happening for LMTP over a pipe (the lmtp transport), but now it is the
443 same for both kinds of LMTP.
445 PH/38 Despite being documented as not happening, Exim was rewriting addresses
446 in header lines that were in fact CNAMEs. This is no longer the case.
448 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
449 and queue runs started by the daemon processed all messages. This has
450 been fixed so that -R and -S can now usefully be given with -q<time>.
452 PH/40 Import PCRE release 6.7 (fixes some bugs).
454 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
456 PH/42 Give an error if -q is specified more than once.
458 PH/43 Renamed the variables $interface_address and $interface_port as
459 $received_ip_address and $received_port, to make it clear that these
460 values apply to message reception, and not to the outgoing interface when
461 a message is delivered. (The old names remain recognized, of course.)
463 PH/44 There was no timeout on the connect() call when using a Unix domain
464 socket in the ${readsocket expansion. There now is.
466 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
467 be meaningful with "accept".
469 SC/01 Eximstats V1.43
470 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
472 SC/02 Eximstats V1.44
473 Use a glob alias rather than an array ref in the generated
474 parser. This improves both readability and performance.
476 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
477 Collect SpamAssassin and rejection statistics.
478 Don't display local sender or destination tables unless
479 there is data to show.
480 Added average volumes into the top table text output.
482 SC/04 Eximstats V1.46
483 Collect data on the number of addresses (recipients)
484 as well as the number of messages.
486 SC/05 Eximstats V1.47
487 Added 'Message too big' to the list of mail rejection
488 reasons (thanks to Marco Gaiarin).
490 SC/06 Eximstats V1.48
491 Mainlog lines which have GMT offsets and are too short to
492 have a flag are now skipped.
494 SC/07 Eximstats V1.49 (Alain Williams)
495 Added the -emptyok flag.
497 SC/08 Eximstats V1.50
498 Fixes for obtaining the IP address from reject messages.
500 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
501 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
502 whitesspace changes from 4.64-PH/27
504 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
507 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
508 are found, allow negative numbers in numeric criteria)
510 JJ/06 exipick.20061117.2, added new $message_body_missing variable
512 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
513 to match changes made in 4.64-PH/43
515 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
517 PH/47 Put in an explicit test for a DNS lookup of an address record where the
518 "domain" is actually an IP address, and force a failure. This locks out
519 those revolvers/nameservers that support "A-for-A" lookups, in
520 contravention of the specifications.
522 PH/48 When a host name was looked up from an IP address, and the subsequent
523 forward lookup of the name timed out, the host name was left in
524 $sender_host_name, contrary to the specification.
526 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
527 restricted to single-key lookups, Exim was not diagnosing an error if
528 * or *@ was used with a query-style lookup.
530 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
532 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
533 long ago, but noone interested enough thought of it. Let's just say that
534 the "1.1" means that there are some new functions that weren't there at
535 some point in the past.
537 PH/51 Error processing for expansion failure of helo_data from an smtp
538 transport during callout processing was broken.
540 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
541 tested/used via the -bh/-bhc/-bs options.
543 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
544 bug, fixed in subsequent PCRE releases).
546 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
547 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
549 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
555 SC/01 Use a glob alias rather than an array ref in eximstats generated
556 parser. This improves both readability and performance.
558 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
559 Don't display local sender or destination tables in eximstats unless
560 there is data to show.
561 Added average volumes into the eximstats top table text output.
563 SC/03 Collect data on the number of addresses (recipients) as well
564 as the number of messages in eximstats.
566 TF/01 Correct an error in the documentation for the redirect router. Exim
567 does (usually) call initgroups() when daemonizing.
569 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
570 with consistent privilege compared to when running as a daemon.
572 TF/03 Note in the spec that $authenticated_id is not set for local
573 submissions from trusted users.
575 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
576 Thanks to Dean Brooks <dean@iglou.com> for the patch.
578 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
579 by adding some example configuration directives to the default
580 configuration file. A little bit of work is required to uncomment the
581 directives and define how usernames and passwords are checked, but
582 there is now a framework to start from.
584 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
585 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
586 without this. I don't know how relevant this is to other LDAP libraries.
588 PH/02 Add the verb name to the "unknown ACL verb" error.
590 PH/03 Magnus Holmgren's patch for filter_prepend_home.
592 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
594 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
595 directory not expanded when it should be if an expanded home directory
596 was set for the address (which is overridden by the transport).
598 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
601 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
602 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
603 because it is too late at that time, and has no effect.
605 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
606 security issue with \' (bugzilla #107). I could not use the
607 PQescapeStringConn() function, because it needs a PGconn value as one of
610 PH/08 When testing addresses using -bt, indicate those final addresses that
611 are duplicates that would not cause an additional delivery. At least one
612 person was confused, thinking that -bt output corresponded to deliveries.
613 (Suppressing duplicates isn't a good idea as you lose the information
614 about possibly different redirections that led to the duplicates.)
616 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
617 systems where poll() doesn't work, in particular OS X.
619 PH/10 Added more information to debugging output for retry time not reached.
621 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
622 operations in malware.c.
624 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
627 PH/13 If write_rejectlog was set false when logging was sent to syslog with
628 syslog_duplication set false, log lines that would normally be written
629 both the the main log and to the reject log were not written to syslog at
632 PH/14 In the default configuration, change the use of "message" in ACL warn
633 statements to "add_header".
635 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
636 not followed by a command (e.g. "seen endif").
638 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
639 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
642 PH/17 Added extra conditions to the default value of delay_warning_condition
646 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
647 { match{$h_precedence:}{(?i)bulk|list|junk} } \
648 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
651 The Auto-Submitted: and various List- headers are standardised, whereas I
652 don't think Precedence: ever was.
654 PH/18 Refactored debugging code in route_finduser() to show more information,
655 in particular, the error code if getpwnam() issues one.
657 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
658 This is apparently needed in addition to the PH/07 change above to avoid
659 any possible encoding problems.
661 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
662 but not after initializing Perl.
664 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
665 output them only if debugging. By default they are written stderr,
666 apparently, which is not desirable.
668 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
671 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
674 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
676 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
677 authenticated or an ident call has been made. Suppress the default
678 values for $authenticated_id and $authenticated_sender (but permit -oMai
679 and -oMas) when testing with -bh.
681 PH/24 Re-jigged the order of the tests in the default configuration so that the
682 tests for valid domains and recipients precede the DNS black list and CSA
683 tests, on the grounds that those ones are more expensive.
685 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
686 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
687 This bug exists in every version of Exim that I still have, right back to
690 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
691 However, an attempt to turn on case-sensitivity in a regex key by
692 including (?-i) didn't work because the subject string was already
693 lowercased, and the effects were non-intuitive. It turns out that a
694 one-line patch can be used to allow (?-i) to work as expected.
700 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
701 other effects) broke the use of negated acl sub-conditions.
703 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
706 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
707 "Deny" causes Exim to reject the incoming connection with a 554 error.
708 Unfortunately, if there is a major crisis, such as a disk failure,
709 tcp-wrappers gives "deny", whereas what one would like would be some
710 kind of temporary error. A kludge has been added to help with this.
711 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
712 554 error is used if errno is still zero or contains ENOENT (which occurs
713 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
716 PH/03 Add -lutil to the default FreeBSD LIBS setting.
718 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
719 errors. Otherwise a message that provokes a temporary error (when other
720 messages do not) can cause a whole host to time out.
722 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
723 addresses were routed directly to files or pipes from a redirect router.
724 File deliveries just didn't batch; pipe deliveries might have suffered
727 PH/06 A failure to get a lock for a hints database would erroneously always say
728 "Failed to get write lock", even when it was really a read lock.
730 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
731 of 0600. This has been changed to use the value of the lockfile_mode
732 option (which defaults to 0600).
734 PH/08 Applied small patch from the Sieve maintainer.
736 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
737 folder from quota calculations, a direct delivery into this folder messed
738 up the contents of the maildirsize file. This was because the regex was
739 used only to exclude .Trash (or whatever) when the size of the mailbox
740 was calculated. There was no check that a delivery was happening into an
741 excluded directory. This bug has been fixed by ignoring all quota
742 processing for deliveries into excluded directories.
744 PH/10 Added the maildirfolder_create_regex option to appendfile.
750 PH/01 The code for finding all the local interface addresses on a FreeBSD
751 system running IPv6 was broken. This may well have applied to all BSD
752 systems, as well as to others that have similar system calls. The broken
753 code found IPv4 interfaces correctly, but gave incorrect values for the
754 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
755 that it would not match correctly against @[] and not recognize the IPv6
758 PH/02 The ipliteral router was not recognizing addresses of the form user@
759 [ipv6:....] because it didn't know about the "ipv6:" prefix.
761 PH/03 Added disable_ipv6.
763 PH/04 Changed $reply_address to use the raw form of the headers instead of the
764 decoded form, because it is most often used to construct To: headers
765 lines in autoreplies, and the decoded form may well be syntactically
766 invalid. However, $reply_address has leading white space removed, and all
767 newlines turned into spaces so that the autoreply transport does not
770 PH/05 If group was specified without a user on a router, and no group or user
771 was specified on a transport, the group from the router was ignored.
773 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
774 for visible compile-time settings that can be used to change these
775 numbers, for those that want even more. Backwards compatibility with old
776 spool files has been maintained. However, going back to a previous Exim
777 release will lost any variables that are in spool files.
779 PH/07 Two small changes when running in the test harness: increase delay when
780 passing a TCP/IP connection to a new process, in case the original
781 process has to generate a bounce, and remove special handling of
782 127.0.0.2 (sic), which is no longer necessary.
784 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
785 be the same on different OS.
787 PH/09 Moved a debug statement in filter processing to avoid a race problem when
790 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
791 whether --show-vars was specified or not
793 JJ/02 exipick: Added support for new ACL variable spool format introduced
796 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
797 syntactically invalid From: or Reply-to: line, and a filter used this to
798 generate an autoreply, and therefore failed to obtain an address for the
799 autoreply, Exim could try to deliver to a non-existent relative file
800 name, causing unrelated and misleading errors. What now happens is that
801 it logs this as a hard delivery error, but does not attempt to create a
804 PH/11 The exinext utility has a -C option for testing purposes, but although
805 the given file was scanned by exinext itself; it wasn't being passed on
806 when Exim was called.
808 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
809 an end-of-file indication when reading a command response.
811 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
812 compiled. In many other places in Exim, IPv6 addresses are always
813 recognized, so I have changed this. It also means that IPv4 domain
814 literals of the form [IPV4:n.n.n.n] are now always recognized.
816 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
817 used if the router is not running as root, for example, when verifying at
818 ACL time, or when using -bh. The debugging output from this situation was
819 non-existent - all you got was a failure to exec. I have made two
822 (a) Failures to set uid/gid, the current directory, or a process leader
823 in a subprocess such as that created by queryprogram now generate
824 suitable debugging ouput when -d is set.
826 (b) The queryprogram router detects when it is not running as root,
827 outputs suitable debugging information if -d is set, and then runs
828 the subprocess without attempting to change uid/gid.
830 PH/15 Minor change to Makefile for building test_host (undocumented testing
833 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
834 additional section of a DNS packet that returns MX or SRV records.
835 Instead, it always explicitly searches for A/AAAA records. This avoids
836 major problems that occur when a DNS server includes only records of one
837 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
838 fixed another bug: if SRV records were looked up and the corresponding
839 address records were *not* found in the additional section, the port
840 values from the SRV records were lost.
842 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
843 using the correct key (the original address) when searching the retry
844 rules in order to find which one to use for generating the retry hint.
846 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
847 adding the default one. Similarly, if it contains a Reply-To: header, the
848 errors_reply_to option, if set, is not used.
850 PH/19 When calculating a retry time, Exim used to measure the "time since
851 failure" by looking at the "first failed" field in the retry record. Now
852 it does not use this if it is later than than the arrival time of the
853 message. Instead it uses the arrival time. This makes for better
854 behaviour in cases where some deliveries succeed, thus re-setting the
855 "first failed" field. An example is a quota failure for a huge message
856 when small messages continue to be delivered. Without this change, the
857 "time since failure" will always be short, possible causing more frequent
858 delivery attempts for the huge message than are intended.
859 [Note: This change was subsequently modified - see PH/04 for 4.62.]
861 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
862 $1, $2, $3) because the numerical variables can be reset during some
863 expansion items (e.g. "match"), thereby losing the authentication data.
865 PH/21 Make -bV show the size of off_t variables so that the test suite can
866 decide whether to run tests for quotas > 2G.
868 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
869 mailbox_size, and mailbox_filecount in the appendfile transport. If a
870 filecount value is greater than 2G or if a quota value is greater than 2G
871 on a system where the size of off_t is not greater than 4, a panic error
874 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
875 never match. The debug and -bh output now contains an explicit error
876 message indicating a malformed IPv4 address or mask.
878 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
879 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
882 PH/25 Do not write to syslog when running in the test harness. The only
883 occasion when this arises is a failure to open the main or panic logs
884 (for which there is an explicit test).
886 PH/26 Added the /no_tell option to "control=freeze".
888 PH/27 If a host name lookup failed very early in a connection, for example, if
889 the IP address matched host_lookup and the reverse lookup yielded a name
890 that did not have a forward lookup, an error message of the form "no IP
891 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
892 could be logged. Now it outputs the IP address instead of "NULL".
894 PH/28 An enabling patch from MH: add new function child_open_exim2() which
895 allows the sender and the authenticated sender to be set when
896 submitting a message from within Exim. Since child_open_exim() is
897 documented for local_scan(), the new function should be too.
899 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
900 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
901 results in an empty string is now treated as unset.
903 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
905 PH/31 Added sender_verify_fail logging option.
907 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
908 needed by RFC 822 but not by RFC 2822 was commented out. I have now
909 tidied the source and removed it altogether.
911 PH/33 When a queue run was abandoned because the load average was too high, a
912 log line was always written; now it is written only if the queue_run log
913 selector is set. In addition, the log line for abandonment now contains
914 information about the queue run such as the pid. This is always present
915 in "start" and "stop" lines but was omitted from the "abandon" line.
917 PH/34 Omit spaces between a header name and the colon in the error message that
918 is given when verify = headers_syntax fails (if there are lots of them,
919 the message gets confusing).
921 PH/35 Change the default for dns_check_names_pattern to allow slashes within
922 names, as there are now some PTR records that contain slashes. This check
923 is only to protect against broken name servers that fall over on strange
924 characters, so the fact that it applies to all lookups doesn't matter.
926 PH/36 Now that the new test suite is complete, we can remove some of the
927 special code in Exim that was needed for the old test suite. For example,
928 sorting DNS records because real resolvers return them in an arbitrary
929 order. The new test suite's fake resolver always returns records in the
932 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
933 bounces) except when queue_only is set, to avoid logging races between
934 the different processes.
936 PH/38 Panic-die if .include specifies a non-absolute path.
938 PH/39 A tweak to the "H" retry rule from its user.
940 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
941 a label. They prevented compilation on older perls.
943 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
944 a warning to be raised on newish perls.
946 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
947 on queue. Changes to match documented behaviour of showing count of
948 messages matching specified criteria.
950 PH/40 Changed the default ident timeout from 30s to 5s.
952 PH/41 Added support for the use of login_cap features, on those BSD systems
953 that have them, for controlling the resources used by pipe deliveries.
955 PH/42 The content-scanning code uses fopen() to create files in which to put
956 message data. Previously it was not paying any attention to the mode of
957 the files. Exim runs with umask(0) because the rest of the code creates
958 files with open(), and sets the required mode explicitly. Thus, these
959 files were ending up world-writeable. This was not a big issue, because,
960 being within the spool directory, they were not world-accessible. I have
961 created a function called modefopen, which takes an additional mode
962 argument. It sets umask(777), creates the file, chmods it to the required
963 mode, then resets the umask. All the relevant calls to fopen() in the
964 content scanning code have been changed to use this function.
966 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
967 to 24 hours. This avoids potential overflow problems when processing G
968 and H retry rules. I suspect nobody ever tinkers with this value.
970 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
972 PH/45 When the plaintext authenticator is running as a client, the server's
973 challenges are checked to ensure they are valid base64 strings. By
974 default, the authentication attempt is cancelled if an invalid string is
975 received. Setting client_ignore_invalid_base64 true ignores these errors.
976 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
977 they are received. Thus, the responses can be made to depend on the
978 challenges. If an invalid string is ignored, an empty string is placed in
981 PH/46 Messages that are created by the autoreply transport now contains a
982 References: header, in accordance with RFCs 2822 and 3834.
984 PH/47 Added authenticated_sender_force to the smtp transport.
986 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
988 PH/49 Installed latest patch from the Sieve maintainer.
990 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
991 was also set, the appendfile transport was unnecessarily scanning a
992 directory of message files (e.g. for maildir delivery) to find the count
993 of files (along with the size), even though it did not need this
994 information. It now does the scan only if it needs to find either the
995 size of the count of files.
997 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
999 PH/52 Two bugs concerned with error handling when the smtp transport is
1002 (i) Exim was not creating retry information for temporary errors given
1003 for individual recipients after the DATA command when the smtp transport
1004 was used in LMTP mode. This meant that they could be retried too
1005 frequently, and not timed out correctly.
1007 (ii) Exim was setting the flag that allows error details to be returned
1008 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1009 recipients that were returned after the DATA command.
1011 PH/53 This is related to PH/52, but is more general: for any failing address,
1012 when detailed error information was permitted to be returned to the
1013 sender, but the error was temporary, then after the final timeout, only
1014 "retry timeout exceeded" was returned. Now it returns the full error as
1015 well as "retry timeout exceeded".
1017 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1018 do this, and (what is worse) MTAs that accept it.
1020 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1021 will now be deprecated.
1023 PH/56 New os.c-cygwin from the Cygwin maintainer.
1025 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1026 formats (previously only available in exim formats via -bpr, -bpru,
1027 and -bpra. Now also available in native and exiqgrep formats)
1029 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1030 with very large, slow to parse queues
1032 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1034 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1036 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1037 responses to authentication challenges, though it was showing the
1038 challenges; (ii) I've removed the CR characters from the debug output for
1041 PH/58 Allow for the insertion of a newline as well as a space when a string
1042 is turned into more than one encoded-word during RFC 2047 encoding. The
1043 Sieve code now uses this.
1045 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1046 data_4xx, lost_connection, tls_required.
1048 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1049 message was being sent as an SMTP response.
1051 PH/61 Add -l and -k options to exicyclog.
1053 PH/62 When verifying, if an address was redirected to one new address, so that
1054 verification continued, and the new address failed or deferred after
1055 having set something in $address_data, the value of $address_data was not
1056 passed back to the ACL. This was different to the case when no
1057 redirection occurred. The value is now passed back in both cases.
1059 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1060 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1061 use login_cap.h, so on its own it isn't the distinguishing feature. The
1062 new name refers directly to the setclassresources() function.
1064 PH/65 Added configuration files for NetBSD3.
1066 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1068 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1069 is preferred over IPv4.
1071 PH/68 The bounce_return_message and bounce_return_body options were not being
1072 honoured for bounces generated during the reception of non-SMTP messages.
1073 In particular, this applied to messages rejected by the ACL. This bug has
1074 been fixed. However, if bounce_return_message is true and bounce_return_
1075 body is false, the headers that are returned for a non-SMTP message
1076 include only those that have been read before the error was detected.
1077 (In the case of an ACL rejection, they have all been read.)
1079 PH/69 The HTML version of the specification is now built in a directory called
1080 spec_html instead of spec.html, because the latter looks like a path with
1081 a MIME-type, and this confuses some software.
1083 PH/70 Catch two compiler warnings in sieve.c.
1085 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1086 function verify_get_ident() calls ip_connect() to connect a socket, but
1087 if the "connect()" function timed out, ip_connect() used to close the
1088 socket. However, verify_get_ident() also closes the socket later, and in
1089 between Exim writes to the log, which may get opened at this point. When
1090 the socket was closed in ip_connect(), the log could get the same file
1091 descriptor number as the socket. This naturally causes chaos. The fix is
1092 not to close the socket in ip_connect(); the socket should be closed by
1093 the function that creates it. There was only one place in the code where
1094 this was missing, in the iplookup router, which I don't think anybody now
1095 uses, but I've fixed it anyway.
1097 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1098 well as to direct DNS lookups. Otherwise the handling of names in host
1099 lists is inconsistent and therefore confusing.
1105 PH/01 Two changes to the default runtime configuration:
1107 (1) Move the checks for relay_from_hosts and authenticated clients from
1108 after to before the (commented out) DNS black list checks.
1110 (2) Add control=submission to the relay_from_hosts and authenticated
1111 clients checks, on the grounds that messages accepted by these
1112 statements are most likely to be submissions.
1114 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1116 (1) Generate an error if the third argument for the ${prvs expansion is
1119 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1122 (3) Reset the variables that are obtained from the first argument of
1123 ${prvscheck and used in the second argument before leaving the code,
1124 because their memory is reclaimed, so using them afterwards may do
1127 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1128 one (it's much easier than Tom thought :-).
1130 (5) Because of (4), we can now allow for the use of $prvscheck_result
1131 inside the third argument.
1133 PH/03 For some reason, the default setting of PATH when running a command from
1134 a pipe transport was just "/usr/bin". I have changed it to
1137 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1138 anything to be listed in the output from -bV.
1140 PH/05 When a filter generated an autoreply, the entire To: header line was
1141 quoted in the delivery log line, like this:
1143 => >A.N.Other <ano@some.domain> <original@ddress> ...
1145 This has been changed so that it extracts the operative address. There
1146 may be more than one such address. If so, they are comma-separated, like
1149 => >ano@some.domain,ona@other.domain <original@ddress> ...
1151 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1152 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1153 not being looked up in the rDNS to get a host name, Exim was showing the
1154 IP address twice in Received: lines, even though the IP addresses were
1155 identical. For example:
1157 Received: from [1.2.3.4] (helo=[1.2.3.4])
1159 However, if the real host name was known, it was omitting the HELO data
1160 if it matched the actual IP address. This has been tidied up so that it
1161 doesn't show the same IP address twice.
1163 PH/07 When both +timestamp and +memory debugging was on, the value given by
1164 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1165 called by the debug printing, thereby overwriting the timestamp buffer.
1166 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1168 PH/08 When the original message was included in an autoreply transport, it
1169 always said "this is a copy of the message, including all the headers",
1170 even if body_only or headers_only was set. It now gives an appropriate
1173 PH/09 Applied a patch from the Sieve maintainer which:
1175 o fixes some comments
1176 o adds the (disabled) notify extension core
1177 o adds some debug output for the result of if/elsif tests
1178 o points to the current vacation draft in the documentation
1179 and documents the missing references header update
1183 o fixes a bug in processing the envelope test (when testing
1184 multiple envelope elements, the last element determinted the
1187 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1188 Electronic Mail") by including:
1190 Auto-submitted: auto-generated
1192 in the messages that it generates (bounce messages and others, such as
1193 warnings). In the case of bounce messages for non-SMTP mesages, there was
1194 also a typo: it was using "Auto_submitted" (underscore instead of
1195 hyphen). Since every message generated by Exim is necessarily in response
1196 to another message, thes have all been changed to:
1198 Auto-Submitted: auto-replied
1200 in accordance with these statements in the RFC:
1202 The auto-replied keyword:
1204 - SHOULD be used on messages sent in direct response to another
1205 message by an automatic process,
1207 - MUST NOT be used on manually-generated messages,
1209 - MAY be used on Delivery Status Notifications (DSNs) and Message
1210 Disposition Notifications (MDNs),
1212 - MUST NOT be used on messages generated by automatic or periodic
1213 processes, except for messages which are automatic responses to
1216 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1217 to the default Received: header definition.
1219 PH/12 Added log selector acl_warn_skipped (default on).
1221 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1222 variables because (a) they are in the wrong storage pool and (b) even if
1223 they were copied, it wouldn't work properly because of the caching.
1225 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1226 checking when decoding. Apparently there are clients that generate
1227 overlong encoded strings. Why am I not surprised?
1229 PH/15 If the first argument of "${if match_address" was not empty, but did not
1230 contain an "@" character, Exim crashed. Now it writes a panic log message
1231 and treats the condition as false.
1233 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1235 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1236 extension "envelope-auth". The code is finished and in agreement with
1237 other implementations, but there is no documentation so far and in fact,
1238 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1239 not changing the active code.
1241 Print executed "if" and "elsif" statements when debugging is used. This
1242 helps a great deal to understand what a filter does.
1244 Document more things not specified clearly in RFC3028. I had all this
1245 sorted out, when out of a sudden new issues came to my mind. Oops."
1247 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1250 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1251 canonical form (as documented). However, after a host name lookup from
1252 the IP address, check_host() was doing a simple string comparison with
1253 addresses acquired from the DNS when checking that the found name did
1254 have the original IP as one of its addresses. Since any found IPv6
1255 addresses are likely to be in abbreviated form, the comparison could
1256 fail. Luckily, there already exists a function for doing the comparison
1257 by converting both addresses to binary, so now that is used instead of
1258 the text comparison.
1260 PH/20 There was another similar case to PH/19, when a complete host name was
1261 given in a host list; looking up its IP address could give an abbreviated
1262 form, whereas the current host's name might or might not be abbreviated.
1263 The same fix has been applied.
1269 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1270 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1273 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1274 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1276 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1278 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1279 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1280 Also, for "Auto-Submitted", treat anything other than "no" as
1281 non-personal, in accordance with RFC 3834. (Previously it treated
1282 anything starting "auto-" as non-personal.)
1284 TF/01 The control=submission/name=... option had a problem with syntax
1285 errors if the name included a slash character. The /name= option
1286 now slurps the rest of the string, so it can include any characters
1287 but it must come last in the list of options (after /sender_retain
1290 PH/05 Some modifications to the interface to the fake nameserver for the new
1298 TK/01 Added the "success_on_redirect" address verification option. See
1299 NewStuff for rationale and an example.
1301 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1303 PH/02 Patch to exigrep to allow it to work on syslog lines.
1305 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1306 fread() to skip over the body file's header line, because in Cygwin the
1307 header line is locked and is inaccessible.
1309 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1310 co-exist for some time) to make it clear that it is the Exim ID that is
1311 referenced, not the Message-ID: header line.
1313 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1314 string_format() function, because snprintf() does not exist on all
1317 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1318 ${stat: expansion item.
1320 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1321 protocol synchronization error", to keep the pedants happy.
1323 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1324 well as for IRIX systems, when gcc is being used. See the host.c source
1327 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1329 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1332 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1333 transport and to the smtp transport in LMTP mode.
1335 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1337 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1338 run a filter in a subprocess. This could lead to confusion in subsequent
1339 lookups in the parent process. There should also be a search_tidyup() at
1340 the end of the subprocess.
1342 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1343 only if the host matched helo_try_verify_hosts, which caused the
1344 verification to occur when the EHLO/HELO command was issued. The ACL just
1345 tested the remembered result. Now, if a previous verification attempt has
1346 not happened, "verify = helo" does it there and then.
1348 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1350 TK/03 Fix log output including CR from clamd.
1352 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1353 exist provoked a memory error which could cause a segfault.
1355 PH/15 Installed PCRE 6.2
1357 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1359 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1360 of the problem. Specifically, suggested +O2 rather than +O1 for the
1363 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1365 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1366 record, the port was not correctly logged when the outgoing_port log
1367 selector was set (it logged the transort's default port).
1369 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1370 fallback_hosts, and "hosts" in the smtp transport.
1372 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1373 host errors such as "Connection refused".
1375 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1376 authentication with radiusclient 0.4.9:
1378 - Error returned from rc_read_config was caught wrongly
1379 - Username/password not passed on to radius server due to wrong length.
1381 The presumption is that some radiusclient API changes for 4.51/PH/17
1382 were not taken care of correctly. The code is still untested by me (my
1383 Linux distribution still has 0.3.2 of radiusclient), but it was
1384 contributed by a Radius user.
1386 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1387 expanding the "port" option of the smtp transport.
1389 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1390 while reading a MIME header. Thanks to Tom Hughes for a patch.
1392 PH/24 Include config.h inside local_scan.h so that configuration settings are
1395 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1396 that in an ACL for RCPT (for example), you can examine exactly what was
1399 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1400 commands, but it was not correctly comparing the address with the actual
1401 client host address. Thus, it would show the EHLO address in Received:
1402 header lines when this was not necessary.
1404 PH/27 Added the % operator to ${eval:}.
1406 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1407 it should be ignoring failures (because with -C, for example, it has lost
1408 privilege). It wasn't ignoring creation failures other than "already
1411 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1414 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1415 because an input line was too long, either on its own, or by virtue of
1416 too many continuations, the temporary file was not being removed, and the
1417 return code was incorrect.
1419 PH/31 Missing "BOOL" in function definition in filtertest.c.
1421 PH/32 Applied Sieve patches from the maintainer.
1423 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1425 PH/33 Added "verify = not_blind".
1427 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1428 Local/Makefile (with some defaults set). These are used in built scripts
1429 such as exicyclog, but they have never been used in the exim_install
1430 script (though there are many overriding facilities there). I have
1431 arranged that the exim_install script now takes note of these two
1434 PH/35 Installed configuration files for Dragonfly.
1436 PH/36 When a locally submitted message by a trusted user did not contain a
1437 From: header, and the sender address was obtained from -f or from an SMTP
1438 MAIL command, and the trusted user did not use -F to supply a sender
1439 name, $originator_name was incorrectly used when constructing a From:
1440 header. Furthermore, $originator_name was used for submission mode
1441 messages from external hosts without From: headers in a similar way,
1442 which is clearly wrong.
1444 PH/37 Added control=suppress_local_fixups.
1446 PH/38 When log_selector = +received_sender was set, and the addition of the
1447 sender made the log line's construction buffer exactly full, or one byte
1448 less than full, an overflow happened when the terminating "\n" was
1451 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1452 when the result of a list match is failure because a DNS lookup failed.
1454 PH/40 RM_COMMAND is now used in the building process.
1456 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1457 the "build-* directories that it finds.
1459 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1460 address in a domain literal was a prefix of an interface address.
1462 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1463 when verifying a sender address, unless rewrite_headers is false.
1465 PH/44 Wrote a long comment about why errors_to addresses are verified as
1466 recipients, not senders.
1468 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1469 the ratelimit ACL was added.
1471 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1473 PH/46 Added extra information about PostgreSQL errors to the error string.
1475 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1476 avoiding the need to install special zones in a real server. This is
1477 backwards compatible; if it can't find the fake resolver, it drops back.
1478 Thus, both old and new test suites can be run.
1480 TF/02 Added util/ratelimit.pl
1482 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1483 clock is set back in time.
1485 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1486 Candler <B.Candler@pobox.com>.
1488 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1489 OK for addresses that are the result of redirections.
1491 PH/48 A number of further additions for the benefit of the new test suite,
1492 including a fake gethostbyname() that interfaces to the fake DNS resolver
1495 TF/06 The fix for widen_domains has also been applied to qualify_single and
1496 search_parents which are the other dnslookup options that can cause
1499 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1502 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1504 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1505 service" if it encountered an MX record with an empty target hostname.
1506 The message is now "an MX or SRV record indicated no SMTP service".
1508 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1509 if the DNS of the sending site is misconfigured. This is quite a
1510 common situation. This change restores the behaviour of treating a
1511 helo verification defer as a failure.
1513 PH/51 If self=fail was set on a router, the bounce message did not include the
1514 actual error message.
1520 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1522 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1523 command itself ended in error, the underlying message about the transport
1524 filter timeout was being overwritten with the pipe command error. Now the
1525 underlying error message should be appended to the second error message.
1527 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1529 PH/02 Reduce the amount of output that "make" produces by default. Full output
1530 can still be requested.
1532 PH/03 The warning log line about a condition test deferring for a "warn" verb
1533 was being output only once per connection, rather than after each
1534 occurrence (because it was using the same function as for successful
1535 "warn" verbs). This seems wrong, so I have changed it.
1537 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1538 it should not have, which might have caused a crash in the right
1539 circumstances, but probably never did.
1541 PH/04 Installed a modified version of Tony Finch's patch to make submission
1542 mode fix the return path as well as the Sender: header line, and to
1543 add a /name= option so that you can make the user's friendly name appear
1546 TF/03 Added the control = fakedefer ACL modifier.
1548 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1549 Mark Lowes for thorough testing.
1551 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1553 TK/03 Merged latest SRS patch from Miles Wilton.
1555 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1556 with the definition in sysexits.h (which is #included earlier).
1557 Fortunately, Exim does not actually use EX_OK. The code used to try to
1558 preserve the sysexits.h value, by assumimg that macro definitions were
1559 scanned for macro replacements. I have been disabused of this notion,
1560 so now the code just undefines EX_OK before #including unistd.h.
1562 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1563 in the smtp transport. When a block could not be written in a single
1564 write() function, the timeout was being re-applied to each part-write.
1565 This seems wrong - if the receiver was accepting one byte at a time it
1566 would take for ever. The timeout is now adjusted when this happens. It
1567 doesn't have to be particularly precise.
1569 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1570 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1572 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1573 without a domain if the check to <postmaster@domain> fails.
1575 SC/01 Eximstats: added -xls and the ability to specify output files
1576 (patch written by Frank Heydlauf).
1578 SC/02 Eximstats: use FileHandles for outputing results.
1580 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1582 SC/04 Eximstats: fixed display of large numbers with -nvr option
1584 SC/05 Eximstats: fixed merging of reports with empty tables.
1586 SC/06 Eximstats: added the -include_original_destination flag
1588 SC/07 Eximstats: removed tabs and trailing whitespace.
1590 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1592 TK/06 MBOX spool code: Add real "From " MBOX separator line
1593 so the .eml file is really in mbox format (even though
1594 most programs do not really care). Patch from Alex Miller.
1596 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1597 The latter is generated from $received_to and is only set if the
1598 message has one envelope recipient. SA can use these headers,
1599 obviously out-of-the-box. Patch from Alex Miller.
1601 PH/08 The ${def test on a variable was returning false if the variable's
1602 value was "0", contrary to what the specification has always said!
1603 The result should be true unless the variable is empty.
1605 PH/09 The syntax error of a character other than { following "${if
1606 def:variable_name" (after optional whitespace) was not being diagnosed.
1607 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1608 accidental colon was present, for example, could give incorrect results.
1610 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1611 result is used (not including appendfile, where other changes are about
1614 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1615 This involved changing a lot of size variables from int to off_t. It
1616 should work with maildirs and everything.
1618 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1619 spamd dying while we are connected to it.
1621 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1624 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1625 function for BDB 4.3.
1627 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1629 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1630 the sources for PCRE have been re-arranged and more files are now
1633 PH/15 The code I had for printing potentially long long variables in PH/11
1634 above was not the best (it lost precision). The length of off_t variables
1635 is now inspected at build time, and an appropriate printing format (%ld
1636 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1637 to be "long long int" or "long int". This is needed for the internal
1638 formatting function string_vformat().
1640 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1641 the configuration file to be ":syslog", then the script "guesses" where
1642 the logs files are, rather than using the compiled in default. In our
1643 case the guess is not the same as the compiled default, so the script
1644 suddenly stopped working when I started to use syslog. The patch checks
1645 to see if log_file_path is "". If so, it attempts to read it from exim
1646 with no configuration file to get the compiled in version, before it
1647 falls back to the previous guessing code."
1649 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1650 implementing BATV in an Exim configuration. See NewStuff for the gory
1653 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1654 Makefile that are specific to HP-UX.
1656 PH/18 If the "use_postmaster" option was set for a recipient callout together
1657 with the "random" option, the postmaster address was used as the MAIL
1658 FROM address for the random test, but not for the subsequent recipient
1659 test. It is now used for both.
1661 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1662 patch removes a few documentation additions to RFC 3028, because the
1663 latest draft now contains them. It adds the new en;ascii-case comparator
1664 and a new error check for 8bit text in MIME parts. Comparator and
1665 require names are now matched exactly. I enabled the subaddress
1666 extension, but it is not well tested yet (read: it works for me)."
1668 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1669 rework some of the code of TK/09 above to avoid the hardwired use of
1670 "%lld" and "long long". Replaced the call to snprintf() with a call to
1673 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1674 records point to non-existent hosts", "retry timeout exceeded", and
1675 "retry time not reached for any host after a long failure period".
1677 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1678 experimental DomainKeys support:
1680 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1681 (2) On an error such as an illegally used "control", the wrong name for
1682 the control was given.
1684 These problems did NOT occur unless DomainKeys support was compiled.
1686 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1688 PH/24 Added ${if match_ip condition.
1690 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1691 databases so that it will be absolutely obvious if a crash occurs in the
1692 DB library. This is a regular occurrence (often caused by mis-matched
1695 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1696 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1697 picked up on a user's system that detects such things. There doesn't seem
1698 to be a gcc warning option for this - only an attribute that has to be
1699 put on the function's prototype. It seems that in Fedora Core 4 they have
1700 set this on a number of new functions. No doubt there will be more in due
1703 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1704 specify a transport. However, if an address that was verified by such a
1705 router was the subject of a callout, Exim crashed because it tried to
1706 read the rcpt_include_affixes from the non-existent transport. Now it
1707 just assumes that the setting of that option is false. This bug was
1708 introduced by 4.51/PH/31.
1710 PH/28 Changed -d+all to exclude +memory, because that information is very
1711 rarely of interest, but it makes the output a lot bigger. People tend to
1712 do -d+all out of habit.
1714 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1715 code in os-type was giving problems when libc.so lives in lib64, like on
1718 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1719 aren't the modern standard, and it seems that some systems' include files
1720 don't always have them. Exim was already checking for some of the newer
1721 ones like T_AAAA, and defining it itself. I've added checks for all the
1722 record types that Exim uses.
1724 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1725 not automatically generating a new one, as it is supposed to. This
1726 prevented TLS from working. If the file did exist, but contained invalid
1727 data, a new version was generated, as expected. It was only the case of a
1728 non-existent file that was broken.
1730 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1731 with a change in libdomainkeys > 0.64.
1733 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1734 from DNS. If the selector record carries the flag, it now has
1735 precedence over the domain-wide flag.
1737 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1739 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1740 the use of a port name that isn't defined in /etc/services) occurred, the
1741 message was deferred as in a normal delivery, and thus remained on the
1742 spool, instead of being failed because of the mua_wrapper setting. This
1743 is now fixed, and I tidied up some of the mua_wrapper messages at the
1746 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1747 the messages in a hash of arrays rather than using individual hashes.
1748 This is a bit cleaner and results in dramatic memory savings, albeit
1749 at a slight CPU cost.
1751 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1752 as requested by Marc Sherman.
1754 SC/10 Eximstats: added histograms for user specified patterns as requested
1757 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1759 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1760 fopen() in the content-scanning modules that did not already have it.
1766 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1767 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1769 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1771 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1773 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1774 to test Sieve filters that use "vacation".
1776 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1777 that changes the way the GnuTLS parameters are stored in the cache file.
1778 The new format can be generated externally. For backward compatibility,
1779 if the data in the cache doesn't make sense, Exim assumes it has read an
1780 old-format file, and it generates new data and writes a new file. This
1781 means that you can't go back to an older release without removing the
1784 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1785 work if there are any delivery delays because "one_time" forces the
1786 parent to be marked "delivered", so its unseen clone is never tried
1787 again. For this reason, Exim now forbids the simultaneous setting of
1790 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1791 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1792 entry if you want to know the details. The fix, however, neglected to
1793 consider the case when local delivery batching is involved. The test for
1794 "previously delivered" was not happening when checking to see if an
1795 address could be batched with a previous (undelivered) one; under
1796 certain circumstances this could lead to multiple deliveries to the same
1799 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1800 in its include files, and this causes problems building Exim.
1802 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1803 header_syntax) but Exim was just ignoring anything given after a slash.
1804 In particular, this caused confusion with an attempt to use "verify =
1805 reverse_host_lookup/defer_ok". An error is now given when options are
1806 supplied for verify items that do not have them. (Maybe reverse_host_
1807 lookup should have a defer_ok option, but that's a different point.)
1809 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1810 defined by RFC 821) to 2048, because there were problems with some AUTH
1811 commands, and RFC 1869 says the size should be increased for extended
1812 SMTP commands that take arguments.
1814 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1817 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1818 "unknown" error; now it says that the functionality isn't in the binary.
1820 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1821 an address' error message when a string expansion fails (syntax or
1822 whatever). Otherwise the password may appear in the log. Following change
1823 PH/42 below, there is no longer a chance of it appearing in a bounce
1826 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1828 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1829 of its addresses was ever tried. (Bugzilla bug #2.)
1831 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1832 the result incorrectly in the debug output. (It correctly added a newline
1833 to what was transported.)
1835 TF/01 Added $received_time.
1837 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1838 commented out examples of how to interface to a virus scanner and to
1839 SpamAssassin. Also added commented examples of av_scanner and
1840 spamd_address settings.
1842 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1843 and controls are allowed in which ACLs. There were a couple of minor
1844 errors. Some of the entries in the conditions table (which is a table of
1845 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1846 negation of where the condition IS allowed.
1848 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1850 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1851 header file does not have a version number, so I've had to invent a new
1852 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1853 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1854 radiusclient), but it was contributed by a Radius user.
1856 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1857 files or directories, for OpenSSL.
1859 PH/19 When an Exim process that is running as root has to create an Exim log
1860 file, it does so in a subprocess that runs as exim:exim so as to get the
1861 ownership right at creation (otherwise, other Exim processes might see
1862 the file with the wrong ownership). There was no test for failure of this
1863 fork() call, which would lead to the process getting stuck as it waited
1864 for a non-existent subprocess. Forks do occasionally fail when resources
1865 run out. I reviewed all the other calls to fork(); they all seem to check
1868 PH/20 When checking for unexpected SMTP input at connect time (before writing
1869 the banner), Exim was not dealing correctly with a non-positive return
1870 from the read() function. If the client had disconnected by this time,
1871 the result was a log entry for a synchronization error with an empty
1872 string after "input=" when read() returned zero. If read() returned -1
1873 (an event I could not check), uninitialized data bytes were printed.
1874 There were reports of junk text (parts of files, etc) appearing after
1877 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1879 PH/22 Added support for macro redefinition, and (re)definition in between
1880 driver and ACL definitions.
1882 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1883 forgetting to use the resulting value; it was using the unexpanded value.
1885 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1886 hadn't been configured. The fix is from Juergen Kreileder, who
1887 understands it better than I do:
1889 "Here's what I see happening with three configured cyrus_sasl
1890 authenticators configured (plain, login, cram-md5):
1892 On startup auth_cyrus_sasl_init() gets called for each of these.
1893 This means three calls to sasl_listmech() without a specified mech_list.
1894 => SASL tests which mechs of all available mechs actually work
1895 => three warnings about OTP not working
1896 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1898 With the patch, sasl_listmech() also gets called three times. But now
1899 SASL's mech_list option is set to the server_mech specified in the the
1900 authenticator. Or in other words, the answer from sasl_listmech()
1901 gets limited to just the mech you're testing for (which is different
1903 => the return list contains just 'plain' or 'login', 'cram-md5' or
1904 nothing depending on the value of ob->server_mech.
1906 I've just tested the patch: Authentication still works fine,
1907 unavailable mechs specified in the exim configuration are still
1908 caught, and the auth.log warnings about OTP are gone."
1910 PH/25 When debugging is enabled, the contents of the command line are added
1911 to the debugging output, even when log_selector=+arguments is not
1914 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1915 answer is "GNU", and only if the return is "GNU/something" is the answer
1918 PH/27 $acl_verify_message is now set immediately after the failure of a
1919 verification in an ACL, and so is available in subsequent modifiers. In
1920 particular, the message can be preserved by coding like this:
1922 warn !verify = sender
1923 set acl_m0 = $acl_verify_message
1925 Previously, $acl_verify_message was set only while expanding "message"
1926 and "log_message" when a very denied access.
1928 PH/28 Modified OS/os.c-Linux with
1930 -#ifndef OS_LOAD_AVERAGE
1931 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1933 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1934 nomenclature these days.)
1936 PH/29 Installed patch from the Sieve maintainer that adds the options
1937 sieve_useraddress and sieve_subaddress to the redirect router.
1939 PH/30 In these circumstances:
1940 . Two addresses routed to the same list of hosts;
1941 . First host does not offer TLS;
1942 . First host accepts first address;
1943 . First host gives temporary error to second address;
1944 . Second host offers TLS and a TLS session is established;
1945 . Second host accepts second address.
1946 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1947 and peerdn, if requested) that were in fact used only for the second
1950 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1951 attention to any local part prefix or suffix that was matched by the
1952 router that accepted the address. It now behaves in the same way as it
1953 does for delivery: the affixes are removed from the local part unless
1954 rcpt_include_affixes is set on the transport.
1956 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1957 timeout during the DATA phase of an incoming message.
1959 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1960 applied a patch sanctioned by the Sieve maintainer.
1962 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1963 the uid or gid is negative. A case of a negative gid caused this to be
1964 noticed. The fix allows for either to be negative.
1966 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1967 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1970 PH/36 The change PH/12 above was broken. Fixed it.
1972 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1973 the grounds that routing the same address twice would always produce the
1974 same answer. This might have been true once, but it is certainly no
1975 longer true now. Routing a child address may depend on the previous
1976 routing that produced that child. Some complicated redirection strategies
1977 went wrong when messages had multiple recipients, and made Exim's
1978 behaviour dependent on the order in which the addresses were given.
1980 I have moved the duplicate checking until after the routing is complete.
1981 Exim scans the addresses that are assigned to local and remote
1982 transports, and removes any duplicates. This means that more work will be
1983 done, as duplicates will always all be routed, but duplicates are
1984 presumably rare, so I don't expect this is of any significance.
1986 For deliveries to pipes, files, and autoreplies, the duplicate checking
1987 still happens during the routing process, since they are not going to be
1990 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1991 It corrects a timeout issue with spamd. This is Ian's comment: "The
1992 background is that sometimes spamd either never reads data from a
1993 connection it has accepted, or it never writes response data. The exiscan
1994 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1995 blindly assumes that writes won't block so it may never time out."
1997 PH/39 Allow G after quota size as well as K and M.
1999 PH/40 The value set for $authenticated_id in an authenticator may not contain
2000 binary zeroes or newlines because the value is written to log lines and
2001 to spool files. There was no check on this. Now the value is run through
2002 the string_printing() function so that such characters are converted to
2003 printable escape sequences.
2005 PH/41 $message_linecount is a new variable that contains the total number of
2006 lines in the message. Compare $body_linecount, which is the count for the
2009 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2010 bounce and delay warning messages, except in certain special cases, which
2013 (a) An SMTP error message from a remote host;
2014 (b) A message specified in a :fail: redirection;
2015 (c) A message specified in a "fail" command in a system filter;
2016 (d) A message specified in a FAIL return from the queryprogram router;
2017 (e) A message specified by the cannot_route_message router option.
2019 In these cases only, Exim does include the error details in bounce and
2020 warning messages. There are also a few cases where bland messages such
2021 as "unrouteable address" or "local delivery error" are given.
2023 PH/43 $value is now also set for the "else" part of a ${run expansion.
2025 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2026 being worked on, but at least Exim now implements the latest version to
2029 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2030 process to complete was treated as a delivery failure, a timeout while
2031 writing the message to the pipe was logged, but erroneously treated as a
2032 successful delivery. Such timeouts include transport filter timeouts. For
2033 consistency with the overall process timeout, these timeouts are now
2034 treated as errors, giving rise to delivery failures by default. However,
2035 there is now a new Boolean option for the pipe transport called
2036 timeout_defer, which, if set TRUE, converts the failures into defers for
2037 both kinds of timeout. A transport filter timeout is now identified in
2040 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2041 systems where "make" and "gmake" are different, calling "gmake" at top
2042 level broke things. I've arranged for the value of $(MAKE) to be passed
2043 from the Makefile to this script so that it can call the same version of
2047 A note about Exim versions 4.44 and 4.50
2048 ----------------------------------------
2050 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2051 changes of various kinds. As a consequence, a big documentation update was
2052 needed. This delayed the release for rather longer than seemed good, especially
2053 in the light of a couple of (minor) security issues. Therefore, the changes
2054 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2055 release. So 4.44 and 4.50 are in effect two different branches that both start
2058 I have left the 4.50 change log unchanged; it contains all the changes since
2059 4.43. The change log for 4.44 is below; many of its items are identical to
2060 those for 4.50. This seems to be the most sensible way to preserve the
2061 historical information.
2067 1. Minor wording change to the doc/README.SIEVE file.
2069 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2070 computation of the current number of files was incorrect.
2072 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2073 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2076 4. Give more explanation in the error message when the command for a transport
2077 filter fails to execute.
2079 5. There are several places where Exim runs a non-Exim command in a
2080 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2081 was being done only for the command run by the queryprogram router. It is
2082 now done for all such subprocesses. The other cases are: ${run, transport
2083 filters, and the commands run by the lmtp and pipe transports.
2085 6. Added CONFIGURE_GROUP build-time option.
2087 7. Some older OS have a limit of 256 on the maximum number of file
2088 descriptors. Exim was using setrlimit() to set 1000 as a large value
2089 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2090 systems. I've change it so that if it can't get 1000, it tries for 256.
2092 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2093 was an oversight, and furthermore, ever since the addition of extra
2094 controls (e.g. 4.43/32), the checks on when to allow different forms of
2095 "control" were broken. There should now be diagnostics for all cases when a
2096 control that does not make sense is encountered.
2098 9. Added the /retain_sender option to "control=submission".
2100 10. $recipients is now available in the predata ACL (oversight).
2102 11. Tidy the search cache before the fork to do a delivery from a message
2103 received from the command line. Otherwise the child will trigger a lookup
2104 failure and thereby defer the delivery if it tries to use (for example) a
2105 cached ldap connection that the parent has called unbind on.
2107 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2108 of $address_data from the recipient verification was clobbered by the
2109 sender verification.
2111 13. The value of address_data from a sender verification is now available in
2112 $sender_address_data in subsequent conditions in the ACL statement.
2114 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2116 15. Added a new option "connect=<time>" to callout options, to set a different
2119 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2120 was its contents. (It was OK if the option was not defined at all.)
2122 17. A "Completed" log line is now written for messages that are removed from
2123 the spool by the -Mrm option.
2125 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2126 information about exactly what failed.
2128 19. Added -dd to debug only the daemon process.
2130 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2131 handles timeouts, both on the server side and network timeouts. Renamed the
2132 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2134 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2135 It is now set to "smtps".
2137 22. $host_address is now set to the target address during the checking of
2138 ignore_target_hosts.
2140 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2141 being passed; this would have caused $sender_host_name to have been used if
2142 matching the list had actually called for a host name (not very likely,
2143 since this list is usually IP addresses). A host name is now passed as
2146 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2147 code that specifies a non-restarting handler (typically sigaction() in
2148 modern systems) in an attempt to fix a rare and obscure crash bug.
2150 25. Narrowed the window for a race in the daemon that could cause it to ignore
2151 SIGCHLD signals. This is not a major problem, because they are used only to
2152 wake it up if nothing else does.
2154 26. A malformed maildirsize file could cause Exim to calculate negative values
2155 for the mailbox size or file count. Odd effects could occur as a result.
2156 The maildirsize information is now recalculated if the size or filecount
2159 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2160 support for a long time. Removed HAVE_SYS_VFS_H.
2162 28. Installed the latest version of exipick from John Jetmore.
2164 29. In an address list, if the pattern was not a regular expression, an empty
2165 subject address (from a bounce message) matched only if the pattern was an
2166 empty string. Non-empty patterns were not even tested. This was the wrong
2167 because it is perfectly reasonable to use an empty address as part of a
2168 database query. An empty address is now tested by patterns that are
2169 lookups. However, all the other forms of pattern expect the subject to
2170 contain a local part and a domain, and therefore, for them, an empty
2171 address still always fails if the pattern is not itself empty.
2173 30. Exim went into a mad DNS loop when attempting to do a callout where the
2174 host was specified on an smtp transport, and looking it up yielded more
2175 than one IP address.
2177 31. Re-factored the code for checking spool and log partition space into a
2178 function that finds that data and another that does the check. The former
2179 is then used to implement four new variables: $spool_space, $log_space,
2180 $spool_inodes, and $log_inodes.
2182 32. The RFC2047 encoding function was originally intended for short strings
2183 such as real names; it was not keeping to the 75-character limit for
2184 encoded words that the RFC imposes. It now respects the limit, and
2185 generates multiple encoded words if necessary. To be on the safe side, I
2186 have increased the buffer size for the ${rfc2047: expansion operator from
2189 33. It is now permitted to omit both strings after an "if" condition; if the
2190 condition is true, the result is "true". As before, when the second string
2191 is omitted, a false condition yields an empty string. This makes it less
2192 cumbersome to write custom ACL and router conditions.
2194 34. Failure to deliver a bounce message always caused it to be frozen, even if
2195 there was an errors_to setting on the router. The errors_to setting is now
2198 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2199 canonical form (fully expanded) before being placed in
2200 $sender_host_address.
2202 36. The table in the code that translates DNS record types into text (T_A to
2203 "A" for instance) was missing entries for NS and CNAME. It is just possible
2204 that this could have caused confusion if both these types were looked up
2205 for the same domain, because the text type is used as part of Exim's
2206 per-process caching. But the chance of anyone hitting this buglet seems
2209 37. The dnsdb lookup has been extended in a number of ways.
2211 (1) There is a new type, "zns", which walks up the domain tree until it
2212 finds some nameserver records. It should be used with care.
2214 (2) There is a new type, "mxh", which is like "mx" except that it returns
2215 just the host names, not the priorities.
2217 (3) It is now possible to give a list of domains (or IP addresses) to be
2218 looked up. The behaviour when one of the lookups defers can be
2219 controlled by a keyword.
2221 (4) It is now possible to specify the separator character for use when
2222 multiple records are returned.
2224 38. The dnslists ACL condition has been extended: it is now possible to supply
2225 a list of IP addresses and/or domains to be looked up in a particular DNS
2228 39. Added log_selector=+queue_time_overall.
2230 40. When running the queue in the test harness, wait just a tad after forking a
2231 delivery process, to get repeatability of debugging output.
2233 41. Include certificate and key file names in error message when GnuTLS fails
2234 to set them up, because the GnuTLS error message doesn't include the name
2235 of the failing file when there is a problem reading it.
2237 42. Allow both -bf and -bF in the same test run.
2239 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2241 44. The "Exiscan patch" is now merged into the mainline Exim source.
2243 45. Sometimes the final signoff response after QUIT could fail to get
2244 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2245 < 0 before doing a fflush(). This bug looks as though it goes back to the
2246 introduction of TLS in release 3.20, but "sometimes" must have been rare
2247 because the tests only now provoked it.
2249 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2250 (this can affect the format of dates).
2252 47. exim_tidydb, when checking for the continued existence of a message for
2253 which it has found a message-specific retry record, was not finding
2254 messages that were in split spool directories. Consequently, it was
2255 deleting retry records that should have stayed in existence.
2257 48. Steve fixed some bugs in eximstats.
2259 49. The SPA authentication driver was not abandoning authentication and moving
2260 on to the next authenticator when an expansion was forced to fail,
2261 contradicting the general specification for all authenticators. Instead it
2262 was generating a temporary error. It now behaves as specified.
2264 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2265 (the order specifies the preference for clients). The order is now AES256,
2266 AES128, 3DES, ARCFOUR128.
2268 51. Small patch to Sieve code - explicitly set From: when generating an
2271 52. Exim crashed if a remote delivery caused a very long error message to be
2272 recorded - for instance if somebody sent an entire SpamAssassin report back
2273 as a large number of 550 error lines. This bug was coincidentally fixed by
2274 increasing the size of one of Exim's internal buffers (big_buffer) that
2275 happened as part of the Exiscan merge. However, to be on the safe side, I
2276 have made the code more robust (and fixed the comments that describe what
2279 53. Now that there can be additional text after "Completed" in log lines (if
2280 the queue_time_overall log selector is set), a one-byte patch to exigrep
2281 was needed to allow it to recognize "Completed" as not the last thing in
2284 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2285 patch that reportedly fixes this has been added. I am not expert enough to
2286 create a test for it. This is what the patch creator wrote:
2288 "I found a little strange behaviour of ldap code when working with
2289 Windows 2003 AD Domain, where users was placed in more than one
2290 Organization Units. When I tried to give exim partial DN, the exit code
2291 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2292 But simultaneously result of request was absolutely normal ldap result,
2293 so I produce this patch..."
2295 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2296 REFERENCE, so I have modified the code to exclude the patch when that macro
2299 55. Some experimental protocols are using DNS PTR records for new purposes. The
2300 keys for these records are domain names, not reversed IP addresses. The
2301 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2302 leaves it alone. Component reversal etc. now happens only for IP addresses.
2305 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2307 57. Double the size of the debug message buffer (to 2048) so that more of very
2308 long debug lines gets shown.
2310 58. The exicyclog utility now does better if the number of log files to keep
2311 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2313 59. Two changes related to the smtp_active_hostname option:
2315 (1) $smtp_active_hostname is now available as a variable.
2316 (2) The default for smtp_banner uses $smtp_active_hostname instead
2317 of $primary_hostname.
2319 60. The host_aton() function is supposed to be passed a string that is known
2320 to be a valid IP address. However, in the case of IPv6 addresses, it was
2321 not checking this. This is a hostage to fortune. Exim now panics and dies
2322 if the condition is not met. A case was found where this could be provoked
2323 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2324 components; fortuitously, this particular loophole had already been fixed
2325 by change 4.50/55 above.
2327 If there are any other similar loopholes, the new check in host_aton()
2328 itself should stop them being exploited. The report I received stated that
2329 data on the command line could provoke the exploit when Exim was running as
2330 exim, but did not say which command line option was involved. All I could
2331 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2332 running as the user.
2335 61. There was a buffer overflow vulnerability in the SPA authentication code
2336 (which came originally from the Samba project). I have added a test to the
2337 spa_base64_to_bits() function which I hope fixes it.
2340 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2341 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2343 63. The daemon start-up calls getloadavg() while still root for those OS that
2344 need the first call to be done as root, but it missed one case: when
2345 deliver_queue_load_max is set with deliver_drop_privilege. This is
2346 necessary for the benefit of the queue runner, because there is no re-exec
2347 when deliver_drop_privilege is set.
2349 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2350 This has been fixed.
2352 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2353 was in use, was not putting the data itself into the right store pool;
2354 consequently, it could be overwritten for a subsequent message in the same
2355 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2358 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2360 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2361 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2362 the function were treating the return as a boolean value, which happened to
2363 work because 0=false and not-0=true, but is not correct code.
2365 68. The host_aton() function was not handling scoped IPv6 addresses (those
2366 with, for example, "%eth0" on the end) correctly.
2368 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2369 negated items (that is, ~something) in unsigned ints. Some compilers
2370 apparently mutter when there is no cast.
2372 70. If an address verification called from an ACL failed, and did not produce a
2373 user-specific message (i.e. there was only a "system" message), nothing was
2374 put in $acl_verify_message. In this situation, it now puts the system
2377 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2378 session; change 4.31/43 added the unwanted input to the log line - except
2379 that it did not do this in the start of session case. It now does.
2381 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2382 This is wrong and can cause the other end to generate a synchronization
2383 error if it is another Exim or anything else that does the synchronization
2384 check. A QUIT command is no longer sent after a timeout.
2386 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2387 during host lookups.
2389 74. The defer_ok option of callout verification was not working if it was used
2390 when verifying addresses in header lines, that is, for this case:
2392 verify = header_sender/callout=defer_ok
2394 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2395 those file descriptors could be used for SMTP connections. If anything
2396 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2397 could be sent to the SMTP client, causing chaos. The daemon now opens
2398 stdin, stdout, and stderr to /dev/null when it puts itself into the
2401 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2402 log by default. The user can override this with suitable Perl magic.
2404 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2405 the log message when discard triggers, was not working for the DATA ACL or
2406 for the non-SMTP ACL.
2408 78. Error message wording change in sieve.c.
2410 79. If smtp_accept_max_per_host was set, the number of connections could be
2411 restricted to fewer than expected, because the daemon was trying to set up
2412 a new connection before checking whether the processes handling previous
2413 connections had finished. The check for completed processes is now done
2414 earlier. On busy systems, this bug wouldn't be noticed because something
2415 else would have woken the daemon, and it would have reaped the completed
2418 80. If a message was submitted locally by a user whose login name contained one
2419 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2420 It caused a spool format error. I have fixed the spool reading code. A
2421 related problem was that the "from" clause in the Received: line became
2422 illegal because of the space(s). It is now covered by ${quote_local_part.
2424 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2427 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2429 83. Patch from Sieve maintainer to fix unterminated string problem in
2430 "vacation" handling.
2432 84. Some minor changes to the Linux configuration files to help with other
2433 OS variants using glibc.
2435 85. One more patch for Sieve to update vacation handling to latest spec.
2438 ----------------------------------------------------
2439 See the note above about the 4.44 and 4.50 releases.
2440 ----------------------------------------------------
2446 1. Change 4.43/35 introduced a bug that caused file counts to be
2447 incorrectly computed when quota_filecount was set in an appendfile
2450 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2451 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2454 3. Give more explanation in the error message when the command for a transport
2455 filter fails to execute.
2457 4. There are several places where Exim runs a non-Exim command in a
2458 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2459 was being done only for the command run by the queryprogram router. It is
2460 now done for all such subprocesses. The other cases are: ${run, transport
2461 filters, and the commands run by the lmtp and pipe transports.
2463 5. Some older OS have a limit of 256 on the maximum number of file
2464 descriptors. Exim was using setrlimit() to set 1000 as a large value
2465 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2466 systems. I've change it so that if it can't get 1000, it tries for 256.
2468 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2469 was an oversight, and furthermore, ever since the addition of extra
2470 controls (e.g. 4.43/32), the checks on when to allow different forms of
2471 "control" were broken. There should now be diagnostics for all cases when a
2472 control that does not make sense is encountered.
2474 7. $recipients is now available in the predata ACL (oversight).
2476 8. Tidy the search cache before the fork to do a delivery from a message
2477 received from the command line. Otherwise the child will trigger a lookup
2478 failure and thereby defer the delivery if it tries to use (for example) a
2479 cached ldap connection that the parent has called unbind on.
2481 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2482 of $address_data from the recipient verification was clobbered by the
2483 sender verification.
2485 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2486 was its contents. (It was OK if the option was not defined at all.)
2488 11. A "Completed" log line is now written for messages that are removed from
2489 the spool by the -Mrm option.
2491 12. $host_address is now set to the target address during the checking of
2492 ignore_target_hosts.
2494 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2495 being passed; this would have caused $sender_host_name to have been used if
2496 matching the list had actually called for a host name (not very likely,
2497 since this list is usually IP addresses). A host name is now passed as
2500 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2501 code that specifies a non-restarting handler (typically sigaction() in
2502 modern systems) in an attempt to fix a rare and obscure crash bug.
2504 15. Narrowed the window for a race in the daemon that could cause it to ignore
2505 SIGCHLD signals. This is not a major problem, because they are used only to
2506 wake it up if nothing else does.
2508 16. A malformed maildirsize file could cause Exim to calculate negative values
2509 for the mailbox size or file count. Odd effects could occur as a result.
2510 The maildirsize information is now recalculated if the size or filecount
2513 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2514 support for a long time. Removed HAVE_SYS_VFS_H.
2516 18. Updated exipick to current release from John Jetmore.
2518 19. Allow an empty sender to be matched against a lookup in an address list.
2519 Previously the only cases considered were a regular expression, or an
2522 20. Exim went into a mad DNS lookup loop when doing a callout where the
2523 host was specified on the transport, if the DNS lookup yielded more than
2526 21. The RFC2047 encoding function was originally intended for short strings
2527 such as real names; it was not keeping to the 75-character limit for
2528 encoded words that the RFC imposes. It now respects the limit, and
2529 generates multiple encoded words if necessary. To be on the safe side, I
2530 have increased the buffer size for the ${rfc2047: expansion operator from
2533 22. Failure to deliver a bounce message always caused it to be frozen, even if
2534 there was an errors_to setting on the router. The errors_to setting is now
2537 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2538 canonical form (fully expanded) before being placed in
2539 $sender_host_address.
2541 24. Updated eximstats to version 1.33
2543 25. Include certificate and key file names in error message when GnuTLS fails
2544 to set them up, because the GnuTLS error message doesn't include the name
2545 of the failing file when there is a problem reading it.
2547 26. Expand error message when OpenSSL has problems setting up cert/key files.
2550 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2551 (this can affect the format of dates).
2553 28. exim_tidydb, when checking for the continued existence of a message for
2554 which it has found a message-specific retry record, was not finding
2555 messages that were in split spool directories. Consequently, it was
2556 deleting retry records that should have stayed in existence.
2558 29. eximstats updated to version 1.35
2559 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2560 1.35 - bugfix such that pie charts by volume are generated correctly
2562 30. The SPA authentication driver was not abandoning authentication and moving
2563 on to the next authenticator when an expansion was forced to fail,
2564 contradicting the general specification for all authenticators. Instead it
2565 was generating a temporary error. It now behaves as specified.
2567 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2568 (the order specifies the preference for clients). The order is now AES256,
2569 AES128, 3DES, ARCFOUR128.
2571 31. Small patch to Sieve code - explicitly set From: when generating an
2574 32. Exim crashed if a remote delivery caused a very long error message to be
2575 recorded - for instance if somebody sent an entire SpamAssassin report back
2576 as a large number of 550 error lines. This bug was coincidentally fixed by
2577 increasing the size of one of Exim's internal buffers (big_buffer) that
2578 happened as part of the Exiscan merge. However, to be on the safe side, I
2579 have made the code more robust (and fixed the comments that describe what
2582 33. Some experimental protocols are using DNS PTR records for new purposes. The
2583 keys for these records are domain names, not reversed IP addresses. The
2584 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2585 leaves it alone. Component reversal etc. now happens only for IP addresses.
2588 34. The host_aton() function is supposed to be passed a string that is known
2589 to be a valid IP address. However, in the case of IPv6 addresses, it was
2590 not checking this. This is a hostage to fortune. Exim now panics and dies
2591 if the condition is not met. A case was found where this could be provoked
2592 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2593 components; fortuitously, this particular loophole had already been fixed
2594 by change 4.50/55 or 4.44/33 above.
2596 If there are any other similar loopholes, the new check in host_aton()
2597 itself should stop them being exploited. The report I received stated that
2598 data on the command line could provoke the exploit when Exim was running as
2599 exim, but did not say which command line option was involved. All I could
2600 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2601 running as the user.
2604 35. There was a buffer overflow vulnerability in the SPA authentication code
2605 (which came originally from the Samba project). I have added a test to the
2606 spa_base64_to_bits() function which I hope fixes it.
2609 36. The daemon start-up calls getloadavg() while still root for those OS that
2610 need the first call to be done as root, but it missed one case: when
2611 deliver_queue_load_max is set with deliver_drop_privilege. This is
2612 necessary for the benefit of the queue runner, because there is no re-exec
2613 when deliver_drop_privilege is set.
2615 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2616 was in use, was not putting the data itself into the right store pool;
2617 consequently, it could be overwritten for a subsequent message in the same
2618 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2621 38. Sometimes the final signoff response after QUIT could fail to get
2622 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2623 < 0 before doing a fflush(). This bug looks as though it goes back to the
2624 introduction of TLS in release 3.20, but "sometimes" must have been rare
2625 because the tests only now provoked it.
2631 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2632 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2633 Now it returns an integer. A number of calls were still expecting a T/F
2634 return. Fortuitously, in all cases, the tests worked in OK situations,
2635 which is the norm. However, things would have gone wrong on any write
2636 failures on the smtp file descriptor. This function is used when sending
2637 messages over SMTP and also when doing verify callouts.
2639 2. When Exim is called to do synchronous delivery of a locally submitted
2640 message (the -odf or -odi options), it no longer closes stderr before doing
2643 3. Implemented the mua_wrapper option.
2645 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2647 5. Implemented the functions header_remove(), header_testname(),
2648 header_add_at_position(), and receive_remove_recipient(), and exported them
2651 6. If an ACL "warn" statement specified the addition of headers, Exim already
2652 inserted X-ACL-Warn: at the start if there was no header name. However, it
2653 was not making this test for the second and subsequent header lines if
2654 there were newlines in the string. This meant that an invalid header could
2655 be inserted if Exim was badly configured.
2657 7. Allow an ACL "warn" statement to add header lines at the start or after all
2658 the Received: headers, as well as at the end.
2660 8. Added the rcpt_4xx retry error code.
2662 9. Added postmaster_mailfrom=xxx to callout verification option.
2664 10. Added mailfrom=xxxx to the callout verification option, for verify=
2667 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2668 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2670 12. Inserted some casts to stop certain compilers warning when using pointer
2671 differences as field lengths or precisions in printf-type calls (mostly
2672 affecting debugging statements).
2674 13. Added optional readline() support for -be (dynamically loaded).
2676 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2677 same clock tick as a message's arrival, so that its received time was the
2678 same as the "first fail" time on the retry record, and that message
2679 remained on the queue past the ultimate address timeout, every queue runner
2680 would try a delivery (because it was past the ultimate address timeout) but
2681 after another failure, the ultimate address timeout, which should have then
2682 bounced the address, did not kick in. This was a "< instead of <=" error;
2683 in most cases the first failure would have been in the next clock tick
2684 after the received time, and all would be well.
2686 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2687 being recognized when the domain list was tested by the match_domain
2688 condition in an expansion string.
2690 16. Added the ${str2b64: operator.
2692 17. Exim was always calling setrlimit() to set a large limit for the number of
2693 processes, without checking whether the existing limit was already
2694 adequate. (It did check for the limit on file descriptors.) Furthermore,
2695 errors from getrlimit() and setrlimit() were being ignored. Now they are
2696 logged to the main and panic logs, but Exim does carry on, to try to do its
2697 job under whatever limits there are.
2699 18. Imported PCRE 5.0.
2701 19. Trivial typo in log message " temporarily refused connection" (the leading
2704 20. If the log selector return_path_on_delivery was set and an address was
2705 redirected to /dev/null, the delivery process crashed because it assumed
2706 that a return path would always be set for a "successful" delivery. In this
2707 case, the whole delivery is bypassed as an optimization, and therefore no
2710 21. Internal re-arrangement: the function for sending a challenge and reading
2711 a response while authentication was assuming a zero-terminated challenge
2712 string. It's now changed to take a pointer and a length, to allow for
2713 binary data in such strings.
2715 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2717 23. Exim was not respecting finduser_retries when seeking the login of the
2718 uid under which it was called; it was always trying 10 times. (The default
2719 setting of finduser_retries is zero.) Also, it was sleeping after the final
2720 failure, which is pointless.
2722 24. Implemented tls_on_connect_ports.
2724 25. Implemented acl_smtp_predata.
2726 26. If the domain in control=submission is set empty, Exim assumes that the
2727 authenticated id is a complete email address when it generates From: or
2728 Sender: header lines.
2730 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2731 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2732 chown and chgrp in /bin and hostname in /usr/bin.
2734 28. Exim was keeping the "process log" file open after each use, just as it
2735 does for the main log. This opens the possibility of it remaining open for
2736 long periods when the USR1 signal hits a daemon. Occasional processlog
2737 errors were reported, that could have been caused by this. Anyway, it seems
2738 much more sensible not to leave this file open at all, so that is what now
2741 29. The long-running daemon process does not normally write to the log once it
2742 has entered its main loop, and it closes the log before doing so. This is
2743 so that log files can straightforwardly be renamed and moved. However,
2744 there are a couple of unusual error situations where the daemon does write
2745 log entries, and I had neglected to close the log afterwards.
2747 30. The text of an SMTP error response that was received during a remote
2748 delivery was being truncated at 512 bytes. This is too short for some of
2749 the long messages that one sometimes sees. I've increased the limit to
2752 31. It is now possible to make retry rules that apply only when a message has a
2753 specific sender, in particular, an empty sender.
2755 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2756 it possible to be selective about when SMTP synchronization is enforced.
2758 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2760 32. Implemented hosts_connection_nolog.
2762 33. Added an ACL for QUIT.
2764 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2767 35. Added mailbox_size and mailbox_filecount to appendfile.
2769 36. Added control = no_multiline_responses to ACLs.
2771 37. There was a bug in the logic of the code that waits for the clock to tick
2772 in the case where the clock went backwards by a substantial amount such
2773 that the microsecond fraction of "now" was more than the microsecond
2774 fraction of "then" (but the whole seconds number was less).
2776 38. Added support for the libradius Radius client library this is found on
2777 FreeBSD (previously only the radiusclient library was supported).
2783 1. When certain lookups returned multiple values in the form name=value, the
2784 quoting of the values was not always being done properly. Specifically:
2785 (a) If the value started with a double quote, but contained no whitespace,
2787 (b) If the value contained whitespace other than a space character (i.e.
2788 tabs or newlines or carriage returns) it was not quoted.
2789 This fix has been applied to the mysql and pgsql lookups by writing a
2790 separate quoting function and calling it from the lookup code. The fix
2791 should probably also be applied to nisplus, ibase and oracle lookups, but
2792 since I cannot test any of those, I have not disturbed their existing code.
2794 2. A hit in the callout cache for a specific address caused a log line with no
2795 reason for rejecting RCPT. Now it says "Previous (cached) callout
2796 verification failure".
2798 3. There was an off-by-one bug in the queryprogram router. An over-long
2799 return line was truncated at 256 instead of 255 characters, thereby
2800 overflowing its buffer with the terminating zero. As well as fixing this, I
2801 have increased the buffer size to 1024 (and made a note to document this).
2803 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2804 when Exim is waiting for an SMTP response from a remote server, Exim
2805 restarts its select() call on the socket, thereby resetting its timeout.
2806 This is not a problem when such interrupts are rare. Somebody set up a cron
2807 job to run exiwhat every 2 minutes, which is less than the normal select()
2808 timeout (5 or 10 minutes). This meant that the select() timeout never
2809 kicked in because it was always reset. I have fixed this by comparing the
2810 time when an interrupt arrives with the time at the start of the first call
2811 to select(). If more time than the timeout has elapsed, the interrupt is
2812 treated as a timeout.
2814 5. Some internal re-factoring in preparation for the addition of Sieve
2815 extensions (by MH). In particular, the "personal" test is moved to a
2816 separate function, and given an option for scanning Cc: and Bcc: (which is
2817 not set for Exim filters).
2819 6. When Exim created an email address using the login of the caller as the
2820 local part (e.g. when creating a From: or Sender: header line), it was not
2821 quoting the local part when it contained special characters such as @.
2823 7. Installed new OpenBSD configuration files.
2825 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2826 try to make them clearer.
2828 9. Callout options, other than the timeout value, were being ignored when
2829 verifying sender addresses in header lines. For example, when using
2831 verify = header_sender/callout=no_cache
2833 the cache was (incorrectly) being used.
2835 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2836 only the Cygwin environment.
2838 11. When return_path_on_delivery was set as a log selector, if different remote
2839 addresses in the same message used different return paths and parallel
2840 remote delivery occurred, the wrong values would sometimes be logged.
2841 (Whenever a remote delivery process finished, the return path value from
2842 the most recently started remote delivery process was logged.)
2844 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2845 lines when AUTH and/or TLS are in use. This is the "received protocol"
2846 field. Exim used to use "asmtp" for authenticated SMTP, without any
2847 indication (in the protocol name) for TLS use. Now it follows the RFC and
2848 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2849 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2850 names appear in log lines as well as in Received: header lines.
2852 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2853 extensions, and comparison tests, and to fix some bugs.
2855 14. Changes to the "personal" filter test:
2857 (1) The test was buggy in that it was just doing the equivalent of
2858 "contains" tests on header lines. For example, if a user's address was
2859 anne@some.where, the "personal" test would incorrectly be true for
2861 To: susanne@some.where
2863 This test is now done by extracting each address from the header in turn,
2864 and checking the entire address. Other tests that are part of "personal"
2865 are now done using regular expressions (for example, to check local parts
2866 of addresses in From: header lines).
2868 (2) The list of non-personal local parts in From: addresses has been
2869 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2870 taken from the Sieve specification recommendations.
2872 (3) If the message contains any header line starting with "List-" it is
2873 treated as non-personal.
2875 (4) The test for "circular" in the Subject: header line has been removed
2876 because it now seems ill-conceived.
2878 15. Minor typos in src/EDITME comments corrected.
2880 16. Installed latest exipick from John Jetmore.
2882 17. If headers_add on a router specified a text string that was too long for
2883 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2884 of string_sprintf() is now avoided.
2886 18. $message_body_size was not set (it was always zero) when running the DATA
2887 ACL and the local_scan() function.
2889 19. For the "mail" command in an Exim filter, no default was being set for
2890 the once_repeat time, causing a random time value to be used if "once" was
2891 specified. (If the value happened to be <= 0, no repeat happened.) The
2892 default is now 0s, meaning "never repeat". The "vacation" command was OK
2893 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2894 (I found it when inspecting the code).
2896 20. There is now an overall timeout for performing a callout verification. It
2897 defaults to 4 times the callout timeout, which applies to individual SMTP
2898 commands during the callout. The overall timeout applies when there is more
2899 than one host that can be tried. The timeout is checked before trying the
2900 next host. This prevents very long delays if there are a large number of
2901 hosts and all are timing out (e.g. when the network connections are timing
2902 out). The value of the overall timeout can be changed by specifying an
2903 additional sub-option for "callout", called "maxwait". For example:
2905 verify = sender/callout=5s,maxwait=20s
2907 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2908 to the end before writing, but this should make it even safer).
2910 22. Exim was forgetting that it had advertised PIPELINING for the second and
2911 subsequent messages on an SMTP connection. It was also not resetting its
2912 memory on STARTTLS and an internal HELO.
2914 23. When Exim logs an SMTP synchronization error within a session, it now
2915 records whether PIPELINING has been advertised or not.
2917 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2918 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2919 rather than long int.
2921 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2923 26. Added the never_mail option to autoreply.
2929 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2930 crash if the getsockname() call failed; this can happen if a connection is
2931 closed very soon after it is established. The problem was simply in the
2932 order in which certain operations were done, causing Exim to try to write
2933 to the SMTP stream before it had set up the file descriptor. The bug has
2934 been fixed by making things happen in the correct order.
2940 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2941 before the connection was closed, thus losing the rejection response.
2943 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2944 some early Solaris releases, but causes trouble in current releases where
2945 socklen_t is defined.
2947 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2950 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2953 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2954 that releasing the top of it at the end releases what was used for sub-
2955 expansions (unless the block got too big). However, discard this block if
2956 the first thing is a variable or header, so that we can use its block when
2957 it is dynamic (useful for very large $message_headers, for example).
2959 6. Lookups now cache *every* query, not just the most recent. A new, separate
2960 store pool is used for this. It can be recovered when all lookup caches are
2961 flushed. Lookups now release memory at the end of their result strings.
2962 This has involved some general refactoring of the lookup sources.
2964 7. Some code has been added to the store_xxx() functions to reduce the amount
2965 of flapping under certain conditions.
2967 8. log_incoming_interface used to affect only the <= reception log lines. Now
2968 it causes the local interface and port to be added to several more SMTP log
2969 lines, for example "SMTP connection from", and rejection lines.
2971 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2973 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2975 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2976 could be overwritten at the end of the current message (or the start of a
2977 new message if it was set in a HELO ACL). The value is now preserved for
2978 the duration of the SMTP connection.
2980 12. If a transport had a headers_rewrite setting, and a matching header line
2981 contained an unqualified address, that address was qualified, even if it
2982 did not match any rewriting rules. The underlying bug was that the values
2983 of the flags that permit the existence of unqualified sender and recipient
2984 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2985 non-local messages, and by -bnq for local messages) were not being
2986 preserved with the message after it was received.
2988 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2989 "next input=" as part of the text comprising the host identity instead of
2990 the correct text. The code was using the same buffer for two different
2991 strings. However, depending on which order the printing function evaluated
2992 its arguments, the bug did not always show up. Under Linux, for example, my
2993 test suite worked just fine.
2995 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2996 doesn't work with some older versions of Perl. It has been changed to "my",
2997 which in any case is probably the better facility to use.
2999 15. A really picky compiler found some instances of statements for creating
3000 error messages that either had too many or two few arguments for the format
3003 16. The size of the buffer for calls to the DNS resolver has been increased
3004 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3005 for addresses that have a lot of PTR records. This alleviates a problem; it
3006 does not fully solve it.
3008 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3009 buffer now truncates the list and logs the incident, which is the same
3010 action as happens when Exim is looking up a host name and its aliases.
3011 Previously in this situation something unpredictable would happen;
3012 sometimes it was "internal error: store_reset failed".
3014 18. If a server dropped the connection unexpectedly when an Exim client was
3015 using GnuTLS and trying to read a response, the client delivery process
3016 crashed while trying to generate an error log message.
3018 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3019 string, for example:
3021 warn message = H1: something\nH2: something
3023 the text was added as a single header line from Exim's point of view
3024 though it ended up OK in the delivered message. However, searching for the
3025 second and subsequent header lines using $h_h2: did not work. This has been
3026 fixed. Similarly, if a system filter added multiple headers in this way,
3027 the routers could not see them.
3029 20. Expanded the error message when iplsearch is called with an invalid key to
3030 suggest using net-iplsearch in a host list.
3032 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3033 ACLs are no longer actually imposed (and a message to that effect is
3036 22. If a "gecos" field in a passwd entry contained escaped characters, in
3037 particular, if it contained a \" sequence, Exim got it wrong when building
3038 a From: or a Sender: header from that name. A second bug also caused
3039 incorrect handling when an unquoted " was present following a character
3040 that needed quoting.
3042 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3043 was not being matched caselessly.
3045 24. Arranged for all hyphens in the exim.8 source to be escaped with
3048 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3049 properly. Recipient callout cache records were still being keyed to include
3050 the sender, even when use_sender was set false. This led to far more
3051 callouts that were necessary. The sender is no longer included in the key
3052 when use_sender is false.
3054 26. Added "control = submission" modifier to ACLs.
3056 27. Added the ${base62d: operator to decode base 62 numbers.
3058 28. dnsdb lookups can now access SRV records.
3060 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3061 the configuration file.
3063 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3064 (-v) mode. This makes the output for a verbose queue run more intelligible.
3066 31. Added a use_postmaster feature to recipient callouts.
3068 32. Added the $body_zerocount variable, containing the number of binary zero
3069 bytes in the message body.
3071 33. The time of last modification of the "new" subdirectory is now used as the
3072 "mailbox time last read" when there is a quota error for a maildir
3075 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3077 35. Added +ignore_unknown as a special item in host lists.
3079 36. Code for decoding IPv6 addresses in host lists is now included, even if
3080 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3081 address was recognized as an IP address, but was then not correctly decoded
3082 into binary, causing unexpected and incorrect effects when compared with
3089 1. Very minor rewording of debugging text in manualroute to say "list of
3090 hosts" instead of "hostlist".
3092 2. If verify=header_syntax was set, and a header line with an unqualified
3093 address (no domain) and a large number of spaces between the end of the
3094 name and the colon was received, the reception process suffered a buffer
3095 overflow, and (when I tested it) crashed. This was caused by some obsolete
3096 code that should have been removed. The fix is to remove it!
3098 3. When running in the test harness, delay a bit after writing a bounce
3099 message to get a bit more predictability in the log output.
3101 4. Added a call to search_tidyup() just before forking a reception process. In
3102 theory, someone could use a lookup in the expansion of smtp_accept_max_
3103 per_host which, without the tidyup, could leave open a database connection.
3105 5. Added the variables $recipient_data and $sender_data which get set from a
3106 lookup success in an ACL "recipients" or "senders" condition, or a router
3107 "senders" option, similar to $domain_data and $local_part_data.
3109 6. Moved the writing of debug_print from before to after the "senders" test
3112 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3113 problems for message scanning, either using a data ACL, or using
3114 local_scan() because the Received: header was not generated till after they
3115 were called (in order to set the time as the time of reception completion).
3116 I have revised the way this works. The header is now generated after the
3117 body is received, but before the ACL or local_scan() are called. After they
3118 are run, the timestamp in the header is updated.
3124 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3125 before starting a queue runner without re-exec. This happened only when
3126 deliver_drop_privilege was set or when the Exim user was set to root. The
3127 effect of the bug was that timeouts during subsequent deliveries caused
3128 crashes instead of being properly handled. The handler is now left at its
3129 default (and expected) setting.
3131 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3132 message, again when deliver_drop_privilege is set or Exim is run as root.
3133 The bug described in (1) was not present in this case, but the tidying up
3134 of the other signals was missing. I have made the two cases consistent.
3136 3. The ignore_target_hosts setting on a manualroute router was being ignored
3137 for hosts that were looked up using the /MX notation.
3139 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3142 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3143 operated on the sender address. After changing the $sender_address to <>
3144 for the sender address verify, Exim was re-instated it as the original
3145 (before rewriting) address, but remembering that it had rewritten it, so it
3146 wasn't rewriting it again. This bug also had the effect of breaking the
3147 sender address verification caching when the sender address was rewritten.
3149 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3150 This has been changed so that if the ip literal address matches
3151 ignore_target_hosts, the router declines.
3153 7. Added expansion conditions match_domain, match_address, and match_local_
3154 part (NOT match_host).
3156 8. The placeholder for the Received: header didn't have a length field set.
3158 9. Added code to Exim itself and to exim_lock to test for a specific race
3159 condition that could lead to file corruption when using MBX delivery. The
3160 issue is with the lockfile that is created in /tmp. If this file is removed
3161 after a process has opened it but before that process has acquired a lock,
3162 there is the potential for a second process to recreate the file and also
3163 acquire a lock. This could lead to two Exim processes writing to the file
3164 at the same time. The added code performs the same test as UW imapd; it
3165 checks after acquiring the lock that its file descriptor still refers to
3166 the same named file.
3168 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3169 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3172 11. Added the smtp_active_hostname option. If used, this will typically be made
3173 to depend on the incoming interface address. Because $interface_address is
3174 not set up until the daemon has forked a reception process, error responses
3175 that can happen earlier (such as "too many connections") no longer contain
3178 12. If an expansion in a condition on a "warn" statement fails because a lookup
3179 defers, the "warn" statement is abandoned, and the next ACL statement is
3180 processed. Previously this caused the whole ACL to be aborted.
3182 13. Added the iplsearch lookup type.
3184 14. Added ident_timeout as a log selector.
3186 15. Added tls_certificate_verified as a log selector.
3188 16. Added a global option tls_require_ciphers (compare the smtp transport
3189 option of the same name). This controls incoming TLS connections.
3191 17. I finally figured out how to make tls_require_ciphers do a similar thing
3192 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3193 before starting the TLS session.
3195 18. Tabs are now shown as \t in -bP output.
3197 19. If the log selector return_path_on_delivery was set, Exim crashed when
3198 bouncing a message because it had too many Received: header lines.
3200 20. If two routers both had headers_remove settings, and the first one included
3201 a superfluous trailing colon, the final name in the first list and the
3202 first name in the second list were incorrectly joined into one item (with a
3203 colon in the middle).
3209 1. Added -C and -D options to the exinext utility, mainly to make it easier
3210 to include in the automated testing, but these could be helpful when
3211 multiple configurations are in use.
3213 2. The exinext utility was not formatting the output nicely when there was
3214 an alternate port involved in the retry record key, nor when there was a
3215 message id as well (for retries that were specific to a specific message
3216 and a specific host). It was also confused by IPv6 addresses, because of
3217 the additional colons they contain. I have fixed the IPv4 problem, and
3218 patched it up to do a reasonable job for IPv6.
3220 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3221 delivery, the log line now contains "pipelined" if PIPELINING was used.
3223 4. An SMTP transport process used to panic and die if the bind() call to set
3224 an explicit outgoing interface failed. This has been changed; it is now
3225 treated in the same way as a connect() failure.
3227 5. A reference to $sender_host_name in the part of a conditional expansion
3228 that was being skipped was still causing a DNS lookup. This no longer
3231 6. The def: expansion condition was not recognizing references to header lines
3232 that used bh_ and bheader_.
3234 7. Added the _cache feature to named lists.
3236 8. The code for checking quota_filecount in the appendfile transport was
3237 allowing one more file than it should have been.
3239 9. For compatibility with Sendmail, the command line option
3247 and sets the incoming protocol and host name (for trusted callers). The
3248 host name and its colon can be omitted when only the protocol is to be set.
3249 Note the Exim already has two private options, -pd and -ps, that refer to
3250 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3251 "s", but I don't think that's a major issue.
3253 10. A number of refactoring changes to the code, none of which should affect
3256 (a) The number of logging options was getting close to filling up the
3257 32-bit word that was used as a bit map. I have split them into two classes:
3258 those that are passed in the argument to log_write(), and those that are
3259 only ever tested independently outside of that function. These are now in
3260 separate 32-bit words, so there is plenty of room for expansion again.
3261 There is no change in the user interface or the logging behaviour.
3263 (b) When building, for example, log lines, the code previously used a
3264 macro that called string_cat() twice, in order to add two strings. This is
3265 not really sufficiently general. Furthermore, there was one instance where
3266 it was actually wrong because one of the argument was used twice, and in
3267 one call a function was used. (As it happened, calling the function twice
3268 did not affect the overall behaviour.) The macro has been replaced by a
3269 function that can join an arbitrary number of extra strings onto a growing
3272 (c) The code for expansion conditions now uses a table and a binary chop
3273 instead of a serial search (which was left over from when there were very
3274 few conditions). Also, it now recognizes conditions like "pam" even when
3275 the relevant support is not compiled in: a suitably worded error message is
3276 given if an attempt is made to use such a condition.
3278 11. Added ${time_interval:xxxxx}.
3280 12. A bug was causing one of the ddress fields not to be passed back correctly
3281 from remote delivery subprocesses. The field in question was not being
3282 subsequently used, so this caused to problems in practice.
3284 13. Added new log selectors queue_time and deliver_time.
3286 14. Might have fixed a bug in maildirsizefile handling that threw up
3287 "unexpected character" debug warnings, and recalculated the data
3288 unnecessarily. In any case, I expanded the warning message to give more
3291 15. Added the message "Restricted characters in address" to the statements in
3292 the default ACL that block characters like @ and % in local parts.
3294 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3295 Three changes have been made:
3297 (a) There was a serious bug; a negative response to MAIL caused the whole
3298 recipient domain to be cached as invalid, thereby blocking all messages
3299 to all local parts at the same domain, from all senders. This bug has
3300 been fixed. The domain is no longer cached after a negative response to
3301 MAIL if the sender used is not empty.
3303 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3306 (c) A new callout option, "use_sender" has been added for people who want
3307 the modified behaviour.
3313 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3316 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3317 indeed breaks things for older releases.
3319 3. Added additional logging to the case where there is a problem reading data
3320 from a filter that is running in a subprocess using a pipe, in order to
3321 try to track down a specific problem.
3323 4. Testing facility fudge: when running in the test harness and attempting
3324 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3325 getting "No route to host". Convert this to a timeout.
3327 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3330 6. Some OS don't have socklen_t but use size_t instead. This affects the
3331 fifth argument of getsockopt() amongst other things. This is now
3332 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3333 can be set for individual OS. I have set it for SunOS5, OSF1, and
3334 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3335 some earlier ones do not.
3337 7. Change 4.30/15 was not doing the test caselessly.
3339 8. The standard form for an IPv6 address literal was being rejected by address
3340 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3341 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3342 this, as well as the form without the "IPv6" on the front (but only when
3343 address literals are enabled, of course).
3345 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3347 10. Exim crashed if a message with an empty sender address specified by -f
3348 encountered a router with an errors_to setting. This could be provoked only
3349 by a command such as
3353 where an empty string was supplied; "<>" did not hit this bug.
3355 11. Installed PCRE release 4.5.
3357 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3358 remained set. It is now erased.
3360 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3361 times from message ids (which are base 36 rather than the normal 62).
3363 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3364 were being counted as actual protocol errors, and logged if the log
3365 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3366 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3367 and DATA following a set of rejected RCPTs do not count as protocol errors.
3368 In other words, Exim assumes they were pipelined, though this may not
3369 actually be the case. Of course, in all cases the client gets an
3370 appropriate error code.
3372 15. If a lookup fails in an ACL condition, a message about the failure may
3373 be available; it is used if testing the ACL cannot continue, because most
3374 such messages specify what the cause of the deferral is. However, some
3375 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3376 that caused an old message to be retained and used if a later statement
3377 caused a defer, replacing the real cause of the deferral.
3379 16. If an IP address had so many PTR records that the DNS lookup buffer
3380 was not large enough to hold them, Exim could crash while trying to process
3381 the truncated data. It now detects and logs this case.
3383 17. Further to 4.21/58, another change has been made: if (and only if) the
3384 first line of a message (the first header line) ends with CRLF, a bare LF
3385 in a subsequent header line has a space inserted after it, so as not to
3386 terminate the header.
3388 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3389 unnecessarily, used atoi() instead of strtol(), and didn't check the
3390 termination when getting file sizes from file names by regex.
3392 19. Completely re-implemented the support for maildirsize files, in the light
3393 of a number of problems with the previous contributed implementation
3394 (4.30/29). In particular:
3396 . If the quota is zero, the maildirsize file is maintained, but no quota is
3399 . If the maildir directory does not exist, it is created before any attempt
3400 to write a maildirsize file.
3402 . The quota value in the file is just a cache; if the quota is changed in
3403 the transport, the new value overrides.
3405 . A regular expression is available for excluding directories from the
3408 20. The autoreply transport checks the characters in options that define the
3409 message's headers; it allows continued headers, but it was checking with
3410 isspace() after an embedded newline instead of explicitly looking for a
3413 21. If all the "regular" hosts to which an address was routed had passed their
3414 expiry times, and had not reached their retry times, the address was
3415 bounced, even if fallback hosts were defined. Now Exim should go on to try
3418 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3419 equivalent code in the SMTP transport. Some hosts send humungous responses
3420 to HELO/EHLO, more than 1024 it seems.
3422 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3423 gives compiler warnings in some environments. I've now done it "properly",
3426 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3427 (because of problems with the built-in one) was declared to return uschar *
3428 instead of char *, causing compiler failure.
3430 25. Fixed a file descriptor leak when processing alias/forward files.
3432 26. Fixed a minor format string issue in dbfn.c.
3434 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3436 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3437 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3440 29. When a sender address is verified, it is cached, to save repeating the test
3441 when there is more than one recipient in a message. However, when the
3442 verification involves a callout, it is possible for different callout
3443 options to be set for different recipients. It is too complicated to keep
3444 track of this in the cache, so now Exim always runs a verification when a
3445 callout is required, relying on the callout cache for the optimization.
3446 The overhead is duplication of the address routing, but this should not be
3449 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3450 to be verified with callout=postmaster, and the main callout worked but the
3451 postmaster check failed, the verification correctly failed. However, if a
3452 subsequent RCPT command asked for sender verification *without* the
3453 postmaster check, incorrect caching caused this verification also to fail,
3456 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3457 it was not caching the DNS options (qualify_single, search_parents) that
3458 were used when the lookup failed. A subsequent lookup with different
3459 options therefore always gave the same answer, though there were cases
3460 where it should not have. (Example: a "domains = !$mx_any" option on a
3461 dnslookup router: the "domains" option is always processed without any
3462 widening, but the router might have qualify_single set.) Now Exim uses the
3463 cached value only when the same options are set.
3465 32. Added John Jetmore's "exipick" utility to the distribution.
3467 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3468 than a timeout (e.g. a certificate is required, and is not provided), an
3469 Exim server now closes the connection immediately. Previously it waited for
3470 the client to close - but if the client is SSL, it seems that they each
3471 wait for each other, leading to a delay before one of them times out.
3473 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3474 maintained 0.8.x compatibility because I don't think many are using it, and
3475 it is clearly obsolete.
3477 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3480 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3481 client certificate was expired. A simple patch fixes this, though I don't
3482 understand the full logic of why the verify callback is called multiple
3485 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3486 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3487 which causes problems with some clients (such as the Certicom SSL Plus
3488 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3489 disables the coutermeasure allowing Eudora to connect."
3491 38. Exim was not checking that a write() to a log file succeeded. This could
3492 lead to Bad Things if a log got too big, in particular if it hit a file
3493 size limit. Exim now panics and dies if it cannot write to a log file, just
3494 as it does if it cannot open a log file.
3496 39. Modified OS/Makefile-Linux so that it now contains
3498 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3500 The two -D definitions ensure that Exim is compiled with large file
3501 support, which makes it possible to handle log files that are bigger than
3504 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3505 instance) a domain was checked against a named list that involved a lookup,
3506 causing $domain_data to be set, then another domain was checked against the
3507 same list, then the first domain was re-checked, the value of $domain_data
3508 after the final check could be wrong. In particular, if the second check
3509 failed, it could be set empty. This bug probably also applied to
3512 41. The strip_trailing_dot option was not being applied to the address given
3513 with the -f command-line option.
3515 42. The code for reading a message's header from the spool was incrementing
3516 $received_count, but never initializing it. This meant that the value was
3517 incorrect (doubled) while delivering a message in the same process in which
3518 it was received. In the most common configuration of Exim, this never
3519 happens - a fresh exec is done - but it can happen when
3520 deliver_drop_privilege is set.
3522 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3523 it now includes up to 150 characters of the unexpected data in the log
3526 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3527 and building data strings. The size of both of these buffers was 10 000
3528 bytes - far larger than anybody would *ever* want, thought I. Needless to
3529 say, somebody hit the limit. I have increased the maximum line length to
3530 20 000 and the maximum data length of concatenated lines to 100 000. I have
3531 also fixed two bugs, because there was no checking on these buffers. Tsk,
3532 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3533 buffer is too small.
3535 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3536 lsearch lookups. Now it does.
3538 46. When parsing a route_list item in a manualroute router, a fixed-length
3539 buffer was used for the list of hosts. I made this 1024 bytes long,
3540 thinking that nobody would ever have a list of hosts that long. Wrong.
3541 Somebody had a whole pile of complicated expansion conditions, and the
3542 string was silently truncated, leading to an expansion error. It turns out
3543 that it is easier to change to an unlimited length (owing to other changes
3544 that have happened since this code was originally written) than to build
3545 structure for giving a limitation error. The length of the item that
3546 expands into the list of hosts is now unlimited.
3548 47. The lsearch lookup could not handle data where the length of text line was
3549 more than 4095 characters. Such lines were truncated, leading to shortened
3550 data being returned. It should now handle lines of any length.
3552 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3553 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3556 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3557 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3558 no longer happens in comment lines. A list of replacements is now placed
3559 at the head of all of the source files, except those whose only change is
3560 to replace PERL_COMMAND in the very first #! line.
3562 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3563 messages on the queue, with a bottom-up merge sort, using code contributed
3564 by Michael Haardt. This should make operations like -bp somewhat faster on
3565 large queues. It won't affect queue runners, except when queue_run_in_order
3568 51. Installed eximstats 1.31 in the distribution.
3570 52. Added support for SRV lookups to the dnslookup router.
3572 53. If an ACL referred to $message_body or $message_body_end, the value was not
3573 reset for any messages that followed in the same SMTP session.
3575 54. The store-handling optimization for building very long strings was not
3576 differentiating between the different store pools. I don't think this
3577 actually made any difference in practice, but I've tidied it.
3579 55. While running the routers to verify a sender address, $sender_address
3580 was still set to the sender address. This is wrong, because when routing to
3581 send a bounce to the sender, it would be empty. Therefore, I have changed
3582 it so that, while verifying a sender address, $sender_address is set to <>.
3583 (There is no change to what happens when verifying a recipient address.)
3585 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3586 target A or AAAA records (if not already returned) without resetting the
3587 qualify_single or search_parents options of the DNS resolver. These are
3588 inappropriate in this case because the targets of MX and SRV records must
3589 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3590 target that, when qualified, matched something in the local domain. These
3591 two options are now turned off when doing these lookups.
3593 57. It seems that at least some releases of Reiserfs (which does not have the
3594 concept of a fixed number of inodes) returns zero and not -1 for the
3595 number of available inodes. This interacted badly with check_spool_inodes,
3596 which assumed that -1 was the "no such thing" setting. What I have done is
3597 to check that the total number of inodes is greater than zero before doing
3598 the test of how many are available.
3600 58. When a "warn" ACL statement has a log_message modifier, the message is
3601 remembered, and not repeated. This is to avoid a lot of repetition when a
3602 message has many recipients that cause the same warning to be written.
3603 Howewer, Exim was preserving the list of already written lines for an
3604 entire SMTP session, which doesn't seem right. The memory is now reset if a
3605 new message is started.
3607 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3608 debugging output unless log_rewrite was also set.
3610 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3611 of (int)(handle) when we know that handle contains (void *)(-1).
3613 61. The Exim daemon panic-logs an error return when it closes the incoming
3614 connection. However "connection reset by peer" seems to be common, and
3615 isn't really an error worthy of noting specially, so that particular error
3618 62. When Exim is trying to find all the local interfaces, it used to panic and
3619 die if the ioctl to get the interface flags failed. However, it seems that
3620 on at least one OS (Solaris 9) it is possible to have an interface that is
3621 included in the list of interfaces, but for which you get a failure error
3622 for this call. This happens when the interface is not "plumbed" into a
3623 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3624 failure of the "get flags" call assumes that the interface is down.
3626 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3627 makes life easier for people who are doing arithmetic on fields extracted
3628 from dates, where you often get leading zeros that should not be
3629 interpreted as octal.
3631 64. Added qualify_domain to the redirect router, to override the global
3634 65. If a pathologically long header line contained very many addresses (the
3635 report of this problem mentioned 10 000) and each of them was rewritten,
3636 Exim could use up a very large amount of memory. (It kept on making new
3637 copies of the header line as it rewrote, and never released the old ones.)
3638 At the expense of a bit more processing, the header rewriting function has
3639 been changed so that it no longer eats memory in this way.
3641 66. The generation of the Received: header has been moved from the time that a
3642 message starts to be received, to the time that it finishes. The timestamp
3643 in the Received: header should now be very close to that of the <= log
3644 line. There are two side-effects of this change:
3646 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3647 logged header lines no longer include the local Received: line, because
3648 it has not yet been created. The same applies to a copy of the message
3649 that is returned to a non-SMTP sender when a message is rejected.
3651 (b) When a filter file is tested using -bf, no additional Received: header
3652 is added to the test message. After some thought, I decided that this
3655 This change does not affect the value of $received_for. It is still set
3656 after address rewriting, but before local_scan() is called.
3658 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3660 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3661 gave an unhelpful panic error message, and a defer error. I have managed to
3662 change this behaviour so that it now rejects any supplied certificate,
3663 which seems right, as the list of acceptable certificates is empty.
3665 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3666 gave an unhelpful defer error. I have not managed to make this reject any
3667 supplied certificates, but the error message it gives is "no certificate
3668 supplied", which is not helpful.
3670 70. exigrep's output now also includes lines that are not associated with any
3671 message, but which match the given pattern. Implemented by a patch from
3672 Martin Sluka, which also tidied up the Perl a bit.
3674 71. Recipient callout verification, like sender verification, was using <> in
3675 the MAIL FROM command. This isn't really the right thing, since the actual
3676 sender may affect whether the remote host accepts the recipient or not. I
3677 have changed it to use the actual sender in the callout; this means that
3678 the cache record is now keyed on a recipient/sender pair, not just the
3679 recipient address. There doesn't seem to be a real danger of callout loops,
3680 since a callout by the remote host to check the sender would use <>.
3681 [SEE ABOVE: changed after hitting problems.]
3683 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3684 temporary errors. However, in the case of such a code being given after
3685 the end of a data transmission (i.e. after ".") Exim was failing to write
3686 a retry record for the message. (Yes, there was some broken host that was
3687 actually sending 8xx at this point.)
3689 73. An unknown lookup type in a host list could cause Exim to panic-die when
3690 the list was checked. (An example that provoked this was putting <; in the
3691 middle of a list instead of at the start.) If this happened during a DATA
3692 ACL check, a -D file could be left lying around. This kind of configuration
3693 error no longer causes Exim to die; instead it causes a defer errror. The
3694 incident is still logged to the main and panic logs.
3696 74. Buglet left over from Exim 3 conversion. The message "too many messages
3697 in one connection" was written to the rejectlog but not the mainlog, except
3698 when address rewriting (yes!) was being logged.
3700 75. Added write_rejectlog option.
3702 76. When a system filter was run not as root (that is, when system_filter_user
3703 was set), the values of the $n variables were not being returned to the
3704 main process; thus, they were not subsequently available in the $sn
3707 77. Added +return_path_on_delivery log selector.
3709 78. A connection timeout was being treated differently from recipients deferred
3710 when testing hosts_max_try with a message that was older than the host's
3711 retry timeout. (The host should not be counted, thus allowing all hosts to
3712 be tried at least once before bouncing.) This may have been the cause of an
3713 occasionally reported bug whereby a message would remain on the queue
3714 longer than the retry timeout, but would be bounced if a delivery was
3715 forced. I say "may" because I never totally pinned down the problem;
3716 setting up timeout/retry tests is difficult. See also the next item.
3718 79. The ultimate address timeout was not being applied to errors that involved
3719 a combination of host plus message (for example, a timeout on a MAIL
3720 command). When an address resolved to a number of possible hosts, and they
3721 were not all tried for each delivery (e.g. because of hosts_max_try), a
3722 message could remain on the queue longer than the retry timeout.
3724 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3727 81. Fixed an obscure SMTP outgoing bug which required at least the following
3728 conditions: (a) there was another message waiting for the same server;
3729 (b) the server returned 5xx to all RCPT commands in the first message so
3730 that the message was not completed; (c) the server dropped the connection
3731 or gave a negative response to the RSET that Exim sends to abort the
3732 transaction. The observed case was a dropped connection after DATA that had
3733 been sent in pipelining mode. That is, the server had advertised PIPELINING
3734 but was not implementing it correctly. The effect of the bug was incorrect
3735 behaviour, such as trying another host, and this could lead to a crash.
3741 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3742 and daemon.c were passed as pointers to ints; they should have been
3743 pointers to socklen_t variables (which are typically unsigned ints).
3745 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3748 3. Fixed a really odd bug that affected only the testing scheme; patching a
3749 certain fixed string in the binary changed the value of another string that
3750 happened to be identical to the end of the original first string.
3752 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3753 name", it returns that address as the IP address. On some operating
3754 systems (e.g. Solaris), it also passes back the IP address string as the
3755 "host name". However, on others (e.g. Linux), it passes back an empty
3756 string. Exim wasn't checking for this, and was changing the host name to an
3757 empty string, assuming it had been canonicized.
3759 5. Although rare, it is permitted to have more than one PTR record for a given
3760 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3761 all the names associated with an address, because they do in Solaris.
3762 However, it seems that they do not in Linux for data that comes from the
3763 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3764 I found this out when I moved to a new Linux workstation and tried to run
3765 the Exim test suite.
3767 To get round this problem I have changed the code so that it now does its
3768 own call to the DNS to look up PTR records when searching for a host name.
3769 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3770 addresses that are only in /etc/hosts are still found.
3772 This behaviour is, however, controlled by an option called host_lookup_
3773 order, which defaults to "bydns:byaddr". If people want to use the other
3774 order, or indeed, just use one or the other means of lookup, they can
3775 specify it in this variable.
3777 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3778 some operating systems, this comes back from gethostbyaddr() as an empty
3779 string, and this is what Exim used to test for. However, it seems that in
3780 other systems, "." is yielded. Exim now tests for this case too.
3782 7. The values of check_spool_space and check_log_space are now held internally
3783 as a number of kilobytes instead of an absolute number of bytes. If a
3784 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3785 kilobyte. This means that much larger values can be stored.
3787 8. Exim monitor: an attempt to get the action menu when not actually pointing
3788 at a message produces an empty menu entitled "No message selected". This
3789 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3790 no entries in it ("Shell widget menu has zero width and/or height"). So I
3791 have added a single, blank menu entry in this case.
3793 9. Added ${quote_local_part.
3795 10. MIME decoding is now applied to the contents of Subject: header lines when
3798 11. Now that a reference to $sender_host_address automatically causes a reverse
3799 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3800 host lookup before query-style lookups in lists that might use this
3801 variable. This has therefore been abolished, and the "net-" prefix is no
3802 longer necessary for query-style lookups.
3804 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3805 have been a typo for LFLAGS, so it has been changed.
3807 13. The install script calls Exim with "-C /dev/null" in order to find the
3808 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3809 to be output. Howeve, since Exim outputs its version number before the
3810 error, it didn't break the script. It just looked ugly. I fixed this by
3811 always allowing "-C /dev/null" if the caller is root.
3813 14. Ignore overlarge ACL variable number when reading spool file - insurance
3814 against a later release with more variables having written the file.
3816 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3817 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3818 this, as well as the form without the "IPv6" on the front.
3820 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3821 OS/Makefile-Darwin file.
3823 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3824 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3826 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3827 variable. However, some OS use ints for inodes. I've added cast to long int
3828 to get rid of the compiler warning.
3830 19. I had forgotten to lock out "/../" in configuration file names when
3831 ALT_CONFIG_PREFIX was set.
3833 20. Routers used for verification do not need to specify transports. However,
3834 if such a router generated a host list, and callout was configured, Exim
3835 crashed, because it could not find a port number from the (non-existent)
3836 transport. It now assumes port 25 in this circumstance.
3838 21. Added the -t option to exigrep.
3840 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3841 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3842 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3843 them is set, LOOKUP_LSEARCH is forced.
3845 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3847 24. Added sender and host information to the "rejected by local_scan()" log
3848 line; previously there was no indication of these.
3850 25. Added .include_if_exists.
3852 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3853 Linux. It turns out that not all file systems support this. Apparently some
3854 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3855 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3856 not supported on the file descriptor, is now ignored when Exim is trying to
3857 sync a directory. This applies only to Linux.
3859 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3861 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3862 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3865 29. Added support for maildirsize files from supplied patch (modified a bit).
3867 30. The use of :fail: followed by an empty string could lead Exim to respond to
3868 sender verification failures with (e.g.):
3870 550 Verification failed for <xxx>
3871 550 Sender verify failed
3873 where the first response line was missing the '-' that indicates it is not
3874 the final line of the response.
3876 31. The loop for finding the name of the user that called Exim had a hardwired
3877 limit of 10; it now uses the value of finduser_retries, which is used for
3878 all other user lookups.
3880 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3883 33. Exim was neglecting to zero errno before one call of strtol() when
3884 expanding a string and expecting an integer value. On some systems this
3885 resulted in spurious "integer overflow" errors. Also, it was casting the
3886 result into an int without checking.
3888 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3889 did not work. The code looks as if it has *never* worked, though it appears
3890 to have been documented since at least releast 1.62. I have made it work.
3892 35. The "timeout_DNS" error in retry rules, also documented since at least
3893 1.62, also never worked. As it isn't clear exactly what this means, and
3894 clearly it isn't a major issue, I have abolished the feature by treating it
3895 as "timeout", and writing a warning to the main and panic logs.
3897 36. The display of retry rules for -brt wasn't always showing the error code
3900 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3901 timeout_connect_A, timeout_connect_MX.
3903 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3904 to the empty sender.
3906 39. The daemon was not analysing the content of -oX till after it had closed
3907 stderr and disconnected from the controlling terminal. This meant that any
3908 syntax errors were only noted on the panic log, and the return code from
3909 the command was 0. By re-arranging the code a little, I've made the
3910 decoding happen first, so such errors now appear on stderr, and the return
3911 code is 1. However, the actual setting up of the sockets still happens in
3912 the disconnected process, so errors there are still only recorded on the
3915 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3916 connections (as happens on some IP stacks) was logged at start up time as
3917 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3918 it from "IPv6 and IPv4", which means that two separate sockets are being
3921 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3922 says whether AF_INET or AF_INET6 was passed as an argument.
3924 42. Exiwhat output was messed up when time zones were included in log
3927 43. Exiwhat now gives more information about the daemon's listening ports,
3928 and whether -tls-on-connect was used.
3930 44. The "port" option of the smtp transport is now expanded.
3932 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3933 silently ignored. Now an error message is written to the main and panic
3936 46. There's a new ACL modifier called "logwrite" which writes to a log file
3937 as soon as it is encountered.
3939 47. Added $local_user_uid and $local_user_gid at routing time.
3941 48. Exim crashed when trying to verify a sender address that was being
3944 49. Exim was recognizing only a space character after ".include". It now also
3945 recognizes a tab character.
3947 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3948 extracting the relevant information from the specification. The man page no
3949 longer contains scrambled data for the -d option, and I've added a section
3950 at the front about calling Exim under different names.
3952 51. Added "extra_headers" argument to the "mail" command in filter files.
3954 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3957 53. Installed eximstats 1.29.
3959 54. Added transport_filter_timeout as a generic transport option.
3961 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3962 Cc: header lines. This was required by RFC 822, but it not required by RFC
3965 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3966 incoming messages that did not have them. Now it does so only if the
3967 message originates locally, that is, if there is no associated remote host
3968 address. When Resent- header lines are present, this applies to the Resent-
3969 lines rather than the non-Resent- lines.
3971 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3972 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3974 58. Messages for configuration errors now include the name of the main
3975 configuration file - useful now that there may be more than one file in a
3976 list (.included file names were always shown).
3978 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3979 for those rare installations that do not start the daemon as root or run it
3980 setuid root. I've cut out the call to initgroups() if the daemon is not
3983 60. The Exim user and group can now be bound into the binary as text strings
3984 that are looked up at the start of Exim's processing.
3986 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3988 62. Added $mailstore_basename variable.
3990 63. Installed patch to sieve.c from Michael Haardt.
3992 64. When Exim failed to open the panic log after failing to open the main log,
3993 the original message it was trying to log was written to stderr and debug
3994 output, but if they were not available (the usual case in production), it
3995 was lost. Now it is written to syslog before the two lines that record the
3996 failures to open the logs.
3998 65. Users' Exim filters run in subprocesses under the user's uid. It is
3999 possible for a "deliver" command or an alias in a "personal" command to
4000 provoke an address rewrite. If logging of address rewriting is configured,
4001 this fails because the process is not running as root or exim. There may be
4002 a better way of dealing with this, but for the moment (because 4.30 needs
4003 to be released), I have disabled address rewrite logging when running a
4004 filter in a non-root, non-exim process.
4010 1. The buildconfig auxiliary program wasn't quoting the value set for
4011 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4012 not defined. This bug was masked in 4.22 by the effect that was fixed in
4015 2. Some messages that were rejected after a message id was allocated were
4016 shown as "incomplete" by exigrep. It no longer does this for messages that
4017 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4019 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4020 have allow_domain_literals set, the ID did not get logged in the <= line.
4021 Domain literals are now always recognized in Message-ID: header lines.
4023 4. The first argument for a ${extract expansion item is the key name or field
4024 number. Leading and trailing spaces in this item were not being ignored,
4025 causing some misleading effects.
4027 5. When deliver_drop_privilege was set, single queue runner processes started
4028 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4029 same command in the process it spins off) were not dropping privilege.
4031 6. When the daemon running as "exim" started a queue runner, it always
4032 re-executed Exim in the spun-off process. This is a waste of effort when
4033 deliver_drop_privilege is set. The new process now just calls the
4034 queue-runner function directly.
4040 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4043 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4044 ignored. Though the use of -oP was forcing the writing of a pid file, it
4045 was always written to the default place.
4047 3. If the message "no IP address found for host xxxx" is generated during
4048 incoming verification, it is now followed by identification of the incoming
4049 connection (so you can more easily find what provoked it).
4051 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4053 5. Added some features to "harden" Exim a bit more against certain attacks:
4055 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4056 be put in Local/Makefile. This is like the never_users runtime option,
4057 but it cannot be overridden. The default setting is "root".
4059 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4060 prefix string with which any file named in a -C command line option
4063 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4064 is retained for -C and -D only if the caller of Exim is root. Without
4065 it, the exim user may also use -C and -D and retain privilege.
4067 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4068 command line option is disabled.
4070 6. Macro names set by the -D option must start with an upper case letter, just
4071 like macro names defined in the configuration file.
4073 7. Added "dereference=" facility to LDAP.
4075 8. Two instances of the typo "uknown" in the source files are fixed.
4077 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4078 the Configure-Makefile script screwed up while processing it.
4080 10. Incorporated PCRE 4.4.
4082 11. The SMTP synchronization check was not operating right at the start of an
4083 SMTP session. For example, it could not catch a HELO sent before the client
4084 waited for the greeting. There is now a check for outstanding input at the
4085 point when the greeting is written. Because of the duplex, asynchronous
4086 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4087 way, but not yet received, when the check is performed.
4089 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4090 on TCP/IP sockets, because this apparently causes some broken clients to
4093 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4094 unchanged) from the Cygwin maintainer.
4096 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4097 was supported instead of testing for mbx. Effectively a typo.
4099 15. The spa authenticator server code was not checking that the input it
4100 received was valid base64.
4102 16. The debug output line for the "set" modifier in ACLs was not showing the
4103 name of the variable that was being set.
4105 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4107 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4108 lookup on its own. Something else was needed to trigger the lookup. For
4109 example, a match in host_lookup or the need for a host name in a host list.
4110 Now, if $sender_host_name is referenced and the host name has not yet been
4111 looked up, a lookup is performed. If the lookup fails, the variable remains
4112 empty, and $host_lookup_failed is set to "1".
4114 19. Added "eqi" as a case-independent comparison operator.
4116 20. The saslauthd authentication condition could segfault if neither service
4117 nor realm was specified.
4119 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4120 error message that was logged was misleading, and incoming SMTP
4121 connections were dropped. The message is now more accurate, and temporary
4122 errors are given to SMTP connections.
4124 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4125 (except RSET) with a 421 error, until QUIT is received. However, it was
4126 failing to send a response to QUIT.
4128 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4129 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4130 ACL is now run after the helo_try_verify_hosts code.
4132 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4133 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4134 case-independent, so other case variants are also recognized). Apparently
4135 some systems use these upper case variants.
4137 25. If more than two messages were waiting for the same host, and a transport
4138 filter was specified for the transport, Exim sent two messages over the
4139 same TCP/IP connection, and then failed with "socket operation on non-
4140 socket" when it tried to send the third.
4142 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4144 27. The extern definition of crypt16() in expand.c was not being excluded when
4145 the OS had its own crypt16() function.
4147 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4148 as a preferred synonym for return_size_limit, both as an option and as an
4151 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4153 30. Changed the default configuration ACL to relax the local part checking rule
4154 for addresses that are not in any local domains. For these addresses,
4155 slashes and pipe symbols are allowed within local parts, but the sequence
4156 /../ is explicitly forbidden.
4158 31. SPA server authentication was not clearing the challenge buffer before
4161 32. log_message in a "warn" ACL statement was writing to the reject log as
4162 well as to the main log, which contradicts the documentation and doesn't
4163 seem right (because no rejection is happening). So I have stopped it.
4165 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4166 However, I am unable to do any testing of this.
4168 34. Fixed an infelicity in the appendfile transport. When checking directories
4169 for a mailbox, to see if any needed to be created, it was accidentally
4170 using path names with one or more superfluous leading slashes; tracing
4171 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4173 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4174 contents are added to the log line that is written for every discarded
4175 recipient. (Previously a log_message setting was ignored.)
4177 36. The ${quote: operator now quotes the string if it is empty.
4179 37. The install script runs exim in order to find its version number. If for
4180 some reason other than non-existence or emptiness, which it checks, it
4181 could not run './exim', it was installing it with an empty version number,
4182 i.e. as "exim-". This error state is now caught, and the installation is
4185 38. An argument was missing from the function that creates an error message
4186 when Exim fails to connect to the socket for saslauthd authentication.
4187 This could cause Exim to crash, or give a corrupted message.
4189 39. Added isip, isip4, and isip6 to ${if conditions.
4191 40. The ACL variables $acl_xx are now saved with the message, and can be
4192 accessed later in routers, transports, and filters.
4194 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4195 strings in the file are not string-expanded.
4197 42. If a MAIL command specified a SIZE value that was too large to fit into an
4198 int variable, the check against message_size_limit failed. Such values are
4199 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4200 day this will have to be increased, but I don't think I want to be around
4201 when emails are that large.
4208 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4209 iconv() is not standard in FreeBSD.
4211 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4212 IPv6 enabled. The observed symptom was a segmentation fault on return from
4213 the function os_common_find_running_interfaces() in src/os.c.
4215 3. In the check_special_case() function in daemon.c I had used "errno" as an
4216 argument name, which causes warnings on some systems. This was basically a
4217 typo, since it was named "eno" in the comments!
4219 4. The code that waits for the clock to tick (at a resolution of some fraction
4220 of a second) so as to ensure message-id uniqueness was always waiting for
4221 at least one whole tick, when it could have waited for less. [This is
4222 almost certainly not relevant at current processor speeds, where it is
4223 unlikely to ever wait at all. But we try to future-proof.]
4225 5. The function that sleeps for a time interval that includes fractions of a
4226 second contained a race. It did not block SIGALRM between setting the
4227 timer, and suspending (a couple of lines later). If the interval was short
4228 and the sigsuspend() was delayed until after it had expired, the suspension
4229 never ended. On busy systems this could lead to processes getting stuck for
4232 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4233 process, before it forks any delivery processes. The open lookup caching
4234 mechanism meant that the open file or database connection was passed into
4235 the delivery process. The problem was that delivery processes always tidy
4236 up cached lookup data. This could cause a problem for the next delivery
4237 process started by the queue runner, because the external queue runner
4238 process does not know about the closure. So the next delivery process
4239 still has data in the lookup cache. In the case of a file lookup, there was
4240 no problem because closing a file descriptor in a subprocess doesn't affect
4241 the parent. However, if the lookup was caching a connection to a database,
4242 the connection was closed, and the second delivery process was likely to
4243 see errors such as "PGSQL: query failed: server closed the connection
4244 unexpectedly". The problem has been fixed by closing all cached lookups
4245 in a queue runner before running a delivery process.
4247 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4248 seem to have the "const" qualifier which it has on other OS. I've
4251 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4252 *of the same type* (client or server) with the same public name that an
4253 error should be diagnosed.
4255 9. When Exim looked up a host name for an IP address, but failed to find the
4256 original IP address when looking up the host name (a safety check), it
4257 output the message "<ip address> does not match any IP for NULL", which was
4258 confusing, to say the least. The bug was that the host name should have
4259 appeared instead of "NULL".
4261 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4262 user that is built into the binary, and the -C or -D options is used, root
4263 privilege is dropped before the configuration file is read. In addition,
4264 logging is switched to stderr instead of the normal log files. If the
4265 configuration then re-defines the Exim user, the unprivileged environment
4266 is probably not what is expected, so Exim logs a panic warning message (but
4269 However, if deliver_drop_privilege is set, the unprivileged state may well
4270 be exactly what is intended, so the warning has been cut out in that case,
4271 and Exim is allowed to try to write to its normal log files.
4277 1. smtp_return_error_details was not giving details for temporary sender
4278 or receiver verification errors.
4280 2. Diagnose a configuration error if two authenticators have the same public
4283 3. Exim used not to create the message log file for a message until the first
4284 delivery attempt. This could be confusing when incoming messages were held
4285 for policy or load reasons. The message log file is now created at the time
4286 the message is received, and an initial "Received" line is written to it.
4288 4. The automatically generated man page for command line options had a minor
4289 bug that caused no ill effects; however, a more serious problem was that
4290 the procedure for building the man page automatically didn't always
4291 operate. Consequently, release 4.20 contains an out-of-date version. This
4292 shouldn't happen again.
4294 5. When building Exim with embedded Perl support, the script that builds the
4295 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4296 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4298 6. The freeze_tell option was not being used for messages that were frozen on
4299 arrival, either by an ACL or by local_scan().
4301 7. Added the smtp_incomplete_transaction log selector.
4303 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4304 was accepting AUTH without a new EHLO.
4306 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4307 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4310 10. Logging of TCP/IP connections (when configured) now happens in the main
4311 daemon process instead of the child process, so that the TCP/IP connection
4312 count is more accurate (but it can never be perfect).
4314 11. The use of "drop" in a nested ACL was not being handled correctly in the
4315 outer ACL. Now, if condition failure induced by the nested "drop" causes
4316 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4317 or "require"), the connection is dropped.
4319 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4320 that yield "discard" can now be used with an "accept" or a "discard" verb,
4321 but an error is generated for any others (because I can't see a useful way
4322 to define what should happen).
4324 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4325 are now processed in the same way as lines in the Exim configuration file.
4326 In particular, continuation lines are supported.
4328 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4330 15. Added -ti meaning -t -i.
4332 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4333 domains, and warn by logging if others are found.
4335 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4336 returned from GIFCONF when Exim is trying to find the list of interfaces on
4337 a host. The code in os.c has been modified to copy each ifreq to an aligned
4338 structure in all cases.
4340 Also, in some cases, the returned ifreq's were being copied to a 'struct
4341 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4342 means the last couple of bytes of an IPv6 address could be chopped if the
4343 ifreq contained only a normal sockaddr (14 bytes storage).
4345 18. Named domain lists were not supported in the hosts_treat_as_local option.
4346 An entry such as +xxxx was not recognized, and was treated as a literal
4349 19. Ensure that header lines added by a DATA ACL are included in the reject log
4350 if the ACL subsequently rejects the message.
4352 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4353 MD5 (which is deprecated).
4355 21. When testing a filter file using -bf, Exim was writing a message when it
4356 took the sender from a "From " line in the message, but it was not doing so
4357 when it took $return_path from a Return-Path: header line. It now does.
4359 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4360 with a valid header line field name (a series of printing characters
4361 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4363 23. Changed "disc" in the source to "disk" to conform to the documentation and
4364 the book and for uniformity.
4366 24. Ignore Sendmail's -Ooption=value command line item.
4368 25. When execve() failed while trying to run a command in a pipe transport,
4369 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4370 could be confused with a return value of 69 from the command itself. This
4371 has been changed to 127, the value the shell returns if it is asked to run
4372 a non-existent command. The wording for the related log line suggests a
4373 non-existent command as the problem.
4375 26. If received_header_text expands to an empty string, do not add a Received:
4376 header line to the message. (Well, it adds a token one on the spool, but
4377 marks it "old" so that it doesn't get used or transmitted.)
4379 27. Installed eximstats 1.28 (addition of -nt option).
4381 28. There was no check for failure on the call to getsockname() in the daemon
4382 code. This can fail if there is a shortage of resources on the system, with
4383 ENOMEM, for example. A temporary error is now given on failure.
4385 29. Contrary to the C standard, it seems that in some environments, the
4386 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4387 program. Exim now does this explicitly; it affects the formatting of
4388 timestamps using strftime().
4390 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4391 complaints. I've now initialized all the variables, to avoid this.
4393 32. Header lines added by a system filter were not being "seen" during
4394 transport-time rewrites.
4396 33. The info_callback() function passed to OpenSSL is set up with type void
4397 (*)(SSL *, int, int), as described somewhere. However, when calling the
4398 function (actually a macro) that sets it up, the type void(*)() is
4399 expected. I've put in a cast to prevent warnings from picky compilers.
4401 34. If a DNS black list lookup found a CNAME record, but there were no A
4402 records associated with the domain it pointed at, Exim crashed.
4404 35. If a DNS black list lookup returned more than one A record, Exim ignored
4405 all but the first. It now scans all returned addresses if a particular IP
4406 value is being sought. In this situation, the contents of the
4407 $dnslist_value variable are a list of all the addresses, separated by a
4410 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4411 to accept a host name and all its aliases if the forward lookup for any of
4412 them yielded the IP address of the incoming connection. Now it accepts only
4413 those names whose forward lookup yields the correct IP address. Any other
4414 names are discarded. This closes a loophole whereby a rogue DNS
4415 administrator could create reverse DNS records to break through a
4416 wildcarded host restriction in an ACL.
4418 37. If a user filter or a system filter that ran in a subprocess used any of
4419 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4420 the wrong values were passed to the pipe command ($thisaddress had the
4421 value of $0, $0 had the value of $1, etc). This bug was introduced by
4422 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4424 38. Improved the line breaking for long SMTP error messages from ACLs.
4425 Previously, if there was no break point between 40 and 75 characters, Exim
4426 left the rest of the message alone. Two changes have been made: (a) I've
4427 reduced the minimum length to 35 characters; (b) if it can't find a break
4428 point between 35 and 75 characters, it looks ahead and uses the first one
4429 that it finds. This may give the occasional overlong line, but at least the
4430 remaining text gets split now.
4432 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4433 file descriptors might be low, and that setting 1000 would always raise it.
4434 It turns out that in some environments, the limit is already over 1000 and
4435 that lowering it causes trouble. So now Exim takes care not to decrease it.
4437 40. When delivering a message, the value of $return_path is set to $sender_
4438 address at the start of routing (routers may change the value). By an
4439 oversight, this default was not being set up when an address was tested by
4440 -bt or -bv, which affected the outcome if any router or filter referred to
4443 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4444 log only when "message" or "log_message" are set. However, if one of the
4445 conditions was an address verification, or a call to a nested ACL, the
4446 messages generated by the underlying test were being passed through. This
4447 no longer happens. The underlying message is available in $acl_verify_
4448 message for both "message" and "log_message" expansions, so it can be
4449 passed through if needed.
4451 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4452 new expansion $bh_ to give the encoded byte string without charset
4453 translation. Translation happens only if iconv() is available; HAVE_ICONV
4454 indicates this at build time. HEADERS_CHARSET gives the charset to
4455 translate to; headers_charset can change it in the configuration, and
4456 "headers charset" can change it in an individual filter file.
4458 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4459 that creates RFC 2047 encoded "words" labels them as that charset instead
4460 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4461 expansion operator; (ii) when Exim creates a From: line for a local
4462 message; (iii) when a header line is rewritten to include a "phrase" part.
4464 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4465 buggy, causing it to skip the first lines of messages whose message ID
4466 ended in 'D'. This would not have bitten before Exim release 4.14, because
4467 message IDs were unlikely to end in 'D' before then. The effect was to have
4468 incorrect size information for certain domains.
4470 45. #include "config.h" was missing at the start of the crypt16.c module. This
4471 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4474 46. If there was a timeout during a "random" callout check, Exim treated it as
4475 a failure of the random address, and carried on sending RSET and the real
4476 address. If the delay was just some slowness somewhere, the response to the
4477 original RCPT would be taken as a response to RSET and so on, causing
4478 mayhem of various kinds.
4480 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4481 when I implemented it. It didn't allow for the fact that some option values
4482 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4483 the right test for positive values.
4485 48. Domain names in DNS records are case-independent. Exim always looks them up
4486 in lower case. Some resolvers return domain names in exactly the case they
4487 appear in the zone file, that is, they may contain uppercase letters. Not
4488 all resolvers do this - some return always lower case. Exim was treating a
4489 change of case by a resolver as a change of domain, similar to a widening
4490 of a domain abbreviation. This triggered its re-routing code and so it was
4491 trying to route what was effectively the same domain again. This normally
4492 caused routing to fail (because the router wouldn't handle the domain
4493 twice). Now Exim checks for this case specially, and just changes the
4494 casing of the domain that it ultimately uses when it transmits the message
4497 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4500 50. If a filter generated a file delivery with a non-absolute name (possible if
4501 no home directory exists for the router), the forbid_file option was not
4504 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4505 the existing equality matching.
4507 52. Exim was using ints instead of ino_t variables in some places where it was
4508 dealing with inode numbers.
4510 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4511 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4512 called TMPDIR, and if it finds it is different, it changes its value.
4514 54. The smtp_printf() function is now made available to local_scan() so
4515 additional output lines can be written before returning. There is also an
4516 smtp_fflush() function to enable the detection of a dropped connection.
4517 The variables smtp_input and smtp_batched_input are exported to
4520 55. Changed the default runtime configuration: the message "Unknown user"
4521 has been removed from the ACL, and instead placed on the localuser router,
4522 using the cannot_route_message feature. This means that any verification
4523 failures that generate their own messages won't get overridden. Similarly,
4524 the "Unrouteable address" message that was in the ACL for unverifiable
4525 relay addresses has also been removed.
4527 56. Added hosts_avoid_esmtp to the smtp transport.
4529 57. The exicyclog script was not checking for the esoteric option
4530 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4531 will work only if exicyclog is run under the appropriate euid.
4533 58. Following a discussion on the list, the rules by which Exim recognises line
4534 endings on incoming messages have been changed. The -dropcr and drop_cr
4535 options are now no-ops, retained only for backwards compatibility. The
4536 following line terminators are recognized: LF CRLF CR. However, special
4537 processing applies to CR:
4539 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4540 nor a local message in the state where . is a terminator.
4542 (ii) If a bare CR is encountered in a header line, an extra space is added
4543 after the line terminator so as not to end the header. The reasoning
4544 behind this is that bare CRs in header lines are most likely either
4545 to be mistakes, or people trying to play silly games.
4547 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4548 was being incorrectly given as 18 bytes larger than it should have been.
4549 This is a VOB (very old bug).
4551 60. This may never have affected anything current, but just in case it has:
4552 When the local host is found other than at the start of a list of hosts,
4553 the local host, those with the same MX, and any that follow, are discarded.
4554 When the list in question was part of a longer list of hosts, the following
4555 hosts (not currently being processed) were also being discarded. This no
4556 longer happens. I'm not sure if this situation could ever has previously
4559 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4562 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4563 header. This is something that is recommended in a new Internet Draft, and
4564 is something that is documented as being done by Sendmail. There are two
4565 possible values. For messages generated by the autoreply transport, Exim
4568 Auto-Submitted: auto-replied
4570 whereas for all other generated messages (e.g. bounces) it adds
4572 Auto-Submitted: auto-generated
4574 63. The "personal" condition in filters now includes a test for the
4575 Auto-Submitted: header. If it contains the string "auto-" the message it
4576 not considered personal.
4578 64. Added rcpt_include_affixes as a generic transport option.
4580 65. Added queue_only_override (default true).
4582 66. Added the syslog_duplication option.
4584 67. If what should have been the first header line of a message consisted of
4585 a space followed by a colon, Exim was mis-interpreting it as a header line.
4586 It isn't of course - it is syntactically invalid and should therefore be
4587 treated as the start of the message body. The misbehaviour could have
4588 caused a number of strange effects, including loss of data in subsequent
4589 header lines, and spool format errors.
4591 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4592 client host had authenticated. This control can now be exercised by an ACL
4593 for more flexibility.
4595 69. By default, callouts do not happen when testing with -bh. There is now a
4596 variant, -bhc, which does actually run the callout code, including
4597 consulting and updating the callout cache.
4599 70. Added support for saslauthd authentication, courtesy of Alexander
4602 71. If statvfs() failed on the spool or log directories while checking their
4603 size for availability, Exim confusingly gave the error "space shortage".
4604 Furthermore, in debugging mode it crashed with a floating point exception.
4605 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4606 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4607 serious problem, Exim now writes to the main and panic logs when this
4608 happens, with details of the failure. It then refuses to accept the
4609 incoming message, giving the message "spool directory problem" or "log
4610 directory problem" with a 421 code for SMTP messages.
4612 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4613 0, 1, and 2 exist, because some OS complain for execs without them (see
4614 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4615 descriptors. However, the code omitted to check that the open succeeded,
4616 causing mysterious errors if for some reason the permissions on /dev/null
4617 got screwed. Now Exim writes a message to the main and panic logs, and
4618 bombs out if it can't open /dev/null.
4620 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4621 interact so that it is all more flexible. It is supposed to remain
4622 backwards compatible. Also added extra_local_interfaces.
4624 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4625 to bomb out with an assertion failure - to the client this appears as a
4626 connection drop. This problem occurs in the part of the code that was taken
4627 from the Samba project. Fortunately, the assertion is in a very simple
4628 function, so I have fixed this by reproducing the function inline in the
4629 one place where it is called, and arranging for authentication to fail
4630 instead of killing the process with assert().
4632 75. The SPA client code was not working when the server requested OEM rather
4633 than Unicode encoding.
4635 76. Added code to make require_files with a specific uid setting more usable in
4636 the case where statting the file as root fails - usually a non-root-mounted
4637 NFS file system. When this happens and the failure is EACCES, Exim now
4638 forks a subprocess and does the per-uid checking as the relevant uid.
4640 77. Added process_log_path.
4642 78. If log_file_path was not explicitly set, a setting of check_log_space or
4643 check_log_inodes was ignored.
4645 79. If a space check for the spool or log partitions fails, the incident is now
4646 logged. Of course, in the latter case the data may get lost...
4648 80. Added the %p formatting code to string_format() so that it can be used to
4649 print addresses in debug_print(). Adjusted all the address printing in the
4650 debugging in store.c to use %p rather than %d.
4652 81. There was a concern that a line of code in smtp_in.c could overflow a
4653 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4654 initially expressed, the concern was not well-founded, because trailing
4655 spaces are removed early. However, if the trailing spaces were followed by
4656 a NULL, they did not get removed, so the overflow was possible. Two fixes
4659 (a) I re-wrote the offending code in a cleaner fashion.
4660 (b) If an incoming SMTP command contains a NULL character, it is rejected
4663 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4664 runs initgroups(), so that if the Exim user is in any additional groups,
4665 they will be used during message reception.
4671 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob