1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.488 2007/03/13 09:50:22 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.
156 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
157 fixed by 4.65/MH/01 (is this a record?) are fixed:
159 (i) An empty string was always treated as zero by the numeric comparison
160 operators. This behaviour has been restored.
162 (ii) It is documented that the numeric comparison operators always treat
163 their arguments as decimal numbers. This was broken in that numbers
164 starting with 0 were being interpreted as octal.
166 While fixing these problems I realized that there was another issue that
167 hadn't been noticed. Values of message_size_limit (both the global option
168 and the transport option) were treated as octal if they started with 0.
169 The documentation was vague. These values are now always treated as
170 decimal, and I will make that clear in the documentation.
176 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
177 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
180 MH/01 Don't check that the operands of numeric comparison operators are
181 integers when their expansion is in "skipping" mode (fixes bug
182 introduced by 4.64-PH/07).
184 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
185 child addresses, Exim now panics and dies. Previously, because the count
186 is held in a short int, deliveries were likely to be lost. As such a
187 large number of recipients for a single message is ridiculous
188 (performance will be very, very poor), I have chosen to impose a limit
189 rather than extend the field.
195 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
196 leftover -K file (the existence of which was triggered by #402).
197 While we were at it, introduced process PID as part of the -K
198 filename. This should rule out race conditions when creating
201 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
202 processing considerably. Previous code took too long for large mails,
203 triggering a timeout which in turn triggers #401.
205 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
206 in the DK code in transports.c. sendfile() is not really portable,
207 hence the _LINUX specificness.
209 TF/01 In the add_headers option to the mail command in an Exim filter,
210 there was a bug that Exim would claim a syntax error in any
211 header after the first one which had an odd number of characters
214 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
215 callout verification, Exim cached a "reject" for the entire domain. This
216 is correct for most verifications, but it is not correct for a recipient
217 verification with use_sender or use_postmaster set, because in that case
218 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
219 case of MAIL FROM:<> rejection from other early rejections (e.g.
220 rejection of HELO). When verifying a recipient using a non-null MAIL
221 address, the cache is ignored if it shows MAIL FROM:<> rejection.
222 Whatever the result of the callout, the value of the domain cache is
223 left unchanged (for any other kind of callout, getting as far as trying
224 RCPT means that the domain itself is ok).
226 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
229 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
230 manifest itself as EPIPE rather than ECONNECT. When tidying away a
231 session, the daemon ignores ECONNECT errors and logs others; it now
232 ignores EPIPE as well.
234 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
235 (quoted-printable decoding).
237 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
238 later the small subsequent patch to fix an introduced bug.
240 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
242 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
244 PH/08 An error is now given if message_size_limit is specified negative.
246 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
247 to be given (somewhat) arbitrary names.
249 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
252 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
253 miscellaneous code fixes
255 PH/10 Added the log_reject_target ACL modifier to specify where to log
258 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
259 hostname. This is wrong, because it relates to the incoming message (and
260 probably the interface on which it is arriving) and not to the outgoing
261 callout (which could be using a different interface). This has been
262 changed to use the value of the helo_data option from the smtp transport
263 instead - this is what is used when a message is actually being sent. If
264 there is no remote transport (possible with a router that sets up host
265 addresses), $smtp_active_hostname is used.
267 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
268 tweaks were necessary in order to get it to work (see also 21 below):
269 (a) The code assumed that strncpy() returns a negative number on buffer
270 overflow, which isn't the case. Replaced with Exim's string_format()
272 (b) There were several signed/unsigned issues. I just did the minimum
273 hacking in of casts. There is scope for a larger refactoring.
274 (c) The code used strcasecmp() which is not a standard C function.
275 Replaced with Exim's strcmpic() function.
276 (d) The code set only $1; it now sets $auth1 as well.
277 (e) A simple test gave the error "authentication client didn't specify
278 service in request". It would seem that Dovecot has changed its
279 interface. Fortunately there's a specification; I followed it and
280 changed what the client sends and it appears to be working now.
282 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
285 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
286 address A is aliased to B and C, where B exists and C does not. Without
287 -v the output is "A verified" because verification stops after a
288 successful redirection if more than one address is generated. However,
289 with -v the child addresses are also verified. Exim was outputting "A
290 failed to verify" and then showing the successful verification for C,
291 with its parentage. It now outputs "B failed to verify", showing B's
292 parentage before showing the successful verification of C.
294 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
295 look up a TXT record in a specific list after matching in a combined
298 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
299 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
300 they consult the DNS. I had assumed they would set it the way they
301 wanted; and indeed my experiments on Linux seem to show that in some
302 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
303 To be on the safe side, however, I have now made the interface to
304 host_find_byname() similar to host_find_bydns(), with an argument
305 containing the DNS resolver options. The host_find_byname() function now
306 sets these options at its start, just as host_find_bydns() does. The smtp
307 transport options dns_qualify_single and dns_search_parents are passed to
308 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
309 of host_find_byname() use the default settings of RES_DEFNAMES
310 (qualify_single) but not RES_DNSRCH (search_parents).
312 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
313 spool_read_header() do less string testing, by means of a preliminary
314 switch on the second character of optional "-foo" lines. (This is
315 overdue, caused by the large number of possibilities that now exist.
316 Originally there were few.) While I was there, I also converted the
317 str(n)cmp tests so they don't re-test the leading "-" and the first
318 character, in the hope this might squeeze out yet more improvement.
320 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
321 flag allowing group syntax was set by the header_syntax check but not
322 turned off, possible causing trouble later; (2) The flag was not being
323 set at all for the header_verify test, causing "group"-style headers to
324 be rejected. I have now set it in this case, and also caused header_
325 verify to ignore an empty address taken from a group. While doing this, I
326 came across some other cases where the code for allowing group syntax
327 while scanning a header line wasn't quite right (mostly, not resetting
328 the flag correctly in the right place). These bugs could have caused
329 trouble for malformed header lines. I hope it is now all correct.
331 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
332 with the "reply" argument non-NULL. The code, however (which originally
333 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
334 but it didn't always do it. This confused somebody who was copying the
335 code for some other use. I have removed all the tests.
337 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
338 feature that was used to support insecure browsers during the U.S. crypto
339 embargo. It requires special client support, and Exim is probably the
340 only MTA that supported it -- and would never use it because real RSA is
341 always available. This code has been removed, because it had the bad
342 effect of slowing Exim down by computing (never used) parameters for the
343 RSA_EXPORT functionality.
345 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
346 authenticator to fail if there's a tab character in the incoming data
347 (there should never be unless someone is messing about, as it's supposed
348 to be base64-encoded). Also added, on Timo's advice, the "secured" option
349 if the connection is using TLS or if the remote IP is the same as the
350 local IP, and the "valid-client-cert option" if a client certificate has
353 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
354 authenticators. This can be used for authorization after authentication
355 succeeds. (In the case of plaintext, it servers for both authentication
358 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
359 if any retry times were supplied.
361 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
362 connection, where there is no client IP address to check. In this
363 situation, the verify now always succeeds.
365 PH/25 Applied John Jetmore's -Mset patch.
367 PH/26 Added -bem to be like -Mset, but loading a message from a file.
369 PH/27 In a string expansion for a processed (not raw) header when multiple
370 headers of the same name were present, leading whitespace was being
371 removed from all of them, but trailing whitespace was being removed only
372 from the last one. Now trailing whitespace is removed from each header
373 before concatenation. Completely empty headers in a concatenation (as
376 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
377 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
379 PH/29 [Removed. This was a change that I later backed out, and forgot to
380 correct the ChangeLog entry (that I had efficiently created) before
381 committing the later change.]
383 PH/30 Exim was sometimes attempting to deliver messages that had suffered
384 address errors (4xx response to RCPT) over the same connection as other
385 messages routed to the same hosts. Such deliveries are always "forced",
386 so retry times are not inspected. This resulted in far too many retries
387 for the affected addresses. The effect occurred only when there were more
388 hosts than the hosts_max_try setting in the smtp transport when it had
389 the 4xx errors. Those hosts that it had tried were not added to the list
390 of hosts for which the message was waiting, so if all were tried, there
391 was no problem. Two fixes have been applied:
393 (i) If there are any address or message errors in an SMTP delivery, none
394 of the hosts (tried or untried) are now added to the list of hosts
395 for which the message is waiting, so the message should not be a
396 candidate for sending over the same connection that was used for a
397 successful delivery of some other message. This seems entirely
398 reasonable: after all the message is NOT "waiting for some host".
399 This is so "obvious" that I'm not sure why it wasn't done
400 previously. Hope I haven't missed anything, but it can't do any
401 harm, as the worst effect is to miss an optimization.
403 (ii) If, despite (i), such a delivery is accidentally attempted, the
404 routing retry time is respected, so at least it doesn't keep
405 hammering the server.
407 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
408 in ${readsocket because some servers need this prod.
410 PH/32 Added some extra debug output when updating a wait-xxx database.
412 PH/33 The hint "could be header name not terminated by colon", which has been
413 given for certain expansion errors for a long time, was not being given
414 for the ${if def:h_colon_omitted{... case.
416 PH/34 The spec says: "With one important exception, whenever a domain list is
417 being scanned, $domain contains the subject domain." There was at least
418 one case where this was not true.
420 PH/35 The error "getsockname() failed: connection reset by peer" was being
421 written to the panic log as well as the main log, but it isn't really
422 panic-worthy as it just means the connection died rather early on. I have
423 removed the panic log writing for the ECONNRESET error when getsockname()
426 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
427 runs only) independently of the message's sender address. This meant
428 that, if the 4xx error was in fact related to the sender, a different
429 message to the same recipient with a different sender could confuse
430 things. In particualar, this can happen when sending to a greylisting
431 server, but other circumstances could also provoke similar problems.
432 I have changed the default so that the retry time for these errors is now
433 based a combination of the sender and recipient addresses. This change
434 can be overridden by setting address_retry_include_sender=false in the
437 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
438 remote server are returned as part of bounce messages. This was not
439 happening for LMTP over a pipe (the lmtp transport), but now it is the
440 same for both kinds of LMTP.
442 PH/38 Despite being documented as not happening, Exim was rewriting addresses
443 in header lines that were in fact CNAMEs. This is no longer the case.
445 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
446 and queue runs started by the daemon processed all messages. This has
447 been fixed so that -R and -S can now usefully be given with -q<time>.
449 PH/40 Import PCRE release 6.7 (fixes some bugs).
451 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
453 PH/42 Give an error if -q is specified more than once.
455 PH/43 Renamed the variables $interface_address and $interface_port as
456 $received_ip_address and $received_port, to make it clear that these
457 values apply to message reception, and not to the outgoing interface when
458 a message is delivered. (The old names remain recognized, of course.)
460 PH/44 There was no timeout on the connect() call when using a Unix domain
461 socket in the ${readsocket expansion. There now is.
463 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
464 be meaningful with "accept".
466 SC/01 Eximstats V1.43
467 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
469 SC/02 Eximstats V1.44
470 Use a glob alias rather than an array ref in the generated
471 parser. This improves both readability and performance.
473 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
474 Collect SpamAssassin and rejection statistics.
475 Don't display local sender or destination tables unless
476 there is data to show.
477 Added average volumes into the top table text output.
479 SC/04 Eximstats V1.46
480 Collect data on the number of addresses (recipients)
481 as well as the number of messages.
483 SC/05 Eximstats V1.47
484 Added 'Message too big' to the list of mail rejection
485 reasons (thanks to Marco Gaiarin).
487 SC/06 Eximstats V1.48
488 Mainlog lines which have GMT offsets and are too short to
489 have a flag are now skipped.
491 SC/07 Eximstats V1.49 (Alain Williams)
492 Added the -emptyok flag.
494 SC/08 Eximstats V1.50
495 Fixes for obtaining the IP address from reject messages.
497 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
498 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
499 whitesspace changes from 4.64-PH/27
501 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
504 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
505 are found, allow negative numbers in numeric criteria)
507 JJ/06 exipick.20061117.2, added new $message_body_missing variable
509 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
510 to match changes made in 4.64-PH/43
512 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
514 PH/47 Put in an explicit test for a DNS lookup of an address record where the
515 "domain" is actually an IP address, and force a failure. This locks out
516 those revolvers/nameservers that support "A-for-A" lookups, in
517 contravention of the specifications.
519 PH/48 When a host name was looked up from an IP address, and the subsequent
520 forward lookup of the name timed out, the host name was left in
521 $sender_host_name, contrary to the specification.
523 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
524 restricted to single-key lookups, Exim was not diagnosing an error if
525 * or *@ was used with a query-style lookup.
527 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
529 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
530 long ago, but noone interested enough thought of it. Let's just say that
531 the "1.1" means that there are some new functions that weren't there at
532 some point in the past.
534 PH/51 Error processing for expansion failure of helo_data from an smtp
535 transport during callout processing was broken.
537 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
538 tested/used via the -bh/-bhc/-bs options.
540 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
541 bug, fixed in subsequent PCRE releases).
543 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
544 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
546 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
552 SC/01 Use a glob alias rather than an array ref in eximstats generated
553 parser. This improves both readability and performance.
555 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
556 Don't display local sender or destination tables in eximstats unless
557 there is data to show.
558 Added average volumes into the eximstats top table text output.
560 SC/03 Collect data on the number of addresses (recipients) as well
561 as the number of messages in eximstats.
563 TF/01 Correct an error in the documentation for the redirect router. Exim
564 does (usually) call initgroups() when daemonizing.
566 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
567 with consistent privilege compared to when running as a daemon.
569 TF/03 Note in the spec that $authenticated_id is not set for local
570 submissions from trusted users.
572 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
573 Thanks to Dean Brooks <dean@iglou.com> for the patch.
575 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
576 by adding some example configuration directives to the default
577 configuration file. A little bit of work is required to uncomment the
578 directives and define how usernames and passwords are checked, but
579 there is now a framework to start from.
581 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
582 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
583 without this. I don't know how relevant this is to other LDAP libraries.
585 PH/02 Add the verb name to the "unknown ACL verb" error.
587 PH/03 Magnus Holmgren's patch for filter_prepend_home.
589 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
591 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
592 directory not expanded when it should be if an expanded home directory
593 was set for the address (which is overridden by the transport).
595 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
598 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
599 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
600 because it is too late at that time, and has no effect.
602 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
603 security issue with \' (bugzilla #107). I could not use the
604 PQescapeStringConn() function, because it needs a PGconn value as one of
607 PH/08 When testing addresses using -bt, indicate those final addresses that
608 are duplicates that would not cause an additional delivery. At least one
609 person was confused, thinking that -bt output corresponded to deliveries.
610 (Suppressing duplicates isn't a good idea as you lose the information
611 about possibly different redirections that led to the duplicates.)
613 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
614 systems where poll() doesn't work, in particular OS X.
616 PH/10 Added more information to debugging output for retry time not reached.
618 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
619 operations in malware.c.
621 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
624 PH/13 If write_rejectlog was set false when logging was sent to syslog with
625 syslog_duplication set false, log lines that would normally be written
626 both the the main log and to the reject log were not written to syslog at
629 PH/14 In the default configuration, change the use of "message" in ACL warn
630 statements to "add_header".
632 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
633 not followed by a command (e.g. "seen endif").
635 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
636 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
639 PH/17 Added extra conditions to the default value of delay_warning_condition
643 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
644 { match{$h_precedence:}{(?i)bulk|list|junk} } \
645 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
648 The Auto-Submitted: and various List- headers are standardised, whereas I
649 don't think Precedence: ever was.
651 PH/18 Refactored debugging code in route_finduser() to show more information,
652 in particular, the error code if getpwnam() issues one.
654 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
655 This is apparently needed in addition to the PH/07 change above to avoid
656 any possible encoding problems.
658 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
659 but not after initializing Perl.
661 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
662 output them only if debugging. By default they are written stderr,
663 apparently, which is not desirable.
665 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
668 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
671 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
673 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
674 authenticated or an ident call has been made. Suppress the default
675 values for $authenticated_id and $authenticated_sender (but permit -oMai
676 and -oMas) when testing with -bh.
678 PH/24 Re-jigged the order of the tests in the default configuration so that the
679 tests for valid domains and recipients precede the DNS black list and CSA
680 tests, on the grounds that those ones are more expensive.
682 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
683 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
684 This bug exists in every version of Exim that I still have, right back to
687 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
688 However, an attempt to turn on case-sensitivity in a regex key by
689 including (?-i) didn't work because the subject string was already
690 lowercased, and the effects were non-intuitive. It turns out that a
691 one-line patch can be used to allow (?-i) to work as expected.
697 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
698 other effects) broke the use of negated acl sub-conditions.
700 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
703 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
704 "Deny" causes Exim to reject the incoming connection with a 554 error.
705 Unfortunately, if there is a major crisis, such as a disk failure,
706 tcp-wrappers gives "deny", whereas what one would like would be some
707 kind of temporary error. A kludge has been added to help with this.
708 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
709 554 error is used if errno is still zero or contains ENOENT (which occurs
710 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
713 PH/03 Add -lutil to the default FreeBSD LIBS setting.
715 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
716 errors. Otherwise a message that provokes a temporary error (when other
717 messages do not) can cause a whole host to time out.
719 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
720 addresses were routed directly to files or pipes from a redirect router.
721 File deliveries just didn't batch; pipe deliveries might have suffered
724 PH/06 A failure to get a lock for a hints database would erroneously always say
725 "Failed to get write lock", even when it was really a read lock.
727 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
728 of 0600. This has been changed to use the value of the lockfile_mode
729 option (which defaults to 0600).
731 PH/08 Applied small patch from the Sieve maintainer.
733 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
734 folder from quota calculations, a direct delivery into this folder messed
735 up the contents of the maildirsize file. This was because the regex was
736 used only to exclude .Trash (or whatever) when the size of the mailbox
737 was calculated. There was no check that a delivery was happening into an
738 excluded directory. This bug has been fixed by ignoring all quota
739 processing for deliveries into excluded directories.
741 PH/10 Added the maildirfolder_create_regex option to appendfile.
747 PH/01 The code for finding all the local interface addresses on a FreeBSD
748 system running IPv6 was broken. This may well have applied to all BSD
749 systems, as well as to others that have similar system calls. The broken
750 code found IPv4 interfaces correctly, but gave incorrect values for the
751 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
752 that it would not match correctly against @[] and not recognize the IPv6
755 PH/02 The ipliteral router was not recognizing addresses of the form user@
756 [ipv6:....] because it didn't know about the "ipv6:" prefix.
758 PH/03 Added disable_ipv6.
760 PH/04 Changed $reply_address to use the raw form of the headers instead of the
761 decoded form, because it is most often used to construct To: headers
762 lines in autoreplies, and the decoded form may well be syntactically
763 invalid. However, $reply_address has leading white space removed, and all
764 newlines turned into spaces so that the autoreply transport does not
767 PH/05 If group was specified without a user on a router, and no group or user
768 was specified on a transport, the group from the router was ignored.
770 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
771 for visible compile-time settings that can be used to change these
772 numbers, for those that want even more. Backwards compatibility with old
773 spool files has been maintained. However, going back to a previous Exim
774 release will lost any variables that are in spool files.
776 PH/07 Two small changes when running in the test harness: increase delay when
777 passing a TCP/IP connection to a new process, in case the original
778 process has to generate a bounce, and remove special handling of
779 127.0.0.2 (sic), which is no longer necessary.
781 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
782 be the same on different OS.
784 PH/09 Moved a debug statement in filter processing to avoid a race problem when
787 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
788 whether --show-vars was specified or not
790 JJ/02 exipick: Added support for new ACL variable spool format introduced
793 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
794 syntactically invalid From: or Reply-to: line, and a filter used this to
795 generate an autoreply, and therefore failed to obtain an address for the
796 autoreply, Exim could try to deliver to a non-existent relative file
797 name, causing unrelated and misleading errors. What now happens is that
798 it logs this as a hard delivery error, but does not attempt to create a
801 PH/11 The exinext utility has a -C option for testing purposes, but although
802 the given file was scanned by exinext itself; it wasn't being passed on
803 when Exim was called.
805 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
806 an end-of-file indication when reading a command response.
808 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
809 compiled. In many other places in Exim, IPv6 addresses are always
810 recognized, so I have changed this. It also means that IPv4 domain
811 literals of the form [IPV4:n.n.n.n] are now always recognized.
813 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
814 used if the router is not running as root, for example, when verifying at
815 ACL time, or when using -bh. The debugging output from this situation was
816 non-existent - all you got was a failure to exec. I have made two
819 (a) Failures to set uid/gid, the current directory, or a process leader
820 in a subprocess such as that created by queryprogram now generate
821 suitable debugging ouput when -d is set.
823 (b) The queryprogram router detects when it is not running as root,
824 outputs suitable debugging information if -d is set, and then runs
825 the subprocess without attempting to change uid/gid.
827 PH/15 Minor change to Makefile for building test_host (undocumented testing
830 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
831 additional section of a DNS packet that returns MX or SRV records.
832 Instead, it always explicitly searches for A/AAAA records. This avoids
833 major problems that occur when a DNS server includes only records of one
834 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
835 fixed another bug: if SRV records were looked up and the corresponding
836 address records were *not* found in the additional section, the port
837 values from the SRV records were lost.
839 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
840 using the correct key (the original address) when searching the retry
841 rules in order to find which one to use for generating the retry hint.
843 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
844 adding the default one. Similarly, if it contains a Reply-To: header, the
845 errors_reply_to option, if set, is not used.
847 PH/19 When calculating a retry time, Exim used to measure the "time since
848 failure" by looking at the "first failed" field in the retry record. Now
849 it does not use this if it is later than than the arrival time of the
850 message. Instead it uses the arrival time. This makes for better
851 behaviour in cases where some deliveries succeed, thus re-setting the
852 "first failed" field. An example is a quota failure for a huge message
853 when small messages continue to be delivered. Without this change, the
854 "time since failure" will always be short, possible causing more frequent
855 delivery attempts for the huge message than are intended.
856 [Note: This change was subsequently modified - see PH/04 for 4.62.]
858 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
859 $1, $2, $3) because the numerical variables can be reset during some
860 expansion items (e.g. "match"), thereby losing the authentication data.
862 PH/21 Make -bV show the size of off_t variables so that the test suite can
863 decide whether to run tests for quotas > 2G.
865 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
866 mailbox_size, and mailbox_filecount in the appendfile transport. If a
867 filecount value is greater than 2G or if a quota value is greater than 2G
868 on a system where the size of off_t is not greater than 4, a panic error
871 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
872 never match. The debug and -bh output now contains an explicit error
873 message indicating a malformed IPv4 address or mask.
875 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
876 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
879 PH/25 Do not write to syslog when running in the test harness. The only
880 occasion when this arises is a failure to open the main or panic logs
881 (for which there is an explicit test).
883 PH/26 Added the /no_tell option to "control=freeze".
885 PH/27 If a host name lookup failed very early in a connection, for example, if
886 the IP address matched host_lookup and the reverse lookup yielded a name
887 that did not have a forward lookup, an error message of the form "no IP
888 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
889 could be logged. Now it outputs the IP address instead of "NULL".
891 PH/28 An enabling patch from MH: add new function child_open_exim2() which
892 allows the sender and the authenticated sender to be set when
893 submitting a message from within Exim. Since child_open_exim() is
894 documented for local_scan(), the new function should be too.
896 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
897 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
898 results in an empty string is now treated as unset.
900 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
902 PH/31 Added sender_verify_fail logging option.
904 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
905 needed by RFC 822 but not by RFC 2822 was commented out. I have now
906 tidied the source and removed it altogether.
908 PH/33 When a queue run was abandoned because the load average was too high, a
909 log line was always written; now it is written only if the queue_run log
910 selector is set. In addition, the log line for abandonment now contains
911 information about the queue run such as the pid. This is always present
912 in "start" and "stop" lines but was omitted from the "abandon" line.
914 PH/34 Omit spaces between a header name and the colon in the error message that
915 is given when verify = headers_syntax fails (if there are lots of them,
916 the message gets confusing).
918 PH/35 Change the default for dns_check_names_pattern to allow slashes within
919 names, as there are now some PTR records that contain slashes. This check
920 is only to protect against broken name servers that fall over on strange
921 characters, so the fact that it applies to all lookups doesn't matter.
923 PH/36 Now that the new test suite is complete, we can remove some of the
924 special code in Exim that was needed for the old test suite. For example,
925 sorting DNS records because real resolvers return them in an arbitrary
926 order. The new test suite's fake resolver always returns records in the
929 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
930 bounces) except when queue_only is set, to avoid logging races between
931 the different processes.
933 PH/38 Panic-die if .include specifies a non-absolute path.
935 PH/39 A tweak to the "H" retry rule from its user.
937 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
938 a label. They prevented compilation on older perls.
940 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
941 a warning to be raised on newish perls.
943 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
944 on queue. Changes to match documented behaviour of showing count of
945 messages matching specified criteria.
947 PH/40 Changed the default ident timeout from 30s to 5s.
949 PH/41 Added support for the use of login_cap features, on those BSD systems
950 that have them, for controlling the resources used by pipe deliveries.
952 PH/42 The content-scanning code uses fopen() to create files in which to put
953 message data. Previously it was not paying any attention to the mode of
954 the files. Exim runs with umask(0) because the rest of the code creates
955 files with open(), and sets the required mode explicitly. Thus, these
956 files were ending up world-writeable. This was not a big issue, because,
957 being within the spool directory, they were not world-accessible. I have
958 created a function called modefopen, which takes an additional mode
959 argument. It sets umask(777), creates the file, chmods it to the required
960 mode, then resets the umask. All the relevant calls to fopen() in the
961 content scanning code have been changed to use this function.
963 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
964 to 24 hours. This avoids potential overflow problems when processing G
965 and H retry rules. I suspect nobody ever tinkers with this value.
967 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
969 PH/45 When the plaintext authenticator is running as a client, the server's
970 challenges are checked to ensure they are valid base64 strings. By
971 default, the authentication attempt is cancelled if an invalid string is
972 received. Setting client_ignore_invalid_base64 true ignores these errors.
973 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
974 they are received. Thus, the responses can be made to depend on the
975 challenges. If an invalid string is ignored, an empty string is placed in
978 PH/46 Messages that are created by the autoreply transport now contains a
979 References: header, in accordance with RFCs 2822 and 3834.
981 PH/47 Added authenticated_sender_force to the smtp transport.
983 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
985 PH/49 Installed latest patch from the Sieve maintainer.
987 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
988 was also set, the appendfile transport was unnecessarily scanning a
989 directory of message files (e.g. for maildir delivery) to find the count
990 of files (along with the size), even though it did not need this
991 information. It now does the scan only if it needs to find either the
992 size of the count of files.
994 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
996 PH/52 Two bugs concerned with error handling when the smtp transport is
999 (i) Exim was not creating retry information for temporary errors given
1000 for individual recipients after the DATA command when the smtp transport
1001 was used in LMTP mode. This meant that they could be retried too
1002 frequently, and not timed out correctly.
1004 (ii) Exim was setting the flag that allows error details to be returned
1005 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1006 recipients that were returned after the DATA command.
1008 PH/53 This is related to PH/52, but is more general: for any failing address,
1009 when detailed error information was permitted to be returned to the
1010 sender, but the error was temporary, then after the final timeout, only
1011 "retry timeout exceeded" was returned. Now it returns the full error as
1012 well as "retry timeout exceeded".
1014 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1015 do this, and (what is worse) MTAs that accept it.
1017 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1018 will now be deprecated.
1020 PH/56 New os.c-cygwin from the Cygwin maintainer.
1022 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1023 formats (previously only available in exim formats via -bpr, -bpru,
1024 and -bpra. Now also available in native and exiqgrep formats)
1026 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1027 with very large, slow to parse queues
1029 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1031 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1033 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1034 responses to authentication challenges, though it was showing the
1035 challenges; (ii) I've removed the CR characters from the debug output for
1038 PH/58 Allow for the insertion of a newline as well as a space when a string
1039 is turned into more than one encoded-word during RFC 2047 encoding. The
1040 Sieve code now uses this.
1042 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1043 data_4xx, lost_connection, tls_required.
1045 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1046 message was being sent as an SMTP response.
1048 PH/61 Add -l and -k options to exicyclog.
1050 PH/62 When verifying, if an address was redirected to one new address, so that
1051 verification continued, and the new address failed or deferred after
1052 having set something in $address_data, the value of $address_data was not
1053 passed back to the ACL. This was different to the case when no
1054 redirection occurred. The value is now passed back in both cases.
1056 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1057 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1058 use login_cap.h, so on its own it isn't the distinguishing feature. The
1059 new name refers directly to the setclassresources() function.
1061 PH/65 Added configuration files for NetBSD3.
1063 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1065 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1066 is preferred over IPv4.
1068 PH/68 The bounce_return_message and bounce_return_body options were not being
1069 honoured for bounces generated during the reception of non-SMTP messages.
1070 In particular, this applied to messages rejected by the ACL. This bug has
1071 been fixed. However, if bounce_return_message is true and bounce_return_
1072 body is false, the headers that are returned for a non-SMTP message
1073 include only those that have been read before the error was detected.
1074 (In the case of an ACL rejection, they have all been read.)
1076 PH/69 The HTML version of the specification is now built in a directory called
1077 spec_html instead of spec.html, because the latter looks like a path with
1078 a MIME-type, and this confuses some software.
1080 PH/70 Catch two compiler warnings in sieve.c.
1082 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1083 function verify_get_ident() calls ip_connect() to connect a socket, but
1084 if the "connect()" function timed out, ip_connect() used to close the
1085 socket. However, verify_get_ident() also closes the socket later, and in
1086 between Exim writes to the log, which may get opened at this point. When
1087 the socket was closed in ip_connect(), the log could get the same file
1088 descriptor number as the socket. This naturally causes chaos. The fix is
1089 not to close the socket in ip_connect(); the socket should be closed by
1090 the function that creates it. There was only one place in the code where
1091 this was missing, in the iplookup router, which I don't think anybody now
1092 uses, but I've fixed it anyway.
1094 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1095 well as to direct DNS lookups. Otherwise the handling of names in host
1096 lists is inconsistent and therefore confusing.
1102 PH/01 Two changes to the default runtime configuration:
1104 (1) Move the checks for relay_from_hosts and authenticated clients from
1105 after to before the (commented out) DNS black list checks.
1107 (2) Add control=submission to the relay_from_hosts and authenticated
1108 clients checks, on the grounds that messages accepted by these
1109 statements are most likely to be submissions.
1111 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1113 (1) Generate an error if the third argument for the ${prvs expansion is
1116 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1119 (3) Reset the variables that are obtained from the first argument of
1120 ${prvscheck and used in the second argument before leaving the code,
1121 because their memory is reclaimed, so using them afterwards may do
1124 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1125 one (it's much easier than Tom thought :-).
1127 (5) Because of (4), we can now allow for the use of $prvscheck_result
1128 inside the third argument.
1130 PH/03 For some reason, the default setting of PATH when running a command from
1131 a pipe transport was just "/usr/bin". I have changed it to
1134 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1135 anything to be listed in the output from -bV.
1137 PH/05 When a filter generated an autoreply, the entire To: header line was
1138 quoted in the delivery log line, like this:
1140 => >A.N.Other <ano@some.domain> <original@ddress> ...
1142 This has been changed so that it extracts the operative address. There
1143 may be more than one such address. If so, they are comma-separated, like
1146 => >ano@some.domain,ona@other.domain <original@ddress> ...
1148 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1149 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1150 not being looked up in the rDNS to get a host name, Exim was showing the
1151 IP address twice in Received: lines, even though the IP addresses were
1152 identical. For example:
1154 Received: from [1.2.3.4] (helo=[1.2.3.4])
1156 However, if the real host name was known, it was omitting the HELO data
1157 if it matched the actual IP address. This has been tidied up so that it
1158 doesn't show the same IP address twice.
1160 PH/07 When both +timestamp and +memory debugging was on, the value given by
1161 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1162 called by the debug printing, thereby overwriting the timestamp buffer.
1163 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1165 PH/08 When the original message was included in an autoreply transport, it
1166 always said "this is a copy of the message, including all the headers",
1167 even if body_only or headers_only was set. It now gives an appropriate
1170 PH/09 Applied a patch from the Sieve maintainer which:
1172 o fixes some comments
1173 o adds the (disabled) notify extension core
1174 o adds some debug output for the result of if/elsif tests
1175 o points to the current vacation draft in the documentation
1176 and documents the missing references header update
1180 o fixes a bug in processing the envelope test (when testing
1181 multiple envelope elements, the last element determinted the
1184 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1185 Electronic Mail") by including:
1187 Auto-submitted: auto-generated
1189 in the messages that it generates (bounce messages and others, such as
1190 warnings). In the case of bounce messages for non-SMTP mesages, there was
1191 also a typo: it was using "Auto_submitted" (underscore instead of
1192 hyphen). Since every message generated by Exim is necessarily in response
1193 to another message, thes have all been changed to:
1195 Auto-Submitted: auto-replied
1197 in accordance with these statements in the RFC:
1199 The auto-replied keyword:
1201 - SHOULD be used on messages sent in direct response to another
1202 message by an automatic process,
1204 - MUST NOT be used on manually-generated messages,
1206 - MAY be used on Delivery Status Notifications (DSNs) and Message
1207 Disposition Notifications (MDNs),
1209 - MUST NOT be used on messages generated by automatic or periodic
1210 processes, except for messages which are automatic responses to
1213 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1214 to the default Received: header definition.
1216 PH/12 Added log selector acl_warn_skipped (default on).
1218 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1219 variables because (a) they are in the wrong storage pool and (b) even if
1220 they were copied, it wouldn't work properly because of the caching.
1222 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1223 checking when decoding. Apparently there are clients that generate
1224 overlong encoded strings. Why am I not surprised?
1226 PH/15 If the first argument of "${if match_address" was not empty, but did not
1227 contain an "@" character, Exim crashed. Now it writes a panic log message
1228 and treats the condition as false.
1230 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1232 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1233 extension "envelope-auth". The code is finished and in agreement with
1234 other implementations, but there is no documentation so far and in fact,
1235 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1236 not changing the active code.
1238 Print executed "if" and "elsif" statements when debugging is used. This
1239 helps a great deal to understand what a filter does.
1241 Document more things not specified clearly in RFC3028. I had all this
1242 sorted out, when out of a sudden new issues came to my mind. Oops."
1244 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1247 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1248 canonical form (as documented). However, after a host name lookup from
1249 the IP address, check_host() was doing a simple string comparison with
1250 addresses acquired from the DNS when checking that the found name did
1251 have the original IP as one of its addresses. Since any found IPv6
1252 addresses are likely to be in abbreviated form, the comparison could
1253 fail. Luckily, there already exists a function for doing the comparison
1254 by converting both addresses to binary, so now that is used instead of
1255 the text comparison.
1257 PH/20 There was another similar case to PH/19, when a complete host name was
1258 given in a host list; looking up its IP address could give an abbreviated
1259 form, whereas the current host's name might or might not be abbreviated.
1260 The same fix has been applied.
1266 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1267 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1270 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1271 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1273 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1275 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1276 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1277 Also, for "Auto-Submitted", treat anything other than "no" as
1278 non-personal, in accordance with RFC 3834. (Previously it treated
1279 anything starting "auto-" as non-personal.)
1281 TF/01 The control=submission/name=... option had a problem with syntax
1282 errors if the name included a slash character. The /name= option
1283 now slurps the rest of the string, so it can include any characters
1284 but it must come last in the list of options (after /sender_retain
1287 PH/05 Some modifications to the interface to the fake nameserver for the new
1295 TK/01 Added the "success_on_redirect" address verification option. See
1296 NewStuff for rationale and an example.
1298 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1300 PH/02 Patch to exigrep to allow it to work on syslog lines.
1302 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1303 fread() to skip over the body file's header line, because in Cygwin the
1304 header line is locked and is inaccessible.
1306 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1307 co-exist for some time) to make it clear that it is the Exim ID that is
1308 referenced, not the Message-ID: header line.
1310 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1311 string_format() function, because snprintf() does not exist on all
1314 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1315 ${stat: expansion item.
1317 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1318 protocol synchronization error", to keep the pedants happy.
1320 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1321 well as for IRIX systems, when gcc is being used. See the host.c source
1324 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1326 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1329 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1330 transport and to the smtp transport in LMTP mode.
1332 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1334 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1335 run a filter in a subprocess. This could lead to confusion in subsequent
1336 lookups in the parent process. There should also be a search_tidyup() at
1337 the end of the subprocess.
1339 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1340 only if the host matched helo_try_verify_hosts, which caused the
1341 verification to occur when the EHLO/HELO command was issued. The ACL just
1342 tested the remembered result. Now, if a previous verification attempt has
1343 not happened, "verify = helo" does it there and then.
1345 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1347 TK/03 Fix log output including CR from clamd.
1349 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1350 exist provoked a memory error which could cause a segfault.
1352 PH/15 Installed PCRE 6.2
1354 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1356 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1357 of the problem. Specifically, suggested +O2 rather than +O1 for the
1360 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1362 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1363 record, the port was not correctly logged when the outgoing_port log
1364 selector was set (it logged the transort's default port).
1366 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1367 fallback_hosts, and "hosts" in the smtp transport.
1369 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1370 host errors such as "Connection refused".
1372 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1373 authentication with radiusclient 0.4.9:
1375 - Error returned from rc_read_config was caught wrongly
1376 - Username/password not passed on to radius server due to wrong length.
1378 The presumption is that some radiusclient API changes for 4.51/PH/17
1379 were not taken care of correctly. The code is still untested by me (my
1380 Linux distribution still has 0.3.2 of radiusclient), but it was
1381 contributed by a Radius user.
1383 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1384 expanding the "port" option of the smtp transport.
1386 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1387 while reading a MIME header. Thanks to Tom Hughes for a patch.
1389 PH/24 Include config.h inside local_scan.h so that configuration settings are
1392 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1393 that in an ACL for RCPT (for example), you can examine exactly what was
1396 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1397 commands, but it was not correctly comparing the address with the actual
1398 client host address. Thus, it would show the EHLO address in Received:
1399 header lines when this was not necessary.
1401 PH/27 Added the % operator to ${eval:}.
1403 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1404 it should be ignoring failures (because with -C, for example, it has lost
1405 privilege). It wasn't ignoring creation failures other than "already
1408 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1411 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1412 because an input line was too long, either on its own, or by virtue of
1413 too many continuations, the temporary file was not being removed, and the
1414 return code was incorrect.
1416 PH/31 Missing "BOOL" in function definition in filtertest.c.
1418 PH/32 Applied Sieve patches from the maintainer.
1420 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1422 PH/33 Added "verify = not_blind".
1424 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1425 Local/Makefile (with some defaults set). These are used in built scripts
1426 such as exicyclog, but they have never been used in the exim_install
1427 script (though there are many overriding facilities there). I have
1428 arranged that the exim_install script now takes note of these two
1431 PH/35 Installed configuration files for Dragonfly.
1433 PH/36 When a locally submitted message by a trusted user did not contain a
1434 From: header, and the sender address was obtained from -f or from an SMTP
1435 MAIL command, and the trusted user did not use -F to supply a sender
1436 name, $originator_name was incorrectly used when constructing a From:
1437 header. Furthermore, $originator_name was used for submission mode
1438 messages from external hosts without From: headers in a similar way,
1439 which is clearly wrong.
1441 PH/37 Added control=suppress_local_fixups.
1443 PH/38 When log_selector = +received_sender was set, and the addition of the
1444 sender made the log line's construction buffer exactly full, or one byte
1445 less than full, an overflow happened when the terminating "\n" was
1448 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1449 when the result of a list match is failure because a DNS lookup failed.
1451 PH/40 RM_COMMAND is now used in the building process.
1453 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1454 the "build-* directories that it finds.
1456 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1457 address in a domain literal was a prefix of an interface address.
1459 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1460 when verifying a sender address, unless rewrite_headers is false.
1462 PH/44 Wrote a long comment about why errors_to addresses are verified as
1463 recipients, not senders.
1465 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1466 the ratelimit ACL was added.
1468 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1470 PH/46 Added extra information about PostgreSQL errors to the error string.
1472 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1473 avoiding the need to install special zones in a real server. This is
1474 backwards compatible; if it can't find the fake resolver, it drops back.
1475 Thus, both old and new test suites can be run.
1477 TF/02 Added util/ratelimit.pl
1479 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1480 clock is set back in time.
1482 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1483 Candler <B.Candler@pobox.com>.
1485 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1486 OK for addresses that are the result of redirections.
1488 PH/48 A number of further additions for the benefit of the new test suite,
1489 including a fake gethostbyname() that interfaces to the fake DNS resolver
1492 TF/06 The fix for widen_domains has also been applied to qualify_single and
1493 search_parents which are the other dnslookup options that can cause
1496 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1499 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1501 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1502 service" if it encountered an MX record with an empty target hostname.
1503 The message is now "an MX or SRV record indicated no SMTP service".
1505 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1506 if the DNS of the sending site is misconfigured. This is quite a
1507 common situation. This change restores the behaviour of treating a
1508 helo verification defer as a failure.
1510 PH/51 If self=fail was set on a router, the bounce message did not include the
1511 actual error message.
1517 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1519 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1520 command itself ended in error, the underlying message about the transport
1521 filter timeout was being overwritten with the pipe command error. Now the
1522 underlying error message should be appended to the second error message.
1524 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1526 PH/02 Reduce the amount of output that "make" produces by default. Full output
1527 can still be requested.
1529 PH/03 The warning log line about a condition test deferring for a "warn" verb
1530 was being output only once per connection, rather than after each
1531 occurrence (because it was using the same function as for successful
1532 "warn" verbs). This seems wrong, so I have changed it.
1534 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1535 it should not have, which might have caused a crash in the right
1536 circumstances, but probably never did.
1538 PH/04 Installed a modified version of Tony Finch's patch to make submission
1539 mode fix the return path as well as the Sender: header line, and to
1540 add a /name= option so that you can make the user's friendly name appear
1543 TF/03 Added the control = fakedefer ACL modifier.
1545 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1546 Mark Lowes for thorough testing.
1548 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1550 TK/03 Merged latest SRS patch from Miles Wilton.
1552 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1553 with the definition in sysexits.h (which is #included earlier).
1554 Fortunately, Exim does not actually use EX_OK. The code used to try to
1555 preserve the sysexits.h value, by assumimg that macro definitions were
1556 scanned for macro replacements. I have been disabused of this notion,
1557 so now the code just undefines EX_OK before #including unistd.h.
1559 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1560 in the smtp transport. When a block could not be written in a single
1561 write() function, the timeout was being re-applied to each part-write.
1562 This seems wrong - if the receiver was accepting one byte at a time it
1563 would take for ever. The timeout is now adjusted when this happens. It
1564 doesn't have to be particularly precise.
1566 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1567 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1569 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1570 without a domain if the check to <postmaster@domain> fails.
1572 SC/01 Eximstats: added -xls and the ability to specify output files
1573 (patch written by Frank Heydlauf).
1575 SC/02 Eximstats: use FileHandles for outputing results.
1577 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1579 SC/04 Eximstats: fixed display of large numbers with -nvr option
1581 SC/05 Eximstats: fixed merging of reports with empty tables.
1583 SC/06 Eximstats: added the -include_original_destination flag
1585 SC/07 Eximstats: removed tabs and trailing whitespace.
1587 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1589 TK/06 MBOX spool code: Add real "From " MBOX separator line
1590 so the .eml file is really in mbox format (even though
1591 most programs do not really care). Patch from Alex Miller.
1593 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1594 The latter is generated from $received_to and is only set if the
1595 message has one envelope recipient. SA can use these headers,
1596 obviously out-of-the-box. Patch from Alex Miller.
1598 PH/08 The ${def test on a variable was returning false if the variable's
1599 value was "0", contrary to what the specification has always said!
1600 The result should be true unless the variable is empty.
1602 PH/09 The syntax error of a character other than { following "${if
1603 def:variable_name" (after optional whitespace) was not being diagnosed.
1604 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1605 accidental colon was present, for example, could give incorrect results.
1607 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1608 result is used (not including appendfile, where other changes are about
1611 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1612 This involved changing a lot of size variables from int to off_t. It
1613 should work with maildirs and everything.
1615 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1616 spamd dying while we are connected to it.
1618 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1621 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1622 function for BDB 4.3.
1624 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1626 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1627 the sources for PCRE have been re-arranged and more files are now
1630 PH/15 The code I had for printing potentially long long variables in PH/11
1631 above was not the best (it lost precision). The length of off_t variables
1632 is now inspected at build time, and an appropriate printing format (%ld
1633 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1634 to be "long long int" or "long int". This is needed for the internal
1635 formatting function string_vformat().
1637 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1638 the configuration file to be ":syslog", then the script "guesses" where
1639 the logs files are, rather than using the compiled in default. In our
1640 case the guess is not the same as the compiled default, so the script
1641 suddenly stopped working when I started to use syslog. The patch checks
1642 to see if log_file_path is "". If so, it attempts to read it from exim
1643 with no configuration file to get the compiled in version, before it
1644 falls back to the previous guessing code."
1646 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1647 implementing BATV in an Exim configuration. See NewStuff for the gory
1650 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1651 Makefile that are specific to HP-UX.
1653 PH/18 If the "use_postmaster" option was set for a recipient callout together
1654 with the "random" option, the postmaster address was used as the MAIL
1655 FROM address for the random test, but not for the subsequent recipient
1656 test. It is now used for both.
1658 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1659 patch removes a few documentation additions to RFC 3028, because the
1660 latest draft now contains them. It adds the new en;ascii-case comparator
1661 and a new error check for 8bit text in MIME parts. Comparator and
1662 require names are now matched exactly. I enabled the subaddress
1663 extension, but it is not well tested yet (read: it works for me)."
1665 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1666 rework some of the code of TK/09 above to avoid the hardwired use of
1667 "%lld" and "long long". Replaced the call to snprintf() with a call to
1670 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1671 records point to non-existent hosts", "retry timeout exceeded", and
1672 "retry time not reached for any host after a long failure period".
1674 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1675 experimental DomainKeys support:
1677 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1678 (2) On an error such as an illegally used "control", the wrong name for
1679 the control was given.
1681 These problems did NOT occur unless DomainKeys support was compiled.
1683 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1685 PH/24 Added ${if match_ip condition.
1687 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1688 databases so that it will be absolutely obvious if a crash occurs in the
1689 DB library. This is a regular occurrence (often caused by mis-matched
1692 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1693 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1694 picked up on a user's system that detects such things. There doesn't seem
1695 to be a gcc warning option for this - only an attribute that has to be
1696 put on the function's prototype. It seems that in Fedora Core 4 they have
1697 set this on a number of new functions. No doubt there will be more in due
1700 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1701 specify a transport. However, if an address that was verified by such a
1702 router was the subject of a callout, Exim crashed because it tried to
1703 read the rcpt_include_affixes from the non-existent transport. Now it
1704 just assumes that the setting of that option is false. This bug was
1705 introduced by 4.51/PH/31.
1707 PH/28 Changed -d+all to exclude +memory, because that information is very
1708 rarely of interest, but it makes the output a lot bigger. People tend to
1709 do -d+all out of habit.
1711 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1712 code in os-type was giving problems when libc.so lives in lib64, like on
1715 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1716 aren't the modern standard, and it seems that some systems' include files
1717 don't always have them. Exim was already checking for some of the newer
1718 ones like T_AAAA, and defining it itself. I've added checks for all the
1719 record types that Exim uses.
1721 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1722 not automatically generating a new one, as it is supposed to. This
1723 prevented TLS from working. If the file did exist, but contained invalid
1724 data, a new version was generated, as expected. It was only the case of a
1725 non-existent file that was broken.
1727 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1728 with a change in libdomainkeys > 0.64.
1730 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1731 from DNS. If the selector record carries the flag, it now has
1732 precedence over the domain-wide flag.
1734 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1736 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1737 the use of a port name that isn't defined in /etc/services) occurred, the
1738 message was deferred as in a normal delivery, and thus remained on the
1739 spool, instead of being failed because of the mua_wrapper setting. This
1740 is now fixed, and I tidied up some of the mua_wrapper messages at the
1743 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1744 the messages in a hash of arrays rather than using individual hashes.
1745 This is a bit cleaner and results in dramatic memory savings, albeit
1746 at a slight CPU cost.
1748 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1749 as requested by Marc Sherman.
1751 SC/10 Eximstats: added histograms for user specified patterns as requested
1754 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1756 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1757 fopen() in the content-scanning modules that did not already have it.
1763 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1764 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1766 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1768 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1770 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1771 to test Sieve filters that use "vacation".
1773 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1774 that changes the way the GnuTLS parameters are stored in the cache file.
1775 The new format can be generated externally. For backward compatibility,
1776 if the data in the cache doesn't make sense, Exim assumes it has read an
1777 old-format file, and it generates new data and writes a new file. This
1778 means that you can't go back to an older release without removing the
1781 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1782 work if there are any delivery delays because "one_time" forces the
1783 parent to be marked "delivered", so its unseen clone is never tried
1784 again. For this reason, Exim now forbids the simultaneous setting of
1787 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1788 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1789 entry if you want to know the details. The fix, however, neglected to
1790 consider the case when local delivery batching is involved. The test for
1791 "previously delivered" was not happening when checking to see if an
1792 address could be batched with a previous (undelivered) one; under
1793 certain circumstances this could lead to multiple deliveries to the same
1796 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1797 in its include files, and this causes problems building Exim.
1799 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1800 header_syntax) but Exim was just ignoring anything given after a slash.
1801 In particular, this caused confusion with an attempt to use "verify =
1802 reverse_host_lookup/defer_ok". An error is now given when options are
1803 supplied for verify items that do not have them. (Maybe reverse_host_
1804 lookup should have a defer_ok option, but that's a different point.)
1806 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1807 defined by RFC 821) to 2048, because there were problems with some AUTH
1808 commands, and RFC 1869 says the size should be increased for extended
1809 SMTP commands that take arguments.
1811 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1814 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1815 "unknown" error; now it says that the functionality isn't in the binary.
1817 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1818 an address' error message when a string expansion fails (syntax or
1819 whatever). Otherwise the password may appear in the log. Following change
1820 PH/42 below, there is no longer a chance of it appearing in a bounce
1823 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1825 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1826 of its addresses was ever tried. (Bugzilla bug #2.)
1828 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1829 the result incorrectly in the debug output. (It correctly added a newline
1830 to what was transported.)
1832 TF/01 Added $received_time.
1834 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1835 commented out examples of how to interface to a virus scanner and to
1836 SpamAssassin. Also added commented examples of av_scanner and
1837 spamd_address settings.
1839 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1840 and controls are allowed in which ACLs. There were a couple of minor
1841 errors. Some of the entries in the conditions table (which is a table of
1842 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1843 negation of where the condition IS allowed.
1845 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1847 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1848 header file does not have a version number, so I've had to invent a new
1849 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1850 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1851 radiusclient), but it was contributed by a Radius user.
1853 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1854 files or directories, for OpenSSL.
1856 PH/19 When an Exim process that is running as root has to create an Exim log
1857 file, it does so in a subprocess that runs as exim:exim so as to get the
1858 ownership right at creation (otherwise, other Exim processes might see
1859 the file with the wrong ownership). There was no test for failure of this
1860 fork() call, which would lead to the process getting stuck as it waited
1861 for a non-existent subprocess. Forks do occasionally fail when resources
1862 run out. I reviewed all the other calls to fork(); they all seem to check
1865 PH/20 When checking for unexpected SMTP input at connect time (before writing
1866 the banner), Exim was not dealing correctly with a non-positive return
1867 from the read() function. If the client had disconnected by this time,
1868 the result was a log entry for a synchronization error with an empty
1869 string after "input=" when read() returned zero. If read() returned -1
1870 (an event I could not check), uninitialized data bytes were printed.
1871 There were reports of junk text (parts of files, etc) appearing after
1874 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1876 PH/22 Added support for macro redefinition, and (re)definition in between
1877 driver and ACL definitions.
1879 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1880 forgetting to use the resulting value; it was using the unexpanded value.
1882 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1883 hadn't been configured. The fix is from Juergen Kreileder, who
1884 understands it better than I do:
1886 "Here's what I see happening with three configured cyrus_sasl
1887 authenticators configured (plain, login, cram-md5):
1889 On startup auth_cyrus_sasl_init() gets called for each of these.
1890 This means three calls to sasl_listmech() without a specified mech_list.
1891 => SASL tests which mechs of all available mechs actually work
1892 => three warnings about OTP not working
1893 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1895 With the patch, sasl_listmech() also gets called three times. But now
1896 SASL's mech_list option is set to the server_mech specified in the the
1897 authenticator. Or in other words, the answer from sasl_listmech()
1898 gets limited to just the mech you're testing for (which is different
1900 => the return list contains just 'plain' or 'login', 'cram-md5' or
1901 nothing depending on the value of ob->server_mech.
1903 I've just tested the patch: Authentication still works fine,
1904 unavailable mechs specified in the exim configuration are still
1905 caught, and the auth.log warnings about OTP are gone."
1907 PH/25 When debugging is enabled, the contents of the command line are added
1908 to the debugging output, even when log_selector=+arguments is not
1911 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1912 answer is "GNU", and only if the return is "GNU/something" is the answer
1915 PH/27 $acl_verify_message is now set immediately after the failure of a
1916 verification in an ACL, and so is available in subsequent modifiers. In
1917 particular, the message can be preserved by coding like this:
1919 warn !verify = sender
1920 set acl_m0 = $acl_verify_message
1922 Previously, $acl_verify_message was set only while expanding "message"
1923 and "log_message" when a very denied access.
1925 PH/28 Modified OS/os.c-Linux with
1927 -#ifndef OS_LOAD_AVERAGE
1928 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1930 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1931 nomenclature these days.)
1933 PH/29 Installed patch from the Sieve maintainer that adds the options
1934 sieve_useraddress and sieve_subaddress to the redirect router.
1936 PH/30 In these circumstances:
1937 . Two addresses routed to the same list of hosts;
1938 . First host does not offer TLS;
1939 . First host accepts first address;
1940 . First host gives temporary error to second address;
1941 . Second host offers TLS and a TLS session is established;
1942 . Second host accepts second address.
1943 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1944 and peerdn, if requested) that were in fact used only for the second
1947 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1948 attention to any local part prefix or suffix that was matched by the
1949 router that accepted the address. It now behaves in the same way as it
1950 does for delivery: the affixes are removed from the local part unless
1951 rcpt_include_affixes is set on the transport.
1953 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1954 timeout during the DATA phase of an incoming message.
1956 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1957 applied a patch sanctioned by the Sieve maintainer.
1959 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1960 the uid or gid is negative. A case of a negative gid caused this to be
1961 noticed. The fix allows for either to be negative.
1963 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1964 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1967 PH/36 The change PH/12 above was broken. Fixed it.
1969 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1970 the grounds that routing the same address twice would always produce the
1971 same answer. This might have been true once, but it is certainly no
1972 longer true now. Routing a child address may depend on the previous
1973 routing that produced that child. Some complicated redirection strategies
1974 went wrong when messages had multiple recipients, and made Exim's
1975 behaviour dependent on the order in which the addresses were given.
1977 I have moved the duplicate checking until after the routing is complete.
1978 Exim scans the addresses that are assigned to local and remote
1979 transports, and removes any duplicates. This means that more work will be
1980 done, as duplicates will always all be routed, but duplicates are
1981 presumably rare, so I don't expect this is of any significance.
1983 For deliveries to pipes, files, and autoreplies, the duplicate checking
1984 still happens during the routing process, since they are not going to be
1987 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1988 It corrects a timeout issue with spamd. This is Ian's comment: "The
1989 background is that sometimes spamd either never reads data from a
1990 connection it has accepted, or it never writes response data. The exiscan
1991 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1992 blindly assumes that writes won't block so it may never time out."
1994 PH/39 Allow G after quota size as well as K and M.
1996 PH/40 The value set for $authenticated_id in an authenticator may not contain
1997 binary zeroes or newlines because the value is written to log lines and
1998 to spool files. There was no check on this. Now the value is run through
1999 the string_printing() function so that such characters are converted to
2000 printable escape sequences.
2002 PH/41 $message_linecount is a new variable that contains the total number of
2003 lines in the message. Compare $body_linecount, which is the count for the
2006 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2007 bounce and delay warning messages, except in certain special cases, which
2010 (a) An SMTP error message from a remote host;
2011 (b) A message specified in a :fail: redirection;
2012 (c) A message specified in a "fail" command in a system filter;
2013 (d) A message specified in a FAIL return from the queryprogram router;
2014 (e) A message specified by the cannot_route_message router option.
2016 In these cases only, Exim does include the error details in bounce and
2017 warning messages. There are also a few cases where bland messages such
2018 as "unrouteable address" or "local delivery error" are given.
2020 PH/43 $value is now also set for the "else" part of a ${run expansion.
2022 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2023 being worked on, but at least Exim now implements the latest version to
2026 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2027 process to complete was treated as a delivery failure, a timeout while
2028 writing the message to the pipe was logged, but erroneously treated as a
2029 successful delivery. Such timeouts include transport filter timeouts. For
2030 consistency with the overall process timeout, these timeouts are now
2031 treated as errors, giving rise to delivery failures by default. However,
2032 there is now a new Boolean option for the pipe transport called
2033 timeout_defer, which, if set TRUE, converts the failures into defers for
2034 both kinds of timeout. A transport filter timeout is now identified in
2037 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2038 systems where "make" and "gmake" are different, calling "gmake" at top
2039 level broke things. I've arranged for the value of $(MAKE) to be passed
2040 from the Makefile to this script so that it can call the same version of
2044 A note about Exim versions 4.44 and 4.50
2045 ----------------------------------------
2047 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2048 changes of various kinds. As a consequence, a big documentation update was
2049 needed. This delayed the release for rather longer than seemed good, especially
2050 in the light of a couple of (minor) security issues. Therefore, the changes
2051 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2052 release. So 4.44 and 4.50 are in effect two different branches that both start
2055 I have left the 4.50 change log unchanged; it contains all the changes since
2056 4.43. The change log for 4.44 is below; many of its items are identical to
2057 those for 4.50. This seems to be the most sensible way to preserve the
2058 historical information.
2064 1. Minor wording change to the doc/README.SIEVE file.
2066 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2067 computation of the current number of files was incorrect.
2069 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2070 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2073 4. Give more explanation in the error message when the command for a transport
2074 filter fails to execute.
2076 5. There are several places where Exim runs a non-Exim command in a
2077 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2078 was being done only for the command run by the queryprogram router. It is
2079 now done for all such subprocesses. The other cases are: ${run, transport
2080 filters, and the commands run by the lmtp and pipe transports.
2082 6. Added CONFIGURE_GROUP build-time option.
2084 7. Some older OS have a limit of 256 on the maximum number of file
2085 descriptors. Exim was using setrlimit() to set 1000 as a large value
2086 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2087 systems. I've change it so that if it can't get 1000, it tries for 256.
2089 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2090 was an oversight, and furthermore, ever since the addition of extra
2091 controls (e.g. 4.43/32), the checks on when to allow different forms of
2092 "control" were broken. There should now be diagnostics for all cases when a
2093 control that does not make sense is encountered.
2095 9. Added the /retain_sender option to "control=submission".
2097 10. $recipients is now available in the predata ACL (oversight).
2099 11. Tidy the search cache before the fork to do a delivery from a message
2100 received from the command line. Otherwise the child will trigger a lookup
2101 failure and thereby defer the delivery if it tries to use (for example) a
2102 cached ldap connection that the parent has called unbind on.
2104 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2105 of $address_data from the recipient verification was clobbered by the
2106 sender verification.
2108 13. The value of address_data from a sender verification is now available in
2109 $sender_address_data in subsequent conditions in the ACL statement.
2111 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2113 15. Added a new option "connect=<time>" to callout options, to set a different
2116 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2117 was its contents. (It was OK if the option was not defined at all.)
2119 17. A "Completed" log line is now written for messages that are removed from
2120 the spool by the -Mrm option.
2122 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2123 information about exactly what failed.
2125 19. Added -dd to debug only the daemon process.
2127 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2128 handles timeouts, both on the server side and network timeouts. Renamed the
2129 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2131 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2132 It is now set to "smtps".
2134 22. $host_address is now set to the target address during the checking of
2135 ignore_target_hosts.
2137 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2138 being passed; this would have caused $sender_host_name to have been used if
2139 matching the list had actually called for a host name (not very likely,
2140 since this list is usually IP addresses). A host name is now passed as
2143 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2144 code that specifies a non-restarting handler (typically sigaction() in
2145 modern systems) in an attempt to fix a rare and obscure crash bug.
2147 25. Narrowed the window for a race in the daemon that could cause it to ignore
2148 SIGCHLD signals. This is not a major problem, because they are used only to
2149 wake it up if nothing else does.
2151 26. A malformed maildirsize file could cause Exim to calculate negative values
2152 for the mailbox size or file count. Odd effects could occur as a result.
2153 The maildirsize information is now recalculated if the size or filecount
2156 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2157 support for a long time. Removed HAVE_SYS_VFS_H.
2159 28. Installed the latest version of exipick from John Jetmore.
2161 29. In an address list, if the pattern was not a regular expression, an empty
2162 subject address (from a bounce message) matched only if the pattern was an
2163 empty string. Non-empty patterns were not even tested. This was the wrong
2164 because it is perfectly reasonable to use an empty address as part of a
2165 database query. An empty address is now tested by patterns that are
2166 lookups. However, all the other forms of pattern expect the subject to
2167 contain a local part and a domain, and therefore, for them, an empty
2168 address still always fails if the pattern is not itself empty.
2170 30. Exim went into a mad DNS loop when attempting to do a callout where the
2171 host was specified on an smtp transport, and looking it up yielded more
2172 than one IP address.
2174 31. Re-factored the code for checking spool and log partition space into a
2175 function that finds that data and another that does the check. The former
2176 is then used to implement four new variables: $spool_space, $log_space,
2177 $spool_inodes, and $log_inodes.
2179 32. The RFC2047 encoding function was originally intended for short strings
2180 such as real names; it was not keeping to the 75-character limit for
2181 encoded words that the RFC imposes. It now respects the limit, and
2182 generates multiple encoded words if necessary. To be on the safe side, I
2183 have increased the buffer size for the ${rfc2047: expansion operator from
2186 33. It is now permitted to omit both strings after an "if" condition; if the
2187 condition is true, the result is "true". As before, when the second string
2188 is omitted, a false condition yields an empty string. This makes it less
2189 cumbersome to write custom ACL and router conditions.
2191 34. Failure to deliver a bounce message always caused it to be frozen, even if
2192 there was an errors_to setting on the router. The errors_to setting is now
2195 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2196 canonical form (fully expanded) before being placed in
2197 $sender_host_address.
2199 36. The table in the code that translates DNS record types into text (T_A to
2200 "A" for instance) was missing entries for NS and CNAME. It is just possible
2201 that this could have caused confusion if both these types were looked up
2202 for the same domain, because the text type is used as part of Exim's
2203 per-process caching. But the chance of anyone hitting this buglet seems
2206 37. The dnsdb lookup has been extended in a number of ways.
2208 (1) There is a new type, "zns", which walks up the domain tree until it
2209 finds some nameserver records. It should be used with care.
2211 (2) There is a new type, "mxh", which is like "mx" except that it returns
2212 just the host names, not the priorities.
2214 (3) It is now possible to give a list of domains (or IP addresses) to be
2215 looked up. The behaviour when one of the lookups defers can be
2216 controlled by a keyword.
2218 (4) It is now possible to specify the separator character for use when
2219 multiple records are returned.
2221 38. The dnslists ACL condition has been extended: it is now possible to supply
2222 a list of IP addresses and/or domains to be looked up in a particular DNS
2225 39. Added log_selector=+queue_time_overall.
2227 40. When running the queue in the test harness, wait just a tad after forking a
2228 delivery process, to get repeatability of debugging output.
2230 41. Include certificate and key file names in error message when GnuTLS fails
2231 to set them up, because the GnuTLS error message doesn't include the name
2232 of the failing file when there is a problem reading it.
2234 42. Allow both -bf and -bF in the same test run.
2236 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2238 44. The "Exiscan patch" is now merged into the mainline Exim source.
2240 45. Sometimes the final signoff response after QUIT could fail to get
2241 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2242 < 0 before doing a fflush(). This bug looks as though it goes back to the
2243 introduction of TLS in release 3.20, but "sometimes" must have been rare
2244 because the tests only now provoked it.
2246 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2247 (this can affect the format of dates).
2249 47. exim_tidydb, when checking for the continued existence of a message for
2250 which it has found a message-specific retry record, was not finding
2251 messages that were in split spool directories. Consequently, it was
2252 deleting retry records that should have stayed in existence.
2254 48. Steve fixed some bugs in eximstats.
2256 49. The SPA authentication driver was not abandoning authentication and moving
2257 on to the next authenticator when an expansion was forced to fail,
2258 contradicting the general specification for all authenticators. Instead it
2259 was generating a temporary error. It now behaves as specified.
2261 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2262 (the order specifies the preference for clients). The order is now AES256,
2263 AES128, 3DES, ARCFOUR128.
2265 51. Small patch to Sieve code - explicitly set From: when generating an
2268 52. Exim crashed if a remote delivery caused a very long error message to be
2269 recorded - for instance if somebody sent an entire SpamAssassin report back
2270 as a large number of 550 error lines. This bug was coincidentally fixed by
2271 increasing the size of one of Exim's internal buffers (big_buffer) that
2272 happened as part of the Exiscan merge. However, to be on the safe side, I
2273 have made the code more robust (and fixed the comments that describe what
2276 53. Now that there can be additional text after "Completed" in log lines (if
2277 the queue_time_overall log selector is set), a one-byte patch to exigrep
2278 was needed to allow it to recognize "Completed" as not the last thing in
2281 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2282 patch that reportedly fixes this has been added. I am not expert enough to
2283 create a test for it. This is what the patch creator wrote:
2285 "I found a little strange behaviour of ldap code when working with
2286 Windows 2003 AD Domain, where users was placed in more than one
2287 Organization Units. When I tried to give exim partial DN, the exit code
2288 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2289 But simultaneously result of request was absolutely normal ldap result,
2290 so I produce this patch..."
2292 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2293 REFERENCE, so I have modified the code to exclude the patch when that macro
2296 55. Some experimental protocols are using DNS PTR records for new purposes. The
2297 keys for these records are domain names, not reversed IP addresses. The
2298 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2299 leaves it alone. Component reversal etc. now happens only for IP addresses.
2302 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2304 57. Double the size of the debug message buffer (to 2048) so that more of very
2305 long debug lines gets shown.
2307 58. The exicyclog utility now does better if the number of log files to keep
2308 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2310 59. Two changes related to the smtp_active_hostname option:
2312 (1) $smtp_active_hostname is now available as a variable.
2313 (2) The default for smtp_banner uses $smtp_active_hostname instead
2314 of $primary_hostname.
2316 60. The host_aton() function is supposed to be passed a string that is known
2317 to be a valid IP address. However, in the case of IPv6 addresses, it was
2318 not checking this. This is a hostage to fortune. Exim now panics and dies
2319 if the condition is not met. A case was found where this could be provoked
2320 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2321 components; fortuitously, this particular loophole had already been fixed
2322 by change 4.50/55 above.
2324 If there are any other similar loopholes, the new check in host_aton()
2325 itself should stop them being exploited. The report I received stated that
2326 data on the command line could provoke the exploit when Exim was running as
2327 exim, but did not say which command line option was involved. All I could
2328 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2329 running as the user.
2332 61. There was a buffer overflow vulnerability in the SPA authentication code
2333 (which came originally from the Samba project). I have added a test to the
2334 spa_base64_to_bits() function which I hope fixes it.
2337 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2338 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2340 63. The daemon start-up calls getloadavg() while still root for those OS that
2341 need the first call to be done as root, but it missed one case: when
2342 deliver_queue_load_max is set with deliver_drop_privilege. This is
2343 necessary for the benefit of the queue runner, because there is no re-exec
2344 when deliver_drop_privilege is set.
2346 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2347 This has been fixed.
2349 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2350 was in use, was not putting the data itself into the right store pool;
2351 consequently, it could be overwritten for a subsequent message in the same
2352 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2355 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2357 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2358 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2359 the function were treating the return as a boolean value, which happened to
2360 work because 0=false and not-0=true, but is not correct code.
2362 68. The host_aton() function was not handling scoped IPv6 addresses (those
2363 with, for example, "%eth0" on the end) correctly.
2365 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2366 negated items (that is, ~something) in unsigned ints. Some compilers
2367 apparently mutter when there is no cast.
2369 70. If an address verification called from an ACL failed, and did not produce a
2370 user-specific message (i.e. there was only a "system" message), nothing was
2371 put in $acl_verify_message. In this situation, it now puts the system
2374 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2375 session; change 4.31/43 added the unwanted input to the log line - except
2376 that it did not do this in the start of session case. It now does.
2378 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2379 This is wrong and can cause the other end to generate a synchronization
2380 error if it is another Exim or anything else that does the synchronization
2381 check. A QUIT command is no longer sent after a timeout.
2383 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2384 during host lookups.
2386 74. The defer_ok option of callout verification was not working if it was used
2387 when verifying addresses in header lines, that is, for this case:
2389 verify = header_sender/callout=defer_ok
2391 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2392 those file descriptors could be used for SMTP connections. If anything
2393 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2394 could be sent to the SMTP client, causing chaos. The daemon now opens
2395 stdin, stdout, and stderr to /dev/null when it puts itself into the
2398 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2399 log by default. The user can override this with suitable Perl magic.
2401 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2402 the log message when discard triggers, was not working for the DATA ACL or
2403 for the non-SMTP ACL.
2405 78. Error message wording change in sieve.c.
2407 79. If smtp_accept_max_per_host was set, the number of connections could be
2408 restricted to fewer than expected, because the daemon was trying to set up
2409 a new connection before checking whether the processes handling previous
2410 connections had finished. The check for completed processes is now done
2411 earlier. On busy systems, this bug wouldn't be noticed because something
2412 else would have woken the daemon, and it would have reaped the completed
2415 80. If a message was submitted locally by a user whose login name contained one
2416 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2417 It caused a spool format error. I have fixed the spool reading code. A
2418 related problem was that the "from" clause in the Received: line became
2419 illegal because of the space(s). It is now covered by ${quote_local_part.
2421 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2424 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2426 83. Patch from Sieve maintainer to fix unterminated string problem in
2427 "vacation" handling.
2429 84. Some minor changes to the Linux configuration files to help with other
2430 OS variants using glibc.
2432 85. One more patch for Sieve to update vacation handling to latest spec.
2435 ----------------------------------------------------
2436 See the note above about the 4.44 and 4.50 releases.
2437 ----------------------------------------------------
2443 1. Change 4.43/35 introduced a bug that caused file counts to be
2444 incorrectly computed when quota_filecount was set in an appendfile
2447 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2448 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2451 3. Give more explanation in the error message when the command for a transport
2452 filter fails to execute.
2454 4. There are several places where Exim runs a non-Exim command in a
2455 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2456 was being done only for the command run by the queryprogram router. It is
2457 now done for all such subprocesses. The other cases are: ${run, transport
2458 filters, and the commands run by the lmtp and pipe transports.
2460 5. Some older OS have a limit of 256 on the maximum number of file
2461 descriptors. Exim was using setrlimit() to set 1000 as a large value
2462 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2463 systems. I've change it so that if it can't get 1000, it tries for 256.
2465 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2466 was an oversight, and furthermore, ever since the addition of extra
2467 controls (e.g. 4.43/32), the checks on when to allow different forms of
2468 "control" were broken. There should now be diagnostics for all cases when a
2469 control that does not make sense is encountered.
2471 7. $recipients is now available in the predata ACL (oversight).
2473 8. Tidy the search cache before the fork to do a delivery from a message
2474 received from the command line. Otherwise the child will trigger a lookup
2475 failure and thereby defer the delivery if it tries to use (for example) a
2476 cached ldap connection that the parent has called unbind on.
2478 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2479 of $address_data from the recipient verification was clobbered by the
2480 sender verification.
2482 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2483 was its contents. (It was OK if the option was not defined at all.)
2485 11. A "Completed" log line is now written for messages that are removed from
2486 the spool by the -Mrm option.
2488 12. $host_address is now set to the target address during the checking of
2489 ignore_target_hosts.
2491 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2492 being passed; this would have caused $sender_host_name to have been used if
2493 matching the list had actually called for a host name (not very likely,
2494 since this list is usually IP addresses). A host name is now passed as
2497 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2498 code that specifies a non-restarting handler (typically sigaction() in
2499 modern systems) in an attempt to fix a rare and obscure crash bug.
2501 15. Narrowed the window for a race in the daemon that could cause it to ignore
2502 SIGCHLD signals. This is not a major problem, because they are used only to
2503 wake it up if nothing else does.
2505 16. A malformed maildirsize file could cause Exim to calculate negative values
2506 for the mailbox size or file count. Odd effects could occur as a result.
2507 The maildirsize information is now recalculated if the size or filecount
2510 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2511 support for a long time. Removed HAVE_SYS_VFS_H.
2513 18. Updated exipick to current release from John Jetmore.
2515 19. Allow an empty sender to be matched against a lookup in an address list.
2516 Previously the only cases considered were a regular expression, or an
2519 20. Exim went into a mad DNS lookup loop when doing a callout where the
2520 host was specified on the transport, if the DNS lookup yielded more than
2523 21. The RFC2047 encoding function was originally intended for short strings
2524 such as real names; it was not keeping to the 75-character limit for
2525 encoded words that the RFC imposes. It now respects the limit, and
2526 generates multiple encoded words if necessary. To be on the safe side, I
2527 have increased the buffer size for the ${rfc2047: expansion operator from
2530 22. Failure to deliver a bounce message always caused it to be frozen, even if
2531 there was an errors_to setting on the router. The errors_to setting is now
2534 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2535 canonical form (fully expanded) before being placed in
2536 $sender_host_address.
2538 24. Updated eximstats to version 1.33
2540 25. Include certificate and key file names in error message when GnuTLS fails
2541 to set them up, because the GnuTLS error message doesn't include the name
2542 of the failing file when there is a problem reading it.
2544 26. Expand error message when OpenSSL has problems setting up cert/key files.
2547 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2548 (this can affect the format of dates).
2550 28. exim_tidydb, when checking for the continued existence of a message for
2551 which it has found a message-specific retry record, was not finding
2552 messages that were in split spool directories. Consequently, it was
2553 deleting retry records that should have stayed in existence.
2555 29. eximstats updated to version 1.35
2556 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2557 1.35 - bugfix such that pie charts by volume are generated correctly
2559 30. The SPA authentication driver was not abandoning authentication and moving
2560 on to the next authenticator when an expansion was forced to fail,
2561 contradicting the general specification for all authenticators. Instead it
2562 was generating a temporary error. It now behaves as specified.
2564 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2565 (the order specifies the preference for clients). The order is now AES256,
2566 AES128, 3DES, ARCFOUR128.
2568 31. Small patch to Sieve code - explicitly set From: when generating an
2571 32. Exim crashed if a remote delivery caused a very long error message to be
2572 recorded - for instance if somebody sent an entire SpamAssassin report back
2573 as a large number of 550 error lines. This bug was coincidentally fixed by
2574 increasing the size of one of Exim's internal buffers (big_buffer) that
2575 happened as part of the Exiscan merge. However, to be on the safe side, I
2576 have made the code more robust (and fixed the comments that describe what
2579 33. Some experimental protocols are using DNS PTR records for new purposes. The
2580 keys for these records are domain names, not reversed IP addresses. The
2581 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2582 leaves it alone. Component reversal etc. now happens only for IP addresses.
2585 34. The host_aton() function is supposed to be passed a string that is known
2586 to be a valid IP address. However, in the case of IPv6 addresses, it was
2587 not checking this. This is a hostage to fortune. Exim now panics and dies
2588 if the condition is not met. A case was found where this could be provoked
2589 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2590 components; fortuitously, this particular loophole had already been fixed
2591 by change 4.50/55 or 4.44/33 above.
2593 If there are any other similar loopholes, the new check in host_aton()
2594 itself should stop them being exploited. The report I received stated that
2595 data on the command line could provoke the exploit when Exim was running as
2596 exim, but did not say which command line option was involved. All I could
2597 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2598 running as the user.
2601 35. There was a buffer overflow vulnerability in the SPA authentication code
2602 (which came originally from the Samba project). I have added a test to the
2603 spa_base64_to_bits() function which I hope fixes it.
2606 36. The daemon start-up calls getloadavg() while still root for those OS that
2607 need the first call to be done as root, but it missed one case: when
2608 deliver_queue_load_max is set with deliver_drop_privilege. This is
2609 necessary for the benefit of the queue runner, because there is no re-exec
2610 when deliver_drop_privilege is set.
2612 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2613 was in use, was not putting the data itself into the right store pool;
2614 consequently, it could be overwritten for a subsequent message in the same
2615 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2618 38. Sometimes the final signoff response after QUIT could fail to get
2619 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2620 < 0 before doing a fflush(). This bug looks as though it goes back to the
2621 introduction of TLS in release 3.20, but "sometimes" must have been rare
2622 because the tests only now provoked it.
2628 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2629 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2630 Now it returns an integer. A number of calls were still expecting a T/F
2631 return. Fortuitously, in all cases, the tests worked in OK situations,
2632 which is the norm. However, things would have gone wrong on any write
2633 failures on the smtp file descriptor. This function is used when sending
2634 messages over SMTP and also when doing verify callouts.
2636 2. When Exim is called to do synchronous delivery of a locally submitted
2637 message (the -odf or -odi options), it no longer closes stderr before doing
2640 3. Implemented the mua_wrapper option.
2642 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2644 5. Implemented the functions header_remove(), header_testname(),
2645 header_add_at_position(), and receive_remove_recipient(), and exported them
2648 6. If an ACL "warn" statement specified the addition of headers, Exim already
2649 inserted X-ACL-Warn: at the start if there was no header name. However, it
2650 was not making this test for the second and subsequent header lines if
2651 there were newlines in the string. This meant that an invalid header could
2652 be inserted if Exim was badly configured.
2654 7. Allow an ACL "warn" statement to add header lines at the start or after all
2655 the Received: headers, as well as at the end.
2657 8. Added the rcpt_4xx retry error code.
2659 9. Added postmaster_mailfrom=xxx to callout verification option.
2661 10. Added mailfrom=xxxx to the callout verification option, for verify=
2664 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2665 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2667 12. Inserted some casts to stop certain compilers warning when using pointer
2668 differences as field lengths or precisions in printf-type calls (mostly
2669 affecting debugging statements).
2671 13. Added optional readline() support for -be (dynamically loaded).
2673 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2674 same clock tick as a message's arrival, so that its received time was the
2675 same as the "first fail" time on the retry record, and that message
2676 remained on the queue past the ultimate address timeout, every queue runner
2677 would try a delivery (because it was past the ultimate address timeout) but
2678 after another failure, the ultimate address timeout, which should have then
2679 bounced the address, did not kick in. This was a "< instead of <=" error;
2680 in most cases the first failure would have been in the next clock tick
2681 after the received time, and all would be well.
2683 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2684 being recognized when the domain list was tested by the match_domain
2685 condition in an expansion string.
2687 16. Added the ${str2b64: operator.
2689 17. Exim was always calling setrlimit() to set a large limit for the number of
2690 processes, without checking whether the existing limit was already
2691 adequate. (It did check for the limit on file descriptors.) Furthermore,
2692 errors from getrlimit() and setrlimit() were being ignored. Now they are
2693 logged to the main and panic logs, but Exim does carry on, to try to do its
2694 job under whatever limits there are.
2696 18. Imported PCRE 5.0.
2698 19. Trivial typo in log message " temporarily refused connection" (the leading
2701 20. If the log selector return_path_on_delivery was set and an address was
2702 redirected to /dev/null, the delivery process crashed because it assumed
2703 that a return path would always be set for a "successful" delivery. In this
2704 case, the whole delivery is bypassed as an optimization, and therefore no
2707 21. Internal re-arrangement: the function for sending a challenge and reading
2708 a response while authentication was assuming a zero-terminated challenge
2709 string. It's now changed to take a pointer and a length, to allow for
2710 binary data in such strings.
2712 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2714 23. Exim was not respecting finduser_retries when seeking the login of the
2715 uid under which it was called; it was always trying 10 times. (The default
2716 setting of finduser_retries is zero.) Also, it was sleeping after the final
2717 failure, which is pointless.
2719 24. Implemented tls_on_connect_ports.
2721 25. Implemented acl_smtp_predata.
2723 26. If the domain in control=submission is set empty, Exim assumes that the
2724 authenticated id is a complete email address when it generates From: or
2725 Sender: header lines.
2727 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2728 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2729 chown and chgrp in /bin and hostname in /usr/bin.
2731 28. Exim was keeping the "process log" file open after each use, just as it
2732 does for the main log. This opens the possibility of it remaining open for
2733 long periods when the USR1 signal hits a daemon. Occasional processlog
2734 errors were reported, that could have been caused by this. Anyway, it seems
2735 much more sensible not to leave this file open at all, so that is what now
2738 29. The long-running daemon process does not normally write to the log once it
2739 has entered its main loop, and it closes the log before doing so. This is
2740 so that log files can straightforwardly be renamed and moved. However,
2741 there are a couple of unusual error situations where the daemon does write
2742 log entries, and I had neglected to close the log afterwards.
2744 30. The text of an SMTP error response that was received during a remote
2745 delivery was being truncated at 512 bytes. This is too short for some of
2746 the long messages that one sometimes sees. I've increased the limit to
2749 31. It is now possible to make retry rules that apply only when a message has a
2750 specific sender, in particular, an empty sender.
2752 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2753 it possible to be selective about when SMTP synchronization is enforced.
2755 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2757 32. Implemented hosts_connection_nolog.
2759 33. Added an ACL for QUIT.
2761 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2764 35. Added mailbox_size and mailbox_filecount to appendfile.
2766 36. Added control = no_multiline_responses to ACLs.
2768 37. There was a bug in the logic of the code that waits for the clock to tick
2769 in the case where the clock went backwards by a substantial amount such
2770 that the microsecond fraction of "now" was more than the microsecond
2771 fraction of "then" (but the whole seconds number was less).
2773 38. Added support for the libradius Radius client library this is found on
2774 FreeBSD (previously only the radiusclient library was supported).
2780 1. When certain lookups returned multiple values in the form name=value, the
2781 quoting of the values was not always being done properly. Specifically:
2782 (a) If the value started with a double quote, but contained no whitespace,
2784 (b) If the value contained whitespace other than a space character (i.e.
2785 tabs or newlines or carriage returns) it was not quoted.
2786 This fix has been applied to the mysql and pgsql lookups by writing a
2787 separate quoting function and calling it from the lookup code. The fix
2788 should probably also be applied to nisplus, ibase and oracle lookups, but
2789 since I cannot test any of those, I have not disturbed their existing code.
2791 2. A hit in the callout cache for a specific address caused a log line with no
2792 reason for rejecting RCPT. Now it says "Previous (cached) callout
2793 verification failure".
2795 3. There was an off-by-one bug in the queryprogram router. An over-long
2796 return line was truncated at 256 instead of 255 characters, thereby
2797 overflowing its buffer with the terminating zero. As well as fixing this, I
2798 have increased the buffer size to 1024 (and made a note to document this).
2800 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2801 when Exim is waiting for an SMTP response from a remote server, Exim
2802 restarts its select() call on the socket, thereby resetting its timeout.
2803 This is not a problem when such interrupts are rare. Somebody set up a cron
2804 job to run exiwhat every 2 minutes, which is less than the normal select()
2805 timeout (5 or 10 minutes). This meant that the select() timeout never
2806 kicked in because it was always reset. I have fixed this by comparing the
2807 time when an interrupt arrives with the time at the start of the first call
2808 to select(). If more time than the timeout has elapsed, the interrupt is
2809 treated as a timeout.
2811 5. Some internal re-factoring in preparation for the addition of Sieve
2812 extensions (by MH). In particular, the "personal" test is moved to a
2813 separate function, and given an option for scanning Cc: and Bcc: (which is
2814 not set for Exim filters).
2816 6. When Exim created an email address using the login of the caller as the
2817 local part (e.g. when creating a From: or Sender: header line), it was not
2818 quoting the local part when it contained special characters such as @.
2820 7. Installed new OpenBSD configuration files.
2822 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2823 try to make them clearer.
2825 9. Callout options, other than the timeout value, were being ignored when
2826 verifying sender addresses in header lines. For example, when using
2828 verify = header_sender/callout=no_cache
2830 the cache was (incorrectly) being used.
2832 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2833 only the Cygwin environment.
2835 11. When return_path_on_delivery was set as a log selector, if different remote
2836 addresses in the same message used different return paths and parallel
2837 remote delivery occurred, the wrong values would sometimes be logged.
2838 (Whenever a remote delivery process finished, the return path value from
2839 the most recently started remote delivery process was logged.)
2841 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2842 lines when AUTH and/or TLS are in use. This is the "received protocol"
2843 field. Exim used to use "asmtp" for authenticated SMTP, without any
2844 indication (in the protocol name) for TLS use. Now it follows the RFC and
2845 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2846 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2847 names appear in log lines as well as in Received: header lines.
2849 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2850 extensions, and comparison tests, and to fix some bugs.
2852 14. Changes to the "personal" filter test:
2854 (1) The test was buggy in that it was just doing the equivalent of
2855 "contains" tests on header lines. For example, if a user's address was
2856 anne@some.where, the "personal" test would incorrectly be true for
2858 To: susanne@some.where
2860 This test is now done by extracting each address from the header in turn,
2861 and checking the entire address. Other tests that are part of "personal"
2862 are now done using regular expressions (for example, to check local parts
2863 of addresses in From: header lines).
2865 (2) The list of non-personal local parts in From: addresses has been
2866 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2867 taken from the Sieve specification recommendations.
2869 (3) If the message contains any header line starting with "List-" it is
2870 treated as non-personal.
2872 (4) The test for "circular" in the Subject: header line has been removed
2873 because it now seems ill-conceived.
2875 15. Minor typos in src/EDITME comments corrected.
2877 16. Installed latest exipick from John Jetmore.
2879 17. If headers_add on a router specified a text string that was too long for
2880 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2881 of string_sprintf() is now avoided.
2883 18. $message_body_size was not set (it was always zero) when running the DATA
2884 ACL and the local_scan() function.
2886 19. For the "mail" command in an Exim filter, no default was being set for
2887 the once_repeat time, causing a random time value to be used if "once" was
2888 specified. (If the value happened to be <= 0, no repeat happened.) The
2889 default is now 0s, meaning "never repeat". The "vacation" command was OK
2890 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2891 (I found it when inspecting the code).
2893 20. There is now an overall timeout for performing a callout verification. It
2894 defaults to 4 times the callout timeout, which applies to individual SMTP
2895 commands during the callout. The overall timeout applies when there is more
2896 than one host that can be tried. The timeout is checked before trying the
2897 next host. This prevents very long delays if there are a large number of
2898 hosts and all are timing out (e.g. when the network connections are timing
2899 out). The value of the overall timeout can be changed by specifying an
2900 additional sub-option for "callout", called "maxwait". For example:
2902 verify = sender/callout=5s,maxwait=20s
2904 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2905 to the end before writing, but this should make it even safer).
2907 22. Exim was forgetting that it had advertised PIPELINING for the second and
2908 subsequent messages on an SMTP connection. It was also not resetting its
2909 memory on STARTTLS and an internal HELO.
2911 23. When Exim logs an SMTP synchronization error within a session, it now
2912 records whether PIPELINING has been advertised or not.
2914 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2915 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2916 rather than long int.
2918 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2920 26. Added the never_mail option to autoreply.
2926 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2927 crash if the getsockname() call failed; this can happen if a connection is
2928 closed very soon after it is established. The problem was simply in the
2929 order in which certain operations were done, causing Exim to try to write
2930 to the SMTP stream before it had set up the file descriptor. The bug has
2931 been fixed by making things happen in the correct order.
2937 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2938 before the connection was closed, thus losing the rejection response.
2940 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2941 some early Solaris releases, but causes trouble in current releases where
2942 socklen_t is defined.
2944 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2947 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2950 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2951 that releasing the top of it at the end releases what was used for sub-
2952 expansions (unless the block got too big). However, discard this block if
2953 the first thing is a variable or header, so that we can use its block when
2954 it is dynamic (useful for very large $message_headers, for example).
2956 6. Lookups now cache *every* query, not just the most recent. A new, separate
2957 store pool is used for this. It can be recovered when all lookup caches are
2958 flushed. Lookups now release memory at the end of their result strings.
2959 This has involved some general refactoring of the lookup sources.
2961 7. Some code has been added to the store_xxx() functions to reduce the amount
2962 of flapping under certain conditions.
2964 8. log_incoming_interface used to affect only the <= reception log lines. Now
2965 it causes the local interface and port to be added to several more SMTP log
2966 lines, for example "SMTP connection from", and rejection lines.
2968 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2970 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2972 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2973 could be overwritten at the end of the current message (or the start of a
2974 new message if it was set in a HELO ACL). The value is now preserved for
2975 the duration of the SMTP connection.
2977 12. If a transport had a headers_rewrite setting, and a matching header line
2978 contained an unqualified address, that address was qualified, even if it
2979 did not match any rewriting rules. The underlying bug was that the values
2980 of the flags that permit the existence of unqualified sender and recipient
2981 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2982 non-local messages, and by -bnq for local messages) were not being
2983 preserved with the message after it was received.
2985 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2986 "next input=" as part of the text comprising the host identity instead of
2987 the correct text. The code was using the same buffer for two different
2988 strings. However, depending on which order the printing function evaluated
2989 its arguments, the bug did not always show up. Under Linux, for example, my
2990 test suite worked just fine.
2992 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2993 doesn't work with some older versions of Perl. It has been changed to "my",
2994 which in any case is probably the better facility to use.
2996 15. A really picky compiler found some instances of statements for creating
2997 error messages that either had too many or two few arguments for the format
3000 16. The size of the buffer for calls to the DNS resolver has been increased
3001 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3002 for addresses that have a lot of PTR records. This alleviates a problem; it
3003 does not fully solve it.
3005 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3006 buffer now truncates the list and logs the incident, which is the same
3007 action as happens when Exim is looking up a host name and its aliases.
3008 Previously in this situation something unpredictable would happen;
3009 sometimes it was "internal error: store_reset failed".
3011 18. If a server dropped the connection unexpectedly when an Exim client was
3012 using GnuTLS and trying to read a response, the client delivery process
3013 crashed while trying to generate an error log message.
3015 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3016 string, for example:
3018 warn message = H1: something\nH2: something
3020 the text was added as a single header line from Exim's point of view
3021 though it ended up OK in the delivered message. However, searching for the
3022 second and subsequent header lines using $h_h2: did not work. This has been
3023 fixed. Similarly, if a system filter added multiple headers in this way,
3024 the routers could not see them.
3026 20. Expanded the error message when iplsearch is called with an invalid key to
3027 suggest using net-iplsearch in a host list.
3029 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3030 ACLs are no longer actually imposed (and a message to that effect is
3033 22. If a "gecos" field in a passwd entry contained escaped characters, in
3034 particular, if it contained a \" sequence, Exim got it wrong when building
3035 a From: or a Sender: header from that name. A second bug also caused
3036 incorrect handling when an unquoted " was present following a character
3037 that needed quoting.
3039 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3040 was not being matched caselessly.
3042 24. Arranged for all hyphens in the exim.8 source to be escaped with
3045 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3046 properly. Recipient callout cache records were still being keyed to include
3047 the sender, even when use_sender was set false. This led to far more
3048 callouts that were necessary. The sender is no longer included in the key
3049 when use_sender is false.
3051 26. Added "control = submission" modifier to ACLs.
3053 27. Added the ${base62d: operator to decode base 62 numbers.
3055 28. dnsdb lookups can now access SRV records.
3057 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3058 the configuration file.
3060 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3061 (-v) mode. This makes the output for a verbose queue run more intelligible.
3063 31. Added a use_postmaster feature to recipient callouts.
3065 32. Added the $body_zerocount variable, containing the number of binary zero
3066 bytes in the message body.
3068 33. The time of last modification of the "new" subdirectory is now used as the
3069 "mailbox time last read" when there is a quota error for a maildir
3072 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3074 35. Added +ignore_unknown as a special item in host lists.
3076 36. Code for decoding IPv6 addresses in host lists is now included, even if
3077 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3078 address was recognized as an IP address, but was then not correctly decoded
3079 into binary, causing unexpected and incorrect effects when compared with
3086 1. Very minor rewording of debugging text in manualroute to say "list of
3087 hosts" instead of "hostlist".
3089 2. If verify=header_syntax was set, and a header line with an unqualified
3090 address (no domain) and a large number of spaces between the end of the
3091 name and the colon was received, the reception process suffered a buffer
3092 overflow, and (when I tested it) crashed. This was caused by some obsolete
3093 code that should have been removed. The fix is to remove it!
3095 3. When running in the test harness, delay a bit after writing a bounce
3096 message to get a bit more predictability in the log output.
3098 4. Added a call to search_tidyup() just before forking a reception process. In
3099 theory, someone could use a lookup in the expansion of smtp_accept_max_
3100 per_host which, without the tidyup, could leave open a database connection.
3102 5. Added the variables $recipient_data and $sender_data which get set from a
3103 lookup success in an ACL "recipients" or "senders" condition, or a router
3104 "senders" option, similar to $domain_data and $local_part_data.
3106 6. Moved the writing of debug_print from before to after the "senders" test
3109 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3110 problems for message scanning, either using a data ACL, or using
3111 local_scan() because the Received: header was not generated till after they
3112 were called (in order to set the time as the time of reception completion).
3113 I have revised the way this works. The header is now generated after the
3114 body is received, but before the ACL or local_scan() are called. After they
3115 are run, the timestamp in the header is updated.
3121 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3122 before starting a queue runner without re-exec. This happened only when
3123 deliver_drop_privilege was set or when the Exim user was set to root. The
3124 effect of the bug was that timeouts during subsequent deliveries caused
3125 crashes instead of being properly handled. The handler is now left at its
3126 default (and expected) setting.
3128 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3129 message, again when deliver_drop_privilege is set or Exim is run as root.
3130 The bug described in (1) was not present in this case, but the tidying up
3131 of the other signals was missing. I have made the two cases consistent.
3133 3. The ignore_target_hosts setting on a manualroute router was being ignored
3134 for hosts that were looked up using the /MX notation.
3136 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3139 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3140 operated on the sender address. After changing the $sender_address to <>
3141 for the sender address verify, Exim was re-instated it as the original
3142 (before rewriting) address, but remembering that it had rewritten it, so it
3143 wasn't rewriting it again. This bug also had the effect of breaking the
3144 sender address verification caching when the sender address was rewritten.
3146 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3147 This has been changed so that if the ip literal address matches
3148 ignore_target_hosts, the router declines.
3150 7. Added expansion conditions match_domain, match_address, and match_local_
3151 part (NOT match_host).
3153 8. The placeholder for the Received: header didn't have a length field set.
3155 9. Added code to Exim itself and to exim_lock to test for a specific race
3156 condition that could lead to file corruption when using MBX delivery. The
3157 issue is with the lockfile that is created in /tmp. If this file is removed
3158 after a process has opened it but before that process has acquired a lock,
3159 there is the potential for a second process to recreate the file and also
3160 acquire a lock. This could lead to two Exim processes writing to the file
3161 at the same time. The added code performs the same test as UW imapd; it
3162 checks after acquiring the lock that its file descriptor still refers to
3163 the same named file.
3165 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3166 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3169 11. Added the smtp_active_hostname option. If used, this will typically be made
3170 to depend on the incoming interface address. Because $interface_address is
3171 not set up until the daemon has forked a reception process, error responses
3172 that can happen earlier (such as "too many connections") no longer contain
3175 12. If an expansion in a condition on a "warn" statement fails because a lookup
3176 defers, the "warn" statement is abandoned, and the next ACL statement is
3177 processed. Previously this caused the whole ACL to be aborted.
3179 13. Added the iplsearch lookup type.
3181 14. Added ident_timeout as a log selector.
3183 15. Added tls_certificate_verified as a log selector.
3185 16. Added a global option tls_require_ciphers (compare the smtp transport
3186 option of the same name). This controls incoming TLS connections.
3188 17. I finally figured out how to make tls_require_ciphers do a similar thing
3189 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3190 before starting the TLS session.
3192 18. Tabs are now shown as \t in -bP output.
3194 19. If the log selector return_path_on_delivery was set, Exim crashed when
3195 bouncing a message because it had too many Received: header lines.
3197 20. If two routers both had headers_remove settings, and the first one included
3198 a superfluous trailing colon, the final name in the first list and the
3199 first name in the second list were incorrectly joined into one item (with a
3200 colon in the middle).
3206 1. Added -C and -D options to the exinext utility, mainly to make it easier
3207 to include in the automated testing, but these could be helpful when
3208 multiple configurations are in use.
3210 2. The exinext utility was not formatting the output nicely when there was
3211 an alternate port involved in the retry record key, nor when there was a
3212 message id as well (for retries that were specific to a specific message
3213 and a specific host). It was also confused by IPv6 addresses, because of
3214 the additional colons they contain. I have fixed the IPv4 problem, and
3215 patched it up to do a reasonable job for IPv6.
3217 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3218 delivery, the log line now contains "pipelined" if PIPELINING was used.
3220 4. An SMTP transport process used to panic and die if the bind() call to set
3221 an explicit outgoing interface failed. This has been changed; it is now
3222 treated in the same way as a connect() failure.
3224 5. A reference to $sender_host_name in the part of a conditional expansion
3225 that was being skipped was still causing a DNS lookup. This no longer
3228 6. The def: expansion condition was not recognizing references to header lines
3229 that used bh_ and bheader_.
3231 7. Added the _cache feature to named lists.
3233 8. The code for checking quota_filecount in the appendfile transport was
3234 allowing one more file than it should have been.
3236 9. For compatibility with Sendmail, the command line option
3244 and sets the incoming protocol and host name (for trusted callers). The
3245 host name and its colon can be omitted when only the protocol is to be set.
3246 Note the Exim already has two private options, -pd and -ps, that refer to
3247 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3248 "s", but I don't think that's a major issue.
3250 10. A number of refactoring changes to the code, none of which should affect
3253 (a) The number of logging options was getting close to filling up the
3254 32-bit word that was used as a bit map. I have split them into two classes:
3255 those that are passed in the argument to log_write(), and those that are
3256 only ever tested independently outside of that function. These are now in
3257 separate 32-bit words, so there is plenty of room for expansion again.
3258 There is no change in the user interface or the logging behaviour.
3260 (b) When building, for example, log lines, the code previously used a
3261 macro that called string_cat() twice, in order to add two strings. This is
3262 not really sufficiently general. Furthermore, there was one instance where
3263 it was actually wrong because one of the argument was used twice, and in
3264 one call a function was used. (As it happened, calling the function twice
3265 did not affect the overall behaviour.) The macro has been replaced by a
3266 function that can join an arbitrary number of extra strings onto a growing
3269 (c) The code for expansion conditions now uses a table and a binary chop
3270 instead of a serial search (which was left over from when there were very
3271 few conditions). Also, it now recognizes conditions like "pam" even when
3272 the relevant support is not compiled in: a suitably worded error message is
3273 given if an attempt is made to use such a condition.
3275 11. Added ${time_interval:xxxxx}.
3277 12. A bug was causing one of the ddress fields not to be passed back correctly
3278 from remote delivery subprocesses. The field in question was not being
3279 subsequently used, so this caused to problems in practice.
3281 13. Added new log selectors queue_time and deliver_time.
3283 14. Might have fixed a bug in maildirsizefile handling that threw up
3284 "unexpected character" debug warnings, and recalculated the data
3285 unnecessarily. In any case, I expanded the warning message to give more
3288 15. Added the message "Restricted characters in address" to the statements in
3289 the default ACL that block characters like @ and % in local parts.
3291 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3292 Three changes have been made:
3294 (a) There was a serious bug; a negative response to MAIL caused the whole
3295 recipient domain to be cached as invalid, thereby blocking all messages
3296 to all local parts at the same domain, from all senders. This bug has
3297 been fixed. The domain is no longer cached after a negative response to
3298 MAIL if the sender used is not empty.
3300 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3303 (c) A new callout option, "use_sender" has been added for people who want
3304 the modified behaviour.
3310 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3313 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3314 indeed breaks things for older releases.
3316 3. Added additional logging to the case where there is a problem reading data
3317 from a filter that is running in a subprocess using a pipe, in order to
3318 try to track down a specific problem.
3320 4. Testing facility fudge: when running in the test harness and attempting
3321 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3322 getting "No route to host". Convert this to a timeout.
3324 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3327 6. Some OS don't have socklen_t but use size_t instead. This affects the
3328 fifth argument of getsockopt() amongst other things. This is now
3329 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3330 can be set for individual OS. I have set it for SunOS5, OSF1, and
3331 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3332 some earlier ones do not.
3334 7. Change 4.30/15 was not doing the test caselessly.
3336 8. The standard form for an IPv6 address literal was being rejected by address
3337 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3338 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3339 this, as well as the form without the "IPv6" on the front (but only when
3340 address literals are enabled, of course).
3342 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3344 10. Exim crashed if a message with an empty sender address specified by -f
3345 encountered a router with an errors_to setting. This could be provoked only
3346 by a command such as
3350 where an empty string was supplied; "<>" did not hit this bug.
3352 11. Installed PCRE release 4.5.
3354 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3355 remained set. It is now erased.
3357 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3358 times from message ids (which are base 36 rather than the normal 62).
3360 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3361 were being counted as actual protocol errors, and logged if the log
3362 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3363 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3364 and DATA following a set of rejected RCPTs do not count as protocol errors.
3365 In other words, Exim assumes they were pipelined, though this may not
3366 actually be the case. Of course, in all cases the client gets an
3367 appropriate error code.
3369 15. If a lookup fails in an ACL condition, a message about the failure may
3370 be available; it is used if testing the ACL cannot continue, because most
3371 such messages specify what the cause of the deferral is. However, some
3372 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3373 that caused an old message to be retained and used if a later statement
3374 caused a defer, replacing the real cause of the deferral.
3376 16. If an IP address had so many PTR records that the DNS lookup buffer
3377 was not large enough to hold them, Exim could crash while trying to process
3378 the truncated data. It now detects and logs this case.
3380 17. Further to 4.21/58, another change has been made: if (and only if) the
3381 first line of a message (the first header line) ends with CRLF, a bare LF
3382 in a subsequent header line has a space inserted after it, so as not to
3383 terminate the header.
3385 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3386 unnecessarily, used atoi() instead of strtol(), and didn't check the
3387 termination when getting file sizes from file names by regex.
3389 19. Completely re-implemented the support for maildirsize files, in the light
3390 of a number of problems with the previous contributed implementation
3391 (4.30/29). In particular:
3393 . If the quota is zero, the maildirsize file is maintained, but no quota is
3396 . If the maildir directory does not exist, it is created before any attempt
3397 to write a maildirsize file.
3399 . The quota value in the file is just a cache; if the quota is changed in
3400 the transport, the new value overrides.
3402 . A regular expression is available for excluding directories from the
3405 20. The autoreply transport checks the characters in options that define the
3406 message's headers; it allows continued headers, but it was checking with
3407 isspace() after an embedded newline instead of explicitly looking for a
3410 21. If all the "regular" hosts to which an address was routed had passed their
3411 expiry times, and had not reached their retry times, the address was
3412 bounced, even if fallback hosts were defined. Now Exim should go on to try
3415 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3416 equivalent code in the SMTP transport. Some hosts send humungous responses
3417 to HELO/EHLO, more than 1024 it seems.
3419 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3420 gives compiler warnings in some environments. I've now done it "properly",
3423 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3424 (because of problems with the built-in one) was declared to return uschar *
3425 instead of char *, causing compiler failure.
3427 25. Fixed a file descriptor leak when processing alias/forward files.
3429 26. Fixed a minor format string issue in dbfn.c.
3431 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3433 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3434 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3437 29. When a sender address is verified, it is cached, to save repeating the test
3438 when there is more than one recipient in a message. However, when the
3439 verification involves a callout, it is possible for different callout
3440 options to be set for different recipients. It is too complicated to keep
3441 track of this in the cache, so now Exim always runs a verification when a
3442 callout is required, relying on the callout cache for the optimization.
3443 The overhead is duplication of the address routing, but this should not be
3446 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3447 to be verified with callout=postmaster, and the main callout worked but the
3448 postmaster check failed, the verification correctly failed. However, if a
3449 subsequent RCPT command asked for sender verification *without* the
3450 postmaster check, incorrect caching caused this verification also to fail,
3453 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3454 it was not caching the DNS options (qualify_single, search_parents) that
3455 were used when the lookup failed. A subsequent lookup with different
3456 options therefore always gave the same answer, though there were cases
3457 where it should not have. (Example: a "domains = !$mx_any" option on a
3458 dnslookup router: the "domains" option is always processed without any
3459 widening, but the router might have qualify_single set.) Now Exim uses the
3460 cached value only when the same options are set.
3462 32. Added John Jetmore's "exipick" utility to the distribution.
3464 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3465 than a timeout (e.g. a certificate is required, and is not provided), an
3466 Exim server now closes the connection immediately. Previously it waited for
3467 the client to close - but if the client is SSL, it seems that they each
3468 wait for each other, leading to a delay before one of them times out.
3470 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3471 maintained 0.8.x compatibility because I don't think many are using it, and
3472 it is clearly obsolete.
3474 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3477 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3478 client certificate was expired. A simple patch fixes this, though I don't
3479 understand the full logic of why the verify callback is called multiple
3482 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3483 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3484 which causes problems with some clients (such as the Certicom SSL Plus
3485 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3486 disables the coutermeasure allowing Eudora to connect."
3488 38. Exim was not checking that a write() to a log file succeeded. This could
3489 lead to Bad Things if a log got too big, in particular if it hit a file
3490 size limit. Exim now panics and dies if it cannot write to a log file, just
3491 as it does if it cannot open a log file.
3493 39. Modified OS/Makefile-Linux so that it now contains
3495 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3497 The two -D definitions ensure that Exim is compiled with large file
3498 support, which makes it possible to handle log files that are bigger than
3501 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3502 instance) a domain was checked against a named list that involved a lookup,
3503 causing $domain_data to be set, then another domain was checked against the
3504 same list, then the first domain was re-checked, the value of $domain_data
3505 after the final check could be wrong. In particular, if the second check
3506 failed, it could be set empty. This bug probably also applied to
3509 41. The strip_trailing_dot option was not being applied to the address given
3510 with the -f command-line option.
3512 42. The code for reading a message's header from the spool was incrementing
3513 $received_count, but never initializing it. This meant that the value was
3514 incorrect (doubled) while delivering a message in the same process in which
3515 it was received. In the most common configuration of Exim, this never
3516 happens - a fresh exec is done - but it can happen when
3517 deliver_drop_privilege is set.
3519 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3520 it now includes up to 150 characters of the unexpected data in the log
3523 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3524 and building data strings. The size of both of these buffers was 10 000
3525 bytes - far larger than anybody would *ever* want, thought I. Needless to
3526 say, somebody hit the limit. I have increased the maximum line length to
3527 20 000 and the maximum data length of concatenated lines to 100 000. I have
3528 also fixed two bugs, because there was no checking on these buffers. Tsk,
3529 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3530 buffer is too small.
3532 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3533 lsearch lookups. Now it does.
3535 46. When parsing a route_list item in a manualroute router, a fixed-length
3536 buffer was used for the list of hosts. I made this 1024 bytes long,
3537 thinking that nobody would ever have a list of hosts that long. Wrong.
3538 Somebody had a whole pile of complicated expansion conditions, and the
3539 string was silently truncated, leading to an expansion error. It turns out
3540 that it is easier to change to an unlimited length (owing to other changes
3541 that have happened since this code was originally written) than to build
3542 structure for giving a limitation error. The length of the item that
3543 expands into the list of hosts is now unlimited.
3545 47. The lsearch lookup could not handle data where the length of text line was
3546 more than 4095 characters. Such lines were truncated, leading to shortened
3547 data being returned. It should now handle lines of any length.
3549 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3550 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3553 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3554 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3555 no longer happens in comment lines. A list of replacements is now placed
3556 at the head of all of the source files, except those whose only change is
3557 to replace PERL_COMMAND in the very first #! line.
3559 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3560 messages on the queue, with a bottom-up merge sort, using code contributed
3561 by Michael Haardt. This should make operations like -bp somewhat faster on
3562 large queues. It won't affect queue runners, except when queue_run_in_order
3565 51. Installed eximstats 1.31 in the distribution.
3567 52. Added support for SRV lookups to the dnslookup router.
3569 53. If an ACL referred to $message_body or $message_body_end, the value was not
3570 reset for any messages that followed in the same SMTP session.
3572 54. The store-handling optimization for building very long strings was not
3573 differentiating between the different store pools. I don't think this
3574 actually made any difference in practice, but I've tidied it.
3576 55. While running the routers to verify a sender address, $sender_address
3577 was still set to the sender address. This is wrong, because when routing to
3578 send a bounce to the sender, it would be empty. Therefore, I have changed
3579 it so that, while verifying a sender address, $sender_address is set to <>.
3580 (There is no change to what happens when verifying a recipient address.)
3582 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3583 target A or AAAA records (if not already returned) without resetting the
3584 qualify_single or search_parents options of the DNS resolver. These are
3585 inappropriate in this case because the targets of MX and SRV records must
3586 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3587 target that, when qualified, matched something in the local domain. These
3588 two options are now turned off when doing these lookups.
3590 57. It seems that at least some releases of Reiserfs (which does not have the
3591 concept of a fixed number of inodes) returns zero and not -1 for the
3592 number of available inodes. This interacted badly with check_spool_inodes,
3593 which assumed that -1 was the "no such thing" setting. What I have done is
3594 to check that the total number of inodes is greater than zero before doing
3595 the test of how many are available.
3597 58. When a "warn" ACL statement has a log_message modifier, the message is
3598 remembered, and not repeated. This is to avoid a lot of repetition when a
3599 message has many recipients that cause the same warning to be written.
3600 Howewer, Exim was preserving the list of already written lines for an
3601 entire SMTP session, which doesn't seem right. The memory is now reset if a
3602 new message is started.
3604 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3605 debugging output unless log_rewrite was also set.
3607 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3608 of (int)(handle) when we know that handle contains (void *)(-1).
3610 61. The Exim daemon panic-logs an error return when it closes the incoming
3611 connection. However "connection reset by peer" seems to be common, and
3612 isn't really an error worthy of noting specially, so that particular error
3615 62. When Exim is trying to find all the local interfaces, it used to panic and
3616 die if the ioctl to get the interface flags failed. However, it seems that
3617 on at least one OS (Solaris 9) it is possible to have an interface that is
3618 included in the list of interfaces, but for which you get a failure error
3619 for this call. This happens when the interface is not "plumbed" into a
3620 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3621 failure of the "get flags" call assumes that the interface is down.
3623 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3624 makes life easier for people who are doing arithmetic on fields extracted
3625 from dates, where you often get leading zeros that should not be
3626 interpreted as octal.
3628 64. Added qualify_domain to the redirect router, to override the global
3631 65. If a pathologically long header line contained very many addresses (the
3632 report of this problem mentioned 10 000) and each of them was rewritten,
3633 Exim could use up a very large amount of memory. (It kept on making new
3634 copies of the header line as it rewrote, and never released the old ones.)
3635 At the expense of a bit more processing, the header rewriting function has
3636 been changed so that it no longer eats memory in this way.
3638 66. The generation of the Received: header has been moved from the time that a
3639 message starts to be received, to the time that it finishes. The timestamp
3640 in the Received: header should now be very close to that of the <= log
3641 line. There are two side-effects of this change:
3643 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3644 logged header lines no longer include the local Received: line, because
3645 it has not yet been created. The same applies to a copy of the message
3646 that is returned to a non-SMTP sender when a message is rejected.
3648 (b) When a filter file is tested using -bf, no additional Received: header
3649 is added to the test message. After some thought, I decided that this
3652 This change does not affect the value of $received_for. It is still set
3653 after address rewriting, but before local_scan() is called.
3655 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3657 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3658 gave an unhelpful panic error message, and a defer error. I have managed to
3659 change this behaviour so that it now rejects any supplied certificate,
3660 which seems right, as the list of acceptable certificates is empty.
3662 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3663 gave an unhelpful defer error. I have not managed to make this reject any
3664 supplied certificates, but the error message it gives is "no certificate
3665 supplied", which is not helpful.
3667 70. exigrep's output now also includes lines that are not associated with any
3668 message, but which match the given pattern. Implemented by a patch from
3669 Martin Sluka, which also tidied up the Perl a bit.
3671 71. Recipient callout verification, like sender verification, was using <> in
3672 the MAIL FROM command. This isn't really the right thing, since the actual
3673 sender may affect whether the remote host accepts the recipient or not. I
3674 have changed it to use the actual sender in the callout; this means that
3675 the cache record is now keyed on a recipient/sender pair, not just the
3676 recipient address. There doesn't seem to be a real danger of callout loops,
3677 since a callout by the remote host to check the sender would use <>.
3678 [SEE ABOVE: changed after hitting problems.]
3680 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3681 temporary errors. However, in the case of such a code being given after
3682 the end of a data transmission (i.e. after ".") Exim was failing to write
3683 a retry record for the message. (Yes, there was some broken host that was
3684 actually sending 8xx at this point.)
3686 73. An unknown lookup type in a host list could cause Exim to panic-die when
3687 the list was checked. (An example that provoked this was putting <; in the
3688 middle of a list instead of at the start.) If this happened during a DATA
3689 ACL check, a -D file could be left lying around. This kind of configuration
3690 error no longer causes Exim to die; instead it causes a defer errror. The
3691 incident is still logged to the main and panic logs.
3693 74. Buglet left over from Exim 3 conversion. The message "too many messages
3694 in one connection" was written to the rejectlog but not the mainlog, except
3695 when address rewriting (yes!) was being logged.
3697 75. Added write_rejectlog option.
3699 76. When a system filter was run not as root (that is, when system_filter_user
3700 was set), the values of the $n variables were not being returned to the
3701 main process; thus, they were not subsequently available in the $sn
3704 77. Added +return_path_on_delivery log selector.
3706 78. A connection timeout was being treated differently from recipients deferred
3707 when testing hosts_max_try with a message that was older than the host's
3708 retry timeout. (The host should not be counted, thus allowing all hosts to
3709 be tried at least once before bouncing.) This may have been the cause of an
3710 occasionally reported bug whereby a message would remain on the queue
3711 longer than the retry timeout, but would be bounced if a delivery was
3712 forced. I say "may" because I never totally pinned down the problem;
3713 setting up timeout/retry tests is difficult. See also the next item.
3715 79. The ultimate address timeout was not being applied to errors that involved
3716 a combination of host plus message (for example, a timeout on a MAIL
3717 command). When an address resolved to a number of possible hosts, and they
3718 were not all tried for each delivery (e.g. because of hosts_max_try), a
3719 message could remain on the queue longer than the retry timeout.
3721 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3724 81. Fixed an obscure SMTP outgoing bug which required at least the following
3725 conditions: (a) there was another message waiting for the same server;
3726 (b) the server returned 5xx to all RCPT commands in the first message so
3727 that the message was not completed; (c) the server dropped the connection
3728 or gave a negative response to the RSET that Exim sends to abort the
3729 transaction. The observed case was a dropped connection after DATA that had
3730 been sent in pipelining mode. That is, the server had advertised PIPELINING
3731 but was not implementing it correctly. The effect of the bug was incorrect
3732 behaviour, such as trying another host, and this could lead to a crash.
3738 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3739 and daemon.c were passed as pointers to ints; they should have been
3740 pointers to socklen_t variables (which are typically unsigned ints).
3742 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3745 3. Fixed a really odd bug that affected only the testing scheme; patching a
3746 certain fixed string in the binary changed the value of another string that
3747 happened to be identical to the end of the original first string.
3749 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3750 name", it returns that address as the IP address. On some operating
3751 systems (e.g. Solaris), it also passes back the IP address string as the
3752 "host name". However, on others (e.g. Linux), it passes back an empty
3753 string. Exim wasn't checking for this, and was changing the host name to an
3754 empty string, assuming it had been canonicized.
3756 5. Although rare, it is permitted to have more than one PTR record for a given
3757 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3758 all the names associated with an address, because they do in Solaris.
3759 However, it seems that they do not in Linux for data that comes from the
3760 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3761 I found this out when I moved to a new Linux workstation and tried to run
3762 the Exim test suite.
3764 To get round this problem I have changed the code so that it now does its
3765 own call to the DNS to look up PTR records when searching for a host name.
3766 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3767 addresses that are only in /etc/hosts are still found.
3769 This behaviour is, however, controlled by an option called host_lookup_
3770 order, which defaults to "bydns:byaddr". If people want to use the other
3771 order, or indeed, just use one or the other means of lookup, they can
3772 specify it in this variable.
3774 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3775 some operating systems, this comes back from gethostbyaddr() as an empty
3776 string, and this is what Exim used to test for. However, it seems that in
3777 other systems, "." is yielded. Exim now tests for this case too.
3779 7. The values of check_spool_space and check_log_space are now held internally
3780 as a number of kilobytes instead of an absolute number of bytes. If a
3781 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3782 kilobyte. This means that much larger values can be stored.
3784 8. Exim monitor: an attempt to get the action menu when not actually pointing
3785 at a message produces an empty menu entitled "No message selected". This
3786 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3787 no entries in it ("Shell widget menu has zero width and/or height"). So I
3788 have added a single, blank menu entry in this case.
3790 9. Added ${quote_local_part.
3792 10. MIME decoding is now applied to the contents of Subject: header lines when
3795 11. Now that a reference to $sender_host_address automatically causes a reverse
3796 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3797 host lookup before query-style lookups in lists that might use this
3798 variable. This has therefore been abolished, and the "net-" prefix is no
3799 longer necessary for query-style lookups.
3801 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3802 have been a typo for LFLAGS, so it has been changed.
3804 13. The install script calls Exim with "-C /dev/null" in order to find the
3805 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3806 to be output. Howeve, since Exim outputs its version number before the
3807 error, it didn't break the script. It just looked ugly. I fixed this by
3808 always allowing "-C /dev/null" if the caller is root.
3810 14. Ignore overlarge ACL variable number when reading spool file - insurance
3811 against a later release with more variables having written the file.
3813 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3814 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3815 this, as well as the form without the "IPv6" on the front.
3817 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3818 OS/Makefile-Darwin file.
3820 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3821 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3823 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3824 variable. However, some OS use ints for inodes. I've added cast to long int
3825 to get rid of the compiler warning.
3827 19. I had forgotten to lock out "/../" in configuration file names when
3828 ALT_CONFIG_PREFIX was set.
3830 20. Routers used for verification do not need to specify transports. However,
3831 if such a router generated a host list, and callout was configured, Exim
3832 crashed, because it could not find a port number from the (non-existent)
3833 transport. It now assumes port 25 in this circumstance.
3835 21. Added the -t option to exigrep.
3837 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3838 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3839 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3840 them is set, LOOKUP_LSEARCH is forced.
3842 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3844 24. Added sender and host information to the "rejected by local_scan()" log
3845 line; previously there was no indication of these.
3847 25. Added .include_if_exists.
3849 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3850 Linux. It turns out that not all file systems support this. Apparently some
3851 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3852 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3853 not supported on the file descriptor, is now ignored when Exim is trying to
3854 sync a directory. This applies only to Linux.
3856 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3858 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3859 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3862 29. Added support for maildirsize files from supplied patch (modified a bit).
3864 30. The use of :fail: followed by an empty string could lead Exim to respond to
3865 sender verification failures with (e.g.):
3867 550 Verification failed for <xxx>
3868 550 Sender verify failed
3870 where the first response line was missing the '-' that indicates it is not
3871 the final line of the response.
3873 31. The loop for finding the name of the user that called Exim had a hardwired
3874 limit of 10; it now uses the value of finduser_retries, which is used for
3875 all other user lookups.
3877 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3880 33. Exim was neglecting to zero errno before one call of strtol() when
3881 expanding a string and expecting an integer value. On some systems this
3882 resulted in spurious "integer overflow" errors. Also, it was casting the
3883 result into an int without checking.
3885 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3886 did not work. The code looks as if it has *never* worked, though it appears
3887 to have been documented since at least releast 1.62. I have made it work.
3889 35. The "timeout_DNS" error in retry rules, also documented since at least
3890 1.62, also never worked. As it isn't clear exactly what this means, and
3891 clearly it isn't a major issue, I have abolished the feature by treating it
3892 as "timeout", and writing a warning to the main and panic logs.
3894 36. The display of retry rules for -brt wasn't always showing the error code
3897 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3898 timeout_connect_A, timeout_connect_MX.
3900 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3901 to the empty sender.
3903 39. The daemon was not analysing the content of -oX till after it had closed
3904 stderr and disconnected from the controlling terminal. This meant that any
3905 syntax errors were only noted on the panic log, and the return code from
3906 the command was 0. By re-arranging the code a little, I've made the
3907 decoding happen first, so such errors now appear on stderr, and the return
3908 code is 1. However, the actual setting up of the sockets still happens in
3909 the disconnected process, so errors there are still only recorded on the
3912 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3913 connections (as happens on some IP stacks) was logged at start up time as
3914 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3915 it from "IPv6 and IPv4", which means that two separate sockets are being
3918 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3919 says whether AF_INET or AF_INET6 was passed as an argument.
3921 42. Exiwhat output was messed up when time zones were included in log
3924 43. Exiwhat now gives more information about the daemon's listening ports,
3925 and whether -tls-on-connect was used.
3927 44. The "port" option of the smtp transport is now expanded.
3929 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3930 silently ignored. Now an error message is written to the main and panic
3933 46. There's a new ACL modifier called "logwrite" which writes to a log file
3934 as soon as it is encountered.
3936 47. Added $local_user_uid and $local_user_gid at routing time.
3938 48. Exim crashed when trying to verify a sender address that was being
3941 49. Exim was recognizing only a space character after ".include". It now also
3942 recognizes a tab character.
3944 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3945 extracting the relevant information from the specification. The man page no
3946 longer contains scrambled data for the -d option, and I've added a section
3947 at the front about calling Exim under different names.
3949 51. Added "extra_headers" argument to the "mail" command in filter files.
3951 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3954 53. Installed eximstats 1.29.
3956 54. Added transport_filter_timeout as a generic transport option.
3958 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3959 Cc: header lines. This was required by RFC 822, but it not required by RFC
3962 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3963 incoming messages that did not have them. Now it does so only if the
3964 message originates locally, that is, if there is no associated remote host
3965 address. When Resent- header lines are present, this applies to the Resent-
3966 lines rather than the non-Resent- lines.
3968 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3969 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3971 58. Messages for configuration errors now include the name of the main
3972 configuration file - useful now that there may be more than one file in a
3973 list (.included file names were always shown).
3975 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3976 for those rare installations that do not start the daemon as root or run it
3977 setuid root. I've cut out the call to initgroups() if the daemon is not
3980 60. The Exim user and group can now be bound into the binary as text strings
3981 that are looked up at the start of Exim's processing.
3983 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3985 62. Added $mailstore_basename variable.
3987 63. Installed patch to sieve.c from Michael Haardt.
3989 64. When Exim failed to open the panic log after failing to open the main log,
3990 the original message it was trying to log was written to stderr and debug
3991 output, but if they were not available (the usual case in production), it
3992 was lost. Now it is written to syslog before the two lines that record the
3993 failures to open the logs.
3995 65. Users' Exim filters run in subprocesses under the user's uid. It is
3996 possible for a "deliver" command or an alias in a "personal" command to
3997 provoke an address rewrite. If logging of address rewriting is configured,
3998 this fails because the process is not running as root or exim. There may be
3999 a better way of dealing with this, but for the moment (because 4.30 needs
4000 to be released), I have disabled address rewrite logging when running a
4001 filter in a non-root, non-exim process.
4007 1. The buildconfig auxiliary program wasn't quoting the value set for
4008 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4009 not defined. This bug was masked in 4.22 by the effect that was fixed in
4012 2. Some messages that were rejected after a message id was allocated were
4013 shown as "incomplete" by exigrep. It no longer does this for messages that
4014 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4016 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4017 have allow_domain_literals set, the ID did not get logged in the <= line.
4018 Domain literals are now always recognized in Message-ID: header lines.
4020 4. The first argument for a ${extract expansion item is the key name or field
4021 number. Leading and trailing spaces in this item were not being ignored,
4022 causing some misleading effects.
4024 5. When deliver_drop_privilege was set, single queue runner processes started
4025 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4026 same command in the process it spins off) were not dropping privilege.
4028 6. When the daemon running as "exim" started a queue runner, it always
4029 re-executed Exim in the spun-off process. This is a waste of effort when
4030 deliver_drop_privilege is set. The new process now just calls the
4031 queue-runner function directly.
4037 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4040 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4041 ignored. Though the use of -oP was forcing the writing of a pid file, it
4042 was always written to the default place.
4044 3. If the message "no IP address found for host xxxx" is generated during
4045 incoming verification, it is now followed by identification of the incoming
4046 connection (so you can more easily find what provoked it).
4048 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4050 5. Added some features to "harden" Exim a bit more against certain attacks:
4052 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4053 be put in Local/Makefile. This is like the never_users runtime option,
4054 but it cannot be overridden. The default setting is "root".
4056 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4057 prefix string with which any file named in a -C command line option
4060 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4061 is retained for -C and -D only if the caller of Exim is root. Without
4062 it, the exim user may also use -C and -D and retain privilege.
4064 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4065 command line option is disabled.
4067 6. Macro names set by the -D option must start with an upper case letter, just
4068 like macro names defined in the configuration file.
4070 7. Added "dereference=" facility to LDAP.
4072 8. Two instances of the typo "uknown" in the source files are fixed.
4074 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4075 the Configure-Makefile script screwed up while processing it.
4077 10. Incorporated PCRE 4.4.
4079 11. The SMTP synchronization check was not operating right at the start of an
4080 SMTP session. For example, it could not catch a HELO sent before the client
4081 waited for the greeting. There is now a check for outstanding input at the
4082 point when the greeting is written. Because of the duplex, asynchronous
4083 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4084 way, but not yet received, when the check is performed.
4086 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4087 on TCP/IP sockets, because this apparently causes some broken clients to
4090 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4091 unchanged) from the Cygwin maintainer.
4093 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4094 was supported instead of testing for mbx. Effectively a typo.
4096 15. The spa authenticator server code was not checking that the input it
4097 received was valid base64.
4099 16. The debug output line for the "set" modifier in ACLs was not showing the
4100 name of the variable that was being set.
4102 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4104 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4105 lookup on its own. Something else was needed to trigger the lookup. For
4106 example, a match in host_lookup or the need for a host name in a host list.
4107 Now, if $sender_host_name is referenced and the host name has not yet been
4108 looked up, a lookup is performed. If the lookup fails, the variable remains
4109 empty, and $host_lookup_failed is set to "1".
4111 19. Added "eqi" as a case-independent comparison operator.
4113 20. The saslauthd authentication condition could segfault if neither service
4114 nor realm was specified.
4116 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4117 error message that was logged was misleading, and incoming SMTP
4118 connections were dropped. The message is now more accurate, and temporary
4119 errors are given to SMTP connections.
4121 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4122 (except RSET) with a 421 error, until QUIT is received. However, it was
4123 failing to send a response to QUIT.
4125 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4126 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4127 ACL is now run after the helo_try_verify_hosts code.
4129 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4130 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4131 case-independent, so other case variants are also recognized). Apparently
4132 some systems use these upper case variants.
4134 25. If more than two messages were waiting for the same host, and a transport
4135 filter was specified for the transport, Exim sent two messages over the
4136 same TCP/IP connection, and then failed with "socket operation on non-
4137 socket" when it tried to send the third.
4139 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4141 27. The extern definition of crypt16() in expand.c was not being excluded when
4142 the OS had its own crypt16() function.
4144 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4145 as a preferred synonym for return_size_limit, both as an option and as an
4148 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4150 30. Changed the default configuration ACL to relax the local part checking rule
4151 for addresses that are not in any local domains. For these addresses,
4152 slashes and pipe symbols are allowed within local parts, but the sequence
4153 /../ is explicitly forbidden.
4155 31. SPA server authentication was not clearing the challenge buffer before
4158 32. log_message in a "warn" ACL statement was writing to the reject log as
4159 well as to the main log, which contradicts the documentation and doesn't
4160 seem right (because no rejection is happening). So I have stopped it.
4162 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4163 However, I am unable to do any testing of this.
4165 34. Fixed an infelicity in the appendfile transport. When checking directories
4166 for a mailbox, to see if any needed to be created, it was accidentally
4167 using path names with one or more superfluous leading slashes; tracing
4168 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4170 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4171 contents are added to the log line that is written for every discarded
4172 recipient. (Previously a log_message setting was ignored.)
4174 36. The ${quote: operator now quotes the string if it is empty.
4176 37. The install script runs exim in order to find its version number. If for
4177 some reason other than non-existence or emptiness, which it checks, it
4178 could not run './exim', it was installing it with an empty version number,
4179 i.e. as "exim-". This error state is now caught, and the installation is
4182 38. An argument was missing from the function that creates an error message
4183 when Exim fails to connect to the socket for saslauthd authentication.
4184 This could cause Exim to crash, or give a corrupted message.
4186 39. Added isip, isip4, and isip6 to ${if conditions.
4188 40. The ACL variables $acl_xx are now saved with the message, and can be
4189 accessed later in routers, transports, and filters.
4191 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4192 strings in the file are not string-expanded.
4194 42. If a MAIL command specified a SIZE value that was too large to fit into an
4195 int variable, the check against message_size_limit failed. Such values are
4196 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4197 day this will have to be increased, but I don't think I want to be around
4198 when emails are that large.
4205 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4206 iconv() is not standard in FreeBSD.
4208 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4209 IPv6 enabled. The observed symptom was a segmentation fault on return from
4210 the function os_common_find_running_interfaces() in src/os.c.
4212 3. In the check_special_case() function in daemon.c I had used "errno" as an
4213 argument name, which causes warnings on some systems. This was basically a
4214 typo, since it was named "eno" in the comments!
4216 4. The code that waits for the clock to tick (at a resolution of some fraction
4217 of a second) so as to ensure message-id uniqueness was always waiting for
4218 at least one whole tick, when it could have waited for less. [This is
4219 almost certainly not relevant at current processor speeds, where it is
4220 unlikely to ever wait at all. But we try to future-proof.]
4222 5. The function that sleeps for a time interval that includes fractions of a
4223 second contained a race. It did not block SIGALRM between setting the
4224 timer, and suspending (a couple of lines later). If the interval was short
4225 and the sigsuspend() was delayed until after it had expired, the suspension
4226 never ended. On busy systems this could lead to processes getting stuck for
4229 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4230 process, before it forks any delivery processes. The open lookup caching
4231 mechanism meant that the open file or database connection was passed into
4232 the delivery process. The problem was that delivery processes always tidy
4233 up cached lookup data. This could cause a problem for the next delivery
4234 process started by the queue runner, because the external queue runner
4235 process does not know about the closure. So the next delivery process
4236 still has data in the lookup cache. In the case of a file lookup, there was
4237 no problem because closing a file descriptor in a subprocess doesn't affect
4238 the parent. However, if the lookup was caching a connection to a database,
4239 the connection was closed, and the second delivery process was likely to
4240 see errors such as "PGSQL: query failed: server closed the connection
4241 unexpectedly". The problem has been fixed by closing all cached lookups
4242 in a queue runner before running a delivery process.
4244 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4245 seem to have the "const" qualifier which it has on other OS. I've
4248 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4249 *of the same type* (client or server) with the same public name that an
4250 error should be diagnosed.
4252 9. When Exim looked up a host name for an IP address, but failed to find the
4253 original IP address when looking up the host name (a safety check), it
4254 output the message "<ip address> does not match any IP for NULL", which was
4255 confusing, to say the least. The bug was that the host name should have
4256 appeared instead of "NULL".
4258 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4259 user that is built into the binary, and the -C or -D options is used, root
4260 privilege is dropped before the configuration file is read. In addition,
4261 logging is switched to stderr instead of the normal log files. If the
4262 configuration then re-defines the Exim user, the unprivileged environment
4263 is probably not what is expected, so Exim logs a panic warning message (but
4266 However, if deliver_drop_privilege is set, the unprivileged state may well
4267 be exactly what is intended, so the warning has been cut out in that case,
4268 and Exim is allowed to try to write to its normal log files.
4274 1. smtp_return_error_details was not giving details for temporary sender
4275 or receiver verification errors.
4277 2. Diagnose a configuration error if two authenticators have the same public
4280 3. Exim used not to create the message log file for a message until the first
4281 delivery attempt. This could be confusing when incoming messages were held
4282 for policy or load reasons. The message log file is now created at the time
4283 the message is received, and an initial "Received" line is written to it.
4285 4. The automatically generated man page for command line options had a minor
4286 bug that caused no ill effects; however, a more serious problem was that
4287 the procedure for building the man page automatically didn't always
4288 operate. Consequently, release 4.20 contains an out-of-date version. This
4289 shouldn't happen again.
4291 5. When building Exim with embedded Perl support, the script that builds the
4292 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4293 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4295 6. The freeze_tell option was not being used for messages that were frozen on
4296 arrival, either by an ACL or by local_scan().
4298 7. Added the smtp_incomplete_transaction log selector.
4300 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4301 was accepting AUTH without a new EHLO.
4303 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4304 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4307 10. Logging of TCP/IP connections (when configured) now happens in the main
4308 daemon process instead of the child process, so that the TCP/IP connection
4309 count is more accurate (but it can never be perfect).
4311 11. The use of "drop" in a nested ACL was not being handled correctly in the
4312 outer ACL. Now, if condition failure induced by the nested "drop" causes
4313 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4314 or "require"), the connection is dropped.
4316 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4317 that yield "discard" can now be used with an "accept" or a "discard" verb,
4318 but an error is generated for any others (because I can't see a useful way
4319 to define what should happen).
4321 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4322 are now processed in the same way as lines in the Exim configuration file.
4323 In particular, continuation lines are supported.
4325 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4327 15. Added -ti meaning -t -i.
4329 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4330 domains, and warn by logging if others are found.
4332 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4333 returned from GIFCONF when Exim is trying to find the list of interfaces on
4334 a host. The code in os.c has been modified to copy each ifreq to an aligned
4335 structure in all cases.
4337 Also, in some cases, the returned ifreq's were being copied to a 'struct
4338 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4339 means the last couple of bytes of an IPv6 address could be chopped if the
4340 ifreq contained only a normal sockaddr (14 bytes storage).
4342 18. Named domain lists were not supported in the hosts_treat_as_local option.
4343 An entry such as +xxxx was not recognized, and was treated as a literal
4346 19. Ensure that header lines added by a DATA ACL are included in the reject log
4347 if the ACL subsequently rejects the message.
4349 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4350 MD5 (which is deprecated).
4352 21. When testing a filter file using -bf, Exim was writing a message when it
4353 took the sender from a "From " line in the message, but it was not doing so
4354 when it took $return_path from a Return-Path: header line. It now does.
4356 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4357 with a valid header line field name (a series of printing characters
4358 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4360 23. Changed "disc" in the source to "disk" to conform to the documentation and
4361 the book and for uniformity.
4363 24. Ignore Sendmail's -Ooption=value command line item.
4365 25. When execve() failed while trying to run a command in a pipe transport,
4366 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4367 could be confused with a return value of 69 from the command itself. This
4368 has been changed to 127, the value the shell returns if it is asked to run
4369 a non-existent command. The wording for the related log line suggests a
4370 non-existent command as the problem.
4372 26. If received_header_text expands to an empty string, do not add a Received:
4373 header line to the message. (Well, it adds a token one on the spool, but
4374 marks it "old" so that it doesn't get used or transmitted.)
4376 27. Installed eximstats 1.28 (addition of -nt option).
4378 28. There was no check for failure on the call to getsockname() in the daemon
4379 code. This can fail if there is a shortage of resources on the system, with
4380 ENOMEM, for example. A temporary error is now given on failure.
4382 29. Contrary to the C standard, it seems that in some environments, the
4383 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4384 program. Exim now does this explicitly; it affects the formatting of
4385 timestamps using strftime().
4387 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4388 complaints. I've now initialized all the variables, to avoid this.
4390 32. Header lines added by a system filter were not being "seen" during
4391 transport-time rewrites.
4393 33. The info_callback() function passed to OpenSSL is set up with type void
4394 (*)(SSL *, int, int), as described somewhere. However, when calling the
4395 function (actually a macro) that sets it up, the type void(*)() is
4396 expected. I've put in a cast to prevent warnings from picky compilers.
4398 34. If a DNS black list lookup found a CNAME record, but there were no A
4399 records associated with the domain it pointed at, Exim crashed.
4401 35. If a DNS black list lookup returned more than one A record, Exim ignored
4402 all but the first. It now scans all returned addresses if a particular IP
4403 value is being sought. In this situation, the contents of the
4404 $dnslist_value variable are a list of all the addresses, separated by a
4407 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4408 to accept a host name and all its aliases if the forward lookup for any of
4409 them yielded the IP address of the incoming connection. Now it accepts only
4410 those names whose forward lookup yields the correct IP address. Any other
4411 names are discarded. This closes a loophole whereby a rogue DNS
4412 administrator could create reverse DNS records to break through a
4413 wildcarded host restriction in an ACL.
4415 37. If a user filter or a system filter that ran in a subprocess used any of
4416 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4417 the wrong values were passed to the pipe command ($thisaddress had the
4418 value of $0, $0 had the value of $1, etc). This bug was introduced by
4419 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4421 38. Improved the line breaking for long SMTP error messages from ACLs.
4422 Previously, if there was no break point between 40 and 75 characters, Exim
4423 left the rest of the message alone. Two changes have been made: (a) I've
4424 reduced the minimum length to 35 characters; (b) if it can't find a break
4425 point between 35 and 75 characters, it looks ahead and uses the first one
4426 that it finds. This may give the occasional overlong line, but at least the
4427 remaining text gets split now.
4429 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4430 file descriptors might be low, and that setting 1000 would always raise it.
4431 It turns out that in some environments, the limit is already over 1000 and
4432 that lowering it causes trouble. So now Exim takes care not to decrease it.
4434 40. When delivering a message, the value of $return_path is set to $sender_
4435 address at the start of routing (routers may change the value). By an
4436 oversight, this default was not being set up when an address was tested by
4437 -bt or -bv, which affected the outcome if any router or filter referred to
4440 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4441 log only when "message" or "log_message" are set. However, if one of the
4442 conditions was an address verification, or a call to a nested ACL, the
4443 messages generated by the underlying test were being passed through. This
4444 no longer happens. The underlying message is available in $acl_verify_
4445 message for both "message" and "log_message" expansions, so it can be
4446 passed through if needed.
4448 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4449 new expansion $bh_ to give the encoded byte string without charset
4450 translation. Translation happens only if iconv() is available; HAVE_ICONV
4451 indicates this at build time. HEADERS_CHARSET gives the charset to
4452 translate to; headers_charset can change it in the configuration, and
4453 "headers charset" can change it in an individual filter file.
4455 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4456 that creates RFC 2047 encoded "words" labels them as that charset instead
4457 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4458 expansion operator; (ii) when Exim creates a From: line for a local
4459 message; (iii) when a header line is rewritten to include a "phrase" part.
4461 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4462 buggy, causing it to skip the first lines of messages whose message ID
4463 ended in 'D'. This would not have bitten before Exim release 4.14, because
4464 message IDs were unlikely to end in 'D' before then. The effect was to have
4465 incorrect size information for certain domains.
4467 45. #include "config.h" was missing at the start of the crypt16.c module. This
4468 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4471 46. If there was a timeout during a "random" callout check, Exim treated it as
4472 a failure of the random address, and carried on sending RSET and the real
4473 address. If the delay was just some slowness somewhere, the response to the
4474 original RCPT would be taken as a response to RSET and so on, causing
4475 mayhem of various kinds.
4477 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4478 when I implemented it. It didn't allow for the fact that some option values
4479 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4480 the right test for positive values.
4482 48. Domain names in DNS records are case-independent. Exim always looks them up
4483 in lower case. Some resolvers return domain names in exactly the case they
4484 appear in the zone file, that is, they may contain uppercase letters. Not
4485 all resolvers do this - some return always lower case. Exim was treating a
4486 change of case by a resolver as a change of domain, similar to a widening
4487 of a domain abbreviation. This triggered its re-routing code and so it was
4488 trying to route what was effectively the same domain again. This normally
4489 caused routing to fail (because the router wouldn't handle the domain
4490 twice). Now Exim checks for this case specially, and just changes the
4491 casing of the domain that it ultimately uses when it transmits the message
4494 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4497 50. If a filter generated a file delivery with a non-absolute name (possible if
4498 no home directory exists for the router), the forbid_file option was not
4501 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4502 the existing equality matching.
4504 52. Exim was using ints instead of ino_t variables in some places where it was
4505 dealing with inode numbers.
4507 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4508 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4509 called TMPDIR, and if it finds it is different, it changes its value.
4511 54. The smtp_printf() function is now made available to local_scan() so
4512 additional output lines can be written before returning. There is also an
4513 smtp_fflush() function to enable the detection of a dropped connection.
4514 The variables smtp_input and smtp_batched_input are exported to
4517 55. Changed the default runtime configuration: the message "Unknown user"
4518 has been removed from the ACL, and instead placed on the localuser router,
4519 using the cannot_route_message feature. This means that any verification
4520 failures that generate their own messages won't get overridden. Similarly,
4521 the "Unrouteable address" message that was in the ACL for unverifiable
4522 relay addresses has also been removed.
4524 56. Added hosts_avoid_esmtp to the smtp transport.
4526 57. The exicyclog script was not checking for the esoteric option
4527 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4528 will work only if exicyclog is run under the appropriate euid.
4530 58. Following a discussion on the list, the rules by which Exim recognises line
4531 endings on incoming messages have been changed. The -dropcr and drop_cr
4532 options are now no-ops, retained only for backwards compatibility. The
4533 following line terminators are recognized: LF CRLF CR. However, special
4534 processing applies to CR:
4536 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4537 nor a local message in the state where . is a terminator.
4539 (ii) If a bare CR is encountered in a header line, an extra space is added
4540 after the line terminator so as not to end the header. The reasoning
4541 behind this is that bare CRs in header lines are most likely either
4542 to be mistakes, or people trying to play silly games.
4544 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4545 was being incorrectly given as 18 bytes larger than it should have been.
4546 This is a VOB (very old bug).
4548 60. This may never have affected anything current, but just in case it has:
4549 When the local host is found other than at the start of a list of hosts,
4550 the local host, those with the same MX, and any that follow, are discarded.
4551 When the list in question was part of a longer list of hosts, the following
4552 hosts (not currently being processed) were also being discarded. This no
4553 longer happens. I'm not sure if this situation could ever has previously
4556 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4559 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4560 header. This is something that is recommended in a new Internet Draft, and
4561 is something that is documented as being done by Sendmail. There are two
4562 possible values. For messages generated by the autoreply transport, Exim
4565 Auto-Submitted: auto-replied
4567 whereas for all other generated messages (e.g. bounces) it adds
4569 Auto-Submitted: auto-generated
4571 63. The "personal" condition in filters now includes a test for the
4572 Auto-Submitted: header. If it contains the string "auto-" the message it
4573 not considered personal.
4575 64. Added rcpt_include_affixes as a generic transport option.
4577 65. Added queue_only_override (default true).
4579 66. Added the syslog_duplication option.
4581 67. If what should have been the first header line of a message consisted of
4582 a space followed by a colon, Exim was mis-interpreting it as a header line.
4583 It isn't of course - it is syntactically invalid and should therefore be
4584 treated as the start of the message body. The misbehaviour could have
4585 caused a number of strange effects, including loss of data in subsequent
4586 header lines, and spool format errors.
4588 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4589 client host had authenticated. This control can now be exercised by an ACL
4590 for more flexibility.
4592 69. By default, callouts do not happen when testing with -bh. There is now a
4593 variant, -bhc, which does actually run the callout code, including
4594 consulting and updating the callout cache.
4596 70. Added support for saslauthd authentication, courtesy of Alexander
4599 71. If statvfs() failed on the spool or log directories while checking their
4600 size for availability, Exim confusingly gave the error "space shortage".
4601 Furthermore, in debugging mode it crashed with a floating point exception.
4602 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4603 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4604 serious problem, Exim now writes to the main and panic logs when this
4605 happens, with details of the failure. It then refuses to accept the
4606 incoming message, giving the message "spool directory problem" or "log
4607 directory problem" with a 421 code for SMTP messages.
4609 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4610 0, 1, and 2 exist, because some OS complain for execs without them (see
4611 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4612 descriptors. However, the code omitted to check that the open succeeded,
4613 causing mysterious errors if for some reason the permissions on /dev/null
4614 got screwed. Now Exim writes a message to the main and panic logs, and
4615 bombs out if it can't open /dev/null.
4617 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4618 interact so that it is all more flexible. It is supposed to remain
4619 backwards compatible. Also added extra_local_interfaces.
4621 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4622 to bomb out with an assertion failure - to the client this appears as a
4623 connection drop. This problem occurs in the part of the code that was taken
4624 from the Samba project. Fortunately, the assertion is in a very simple
4625 function, so I have fixed this by reproducing the function inline in the
4626 one place where it is called, and arranging for authentication to fail
4627 instead of killing the process with assert().
4629 75. The SPA client code was not working when the server requested OEM rather
4630 than Unicode encoding.
4632 76. Added code to make require_files with a specific uid setting more usable in
4633 the case where statting the file as root fails - usually a non-root-mounted
4634 NFS file system. When this happens and the failure is EACCES, Exim now
4635 forks a subprocess and does the per-uid checking as the relevant uid.
4637 77. Added process_log_path.
4639 78. If log_file_path was not explicitly set, a setting of check_log_space or
4640 check_log_inodes was ignored.
4642 79. If a space check for the spool or log partitions fails, the incident is now
4643 logged. Of course, in the latter case the data may get lost...
4645 80. Added the %p formatting code to string_format() so that it can be used to
4646 print addresses in debug_print(). Adjusted all the address printing in the
4647 debugging in store.c to use %p rather than %d.
4649 81. There was a concern that a line of code in smtp_in.c could overflow a
4650 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4651 initially expressed, the concern was not well-founded, because trailing
4652 spaces are removed early. However, if the trailing spaces were followed by
4653 a NULL, they did not get removed, so the overflow was possible. Two fixes
4656 (a) I re-wrote the offending code in a cleaner fashion.
4657 (b) If an incoming SMTP command contains a NULL character, it is rejected
4660 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4661 runs initgroups(), so that if the Exim user is in any additional groups,
4662 they will be used during message reception.
4668 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob