1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.485 2007/02/26 14:07:04 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.
140 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
141 fixed by 4.65/MH/01 (is this a record?) are fixed:
143 (i) An empty string was always treated as zero by the numeric comparison
144 operators. This behaviour has been restored.
146 (ii) It is documented that the numeric comparison operators always treat
147 their arguments as decimal numbers. This was broken in that numbers
148 starting with 0 were being interpreted as octal.
150 While fixing these problems I realized that there was another issue that
151 hadn't been noticed. Values of message_size_limit (both the global option
152 and the transport option) were treated as octal if they started with 0.
153 The documentation was vague. These values are now always treated as
154 decimal, and I will make that clear in the documentation.
160 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
161 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
164 MH/01 Don't check that the operands of numeric comparison operators are
165 integers when their expansion is in "skipping" mode (fixes bug
166 introduced by 4.64-PH/07).
168 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
169 child addresses, Exim now panics and dies. Previously, because the count
170 is held in a short int, deliveries were likely to be lost. As such a
171 large number of recipients for a single message is ridiculous
172 (performance will be very, very poor), I have chosen to impose a limit
173 rather than extend the field.
179 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
180 leftover -K file (the existence of which was triggered by #402).
181 While we were at it, introduced process PID as part of the -K
182 filename. This should rule out race conditions when creating
185 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
186 processing considerably. Previous code took too long for large mails,
187 triggering a timeout which in turn triggers #401.
189 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
190 in the DK code in transports.c. sendfile() is not really portable,
191 hence the _LINUX specificness.
193 TF/01 In the add_headers option to the mail command in an Exim filter,
194 there was a bug that Exim would claim a syntax error in any
195 header after the first one which had an odd number of characters
198 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
199 callout verification, Exim cached a "reject" for the entire domain. This
200 is correct for most verifications, but it is not correct for a recipient
201 verification with use_sender or use_postmaster set, because in that case
202 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
203 case of MAIL FROM:<> rejection from other early rejections (e.g.
204 rejection of HELO). When verifying a recipient using a non-null MAIL
205 address, the cache is ignored if it shows MAIL FROM:<> rejection.
206 Whatever the result of the callout, the value of the domain cache is
207 left unchanged (for any other kind of callout, getting as far as trying
208 RCPT means that the domain itself is ok).
210 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
213 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
214 manifest itself as EPIPE rather than ECONNECT. When tidying away a
215 session, the daemon ignores ECONNECT errors and logs others; it now
216 ignores EPIPE as well.
218 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
219 (quoted-printable decoding).
221 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
222 later the small subsequent patch to fix an introduced bug.
224 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
226 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
228 PH/08 An error is now given if message_size_limit is specified negative.
230 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
231 to be given (somewhat) arbitrary names.
233 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
236 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
237 miscellaneous code fixes
239 PH/10 Added the log_reject_target ACL modifier to specify where to log
242 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
243 hostname. This is wrong, because it relates to the incoming message (and
244 probably the interface on which it is arriving) and not to the outgoing
245 callout (which could be using a different interface). This has been
246 changed to use the value of the helo_data option from the smtp transport
247 instead - this is what is used when a message is actually being sent. If
248 there is no remote transport (possible with a router that sets up host
249 addresses), $smtp_active_hostname is used.
251 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
252 tweaks were necessary in order to get it to work (see also 21 below):
253 (a) The code assumed that strncpy() returns a negative number on buffer
254 overflow, which isn't the case. Replaced with Exim's string_format()
256 (b) There were several signed/unsigned issues. I just did the minimum
257 hacking in of casts. There is scope for a larger refactoring.
258 (c) The code used strcasecmp() which is not a standard C function.
259 Replaced with Exim's strcmpic() function.
260 (d) The code set only $1; it now sets $auth1 as well.
261 (e) A simple test gave the error "authentication client didn't specify
262 service in request". It would seem that Dovecot has changed its
263 interface. Fortunately there's a specification; I followed it and
264 changed what the client sends and it appears to be working now.
266 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
269 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
270 address A is aliased to B and C, where B exists and C does not. Without
271 -v the output is "A verified" because verification stops after a
272 successful redirection if more than one address is generated. However,
273 with -v the child addresses are also verified. Exim was outputting "A
274 failed to verify" and then showing the successful verification for C,
275 with its parentage. It now outputs "B failed to verify", showing B's
276 parentage before showing the successful verification of C.
278 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
279 look up a TXT record in a specific list after matching in a combined
282 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
283 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
284 they consult the DNS. I had assumed they would set it the way they
285 wanted; and indeed my experiments on Linux seem to show that in some
286 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
287 To be on the safe side, however, I have now made the interface to
288 host_find_byname() similar to host_find_bydns(), with an argument
289 containing the DNS resolver options. The host_find_byname() function now
290 sets these options at its start, just as host_find_bydns() does. The smtp
291 transport options dns_qualify_single and dns_search_parents are passed to
292 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
293 of host_find_byname() use the default settings of RES_DEFNAMES
294 (qualify_single) but not RES_DNSRCH (search_parents).
296 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
297 spool_read_header() do less string testing, by means of a preliminary
298 switch on the second character of optional "-foo" lines. (This is
299 overdue, caused by the large number of possibilities that now exist.
300 Originally there were few.) While I was there, I also converted the
301 str(n)cmp tests so they don't re-test the leading "-" and the first
302 character, in the hope this might squeeze out yet more improvement.
304 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
305 flag allowing group syntax was set by the header_syntax check but not
306 turned off, possible causing trouble later; (2) The flag was not being
307 set at all for the header_verify test, causing "group"-style headers to
308 be rejected. I have now set it in this case, and also caused header_
309 verify to ignore an empty address taken from a group. While doing this, I
310 came across some other cases where the code for allowing group syntax
311 while scanning a header line wasn't quite right (mostly, not resetting
312 the flag correctly in the right place). These bugs could have caused
313 trouble for malformed header lines. I hope it is now all correct.
315 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
316 with the "reply" argument non-NULL. The code, however (which originally
317 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
318 but it didn't always do it. This confused somebody who was copying the
319 code for some other use. I have removed all the tests.
321 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
322 feature that was used to support insecure browsers during the U.S. crypto
323 embargo. It requires special client support, and Exim is probably the
324 only MTA that supported it -- and would never use it because real RSA is
325 always available. This code has been removed, because it had the bad
326 effect of slowing Exim down by computing (never used) parameters for the
327 RSA_EXPORT functionality.
329 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
330 authenticator to fail if there's a tab character in the incoming data
331 (there should never be unless someone is messing about, as it's supposed
332 to be base64-encoded). Also added, on Timo's advice, the "secured" option
333 if the connection is using TLS or if the remote IP is the same as the
334 local IP, and the "valid-client-cert option" if a client certificate has
337 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
338 authenticators. This can be used for authorization after authentication
339 succeeds. (In the case of plaintext, it servers for both authentication
342 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
343 if any retry times were supplied.
345 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
346 connection, where there is no client IP address to check. In this
347 situation, the verify now always succeeds.
349 PH/25 Applied John Jetmore's -Mset patch.
351 PH/26 Added -bem to be like -Mset, but loading a message from a file.
353 PH/27 In a string expansion for a processed (not raw) header when multiple
354 headers of the same name were present, leading whitespace was being
355 removed from all of them, but trailing whitespace was being removed only
356 from the last one. Now trailing whitespace is removed from each header
357 before concatenation. Completely empty headers in a concatenation (as
360 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
361 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
363 PH/29 [Removed. This was a change that I later backed out, and forgot to
364 correct the ChangeLog entry (that I had efficiently created) before
365 committing the later change.]
367 PH/30 Exim was sometimes attempting to deliver messages that had suffered
368 address errors (4xx response to RCPT) over the same connection as other
369 messages routed to the same hosts. Such deliveries are always "forced",
370 so retry times are not inspected. This resulted in far too many retries
371 for the affected addresses. The effect occurred only when there were more
372 hosts than the hosts_max_try setting in the smtp transport when it had
373 the 4xx errors. Those hosts that it had tried were not added to the list
374 of hosts for which the message was waiting, so if all were tried, there
375 was no problem. Two fixes have been applied:
377 (i) If there are any address or message errors in an SMTP delivery, none
378 of the hosts (tried or untried) are now added to the list of hosts
379 for which the message is waiting, so the message should not be a
380 candidate for sending over the same connection that was used for a
381 successful delivery of some other message. This seems entirely
382 reasonable: after all the message is NOT "waiting for some host".
383 This is so "obvious" that I'm not sure why it wasn't done
384 previously. Hope I haven't missed anything, but it can't do any
385 harm, as the worst effect is to miss an optimization.
387 (ii) If, despite (i), such a delivery is accidentally attempted, the
388 routing retry time is respected, so at least it doesn't keep
389 hammering the server.
391 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
392 in ${readsocket because some servers need this prod.
394 PH/32 Added some extra debug output when updating a wait-xxx database.
396 PH/33 The hint "could be header name not terminated by colon", which has been
397 given for certain expansion errors for a long time, was not being given
398 for the ${if def:h_colon_omitted{... case.
400 PH/34 The spec says: "With one important exception, whenever a domain list is
401 being scanned, $domain contains the subject domain." There was at least
402 one case where this was not true.
404 PH/35 The error "getsockname() failed: connection reset by peer" was being
405 written to the panic log as well as the main log, but it isn't really
406 panic-worthy as it just means the connection died rather early on. I have
407 removed the panic log writing for the ECONNRESET error when getsockname()
410 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
411 runs only) independently of the message's sender address. This meant
412 that, if the 4xx error was in fact related to the sender, a different
413 message to the same recipient with a different sender could confuse
414 things. In particualar, this can happen when sending to a greylisting
415 server, but other circumstances could also provoke similar problems.
416 I have changed the default so that the retry time for these errors is now
417 based a combination of the sender and recipient addresses. This change
418 can be overridden by setting address_retry_include_sender=false in the
421 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
422 remote server are returned as part of bounce messages. This was not
423 happening for LMTP over a pipe (the lmtp transport), but now it is the
424 same for both kinds of LMTP.
426 PH/38 Despite being documented as not happening, Exim was rewriting addresses
427 in header lines that were in fact CNAMEs. This is no longer the case.
429 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
430 and queue runs started by the daemon processed all messages. This has
431 been fixed so that -R and -S can now usefully be given with -q<time>.
433 PH/40 Import PCRE release 6.7 (fixes some bugs).
435 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
437 PH/42 Give an error if -q is specified more than once.
439 PH/43 Renamed the variables $interface_address and $interface_port as
440 $received_ip_address and $received_port, to make it clear that these
441 values apply to message reception, and not to the outgoing interface when
442 a message is delivered. (The old names remain recognized, of course.)
444 PH/44 There was no timeout on the connect() call when using a Unix domain
445 socket in the ${readsocket expansion. There now is.
447 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
448 be meaningful with "accept".
450 SC/01 Eximstats V1.43
451 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
453 SC/02 Eximstats V1.44
454 Use a glob alias rather than an array ref in the generated
455 parser. This improves both readability and performance.
457 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
458 Collect SpamAssassin and rejection statistics.
459 Don't display local sender or destination tables unless
460 there is data to show.
461 Added average volumes into the top table text output.
463 SC/04 Eximstats V1.46
464 Collect data on the number of addresses (recipients)
465 as well as the number of messages.
467 SC/05 Eximstats V1.47
468 Added 'Message too big' to the list of mail rejection
469 reasons (thanks to Marco Gaiarin).
471 SC/06 Eximstats V1.48
472 Mainlog lines which have GMT offsets and are too short to
473 have a flag are now skipped.
475 SC/07 Eximstats V1.49 (Alain Williams)
476 Added the -emptyok flag.
478 SC/08 Eximstats V1.50
479 Fixes for obtaining the IP address from reject messages.
481 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
482 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
483 whitesspace changes from 4.64-PH/27
485 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
488 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
489 are found, allow negative numbers in numeric criteria)
491 JJ/06 exipick.20061117.2, added new $message_body_missing variable
493 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
494 to match changes made in 4.64-PH/43
496 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
498 PH/47 Put in an explicit test for a DNS lookup of an address record where the
499 "domain" is actually an IP address, and force a failure. This locks out
500 those revolvers/nameservers that support "A-for-A" lookups, in
501 contravention of the specifications.
503 PH/48 When a host name was looked up from an IP address, and the subsequent
504 forward lookup of the name timed out, the host name was left in
505 $sender_host_name, contrary to the specification.
507 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
508 restricted to single-key lookups, Exim was not diagnosing an error if
509 * or *@ was used with a query-style lookup.
511 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
513 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
514 long ago, but noone interested enough thought of it. Let's just say that
515 the "1.1" means that there are some new functions that weren't there at
516 some point in the past.
518 PH/51 Error processing for expansion failure of helo_data from an smtp
519 transport during callout processing was broken.
521 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
522 tested/used via the -bh/-bhc/-bs options.
524 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
525 bug, fixed in subsequent PCRE releases).
527 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
528 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
530 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
536 SC/01 Use a glob alias rather than an array ref in eximstats generated
537 parser. This improves both readability and performance.
539 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
540 Don't display local sender or destination tables in eximstats unless
541 there is data to show.
542 Added average volumes into the eximstats top table text output.
544 SC/03 Collect data on the number of addresses (recipients) as well
545 as the number of messages in eximstats.
547 TF/01 Correct an error in the documentation for the redirect router. Exim
548 does (usually) call initgroups() when daemonizing.
550 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
551 with consistent privilege compared to when running as a daemon.
553 TF/03 Note in the spec that $authenticated_id is not set for local
554 submissions from trusted users.
556 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
557 Thanks to Dean Brooks <dean@iglou.com> for the patch.
559 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
560 by adding some example configuration directives to the default
561 configuration file. A little bit of work is required to uncomment the
562 directives and define how usernames and passwords are checked, but
563 there is now a framework to start from.
565 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
566 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
567 without this. I don't know how relevant this is to other LDAP libraries.
569 PH/02 Add the verb name to the "unknown ACL verb" error.
571 PH/03 Magnus Holmgren's patch for filter_prepend_home.
573 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
575 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
576 directory not expanded when it should be if an expanded home directory
577 was set for the address (which is overridden by the transport).
579 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
582 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
583 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
584 because it is too late at that time, and has no effect.
586 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
587 security issue with \' (bugzilla #107). I could not use the
588 PQescapeStringConn() function, because it needs a PGconn value as one of
591 PH/08 When testing addresses using -bt, indicate those final addresses that
592 are duplicates that would not cause an additional delivery. At least one
593 person was confused, thinking that -bt output corresponded to deliveries.
594 (Suppressing duplicates isn't a good idea as you lose the information
595 about possibly different redirections that led to the duplicates.)
597 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
598 systems where poll() doesn't work, in particular OS X.
600 PH/10 Added more information to debugging output for retry time not reached.
602 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
603 operations in malware.c.
605 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
608 PH/13 If write_rejectlog was set false when logging was sent to syslog with
609 syslog_duplication set false, log lines that would normally be written
610 both the the main log and to the reject log were not written to syslog at
613 PH/14 In the default configuration, change the use of "message" in ACL warn
614 statements to "add_header".
616 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
617 not followed by a command (e.g. "seen endif").
619 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
620 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
623 PH/17 Added extra conditions to the default value of delay_warning_condition
627 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
628 { match{$h_precedence:}{(?i)bulk|list|junk} } \
629 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
632 The Auto-Submitted: and various List- headers are standardised, whereas I
633 don't think Precedence: ever was.
635 PH/18 Refactored debugging code in route_finduser() to show more information,
636 in particular, the error code if getpwnam() issues one.
638 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
639 This is apparently needed in addition to the PH/07 change above to avoid
640 any possible encoding problems.
642 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
643 but not after initializing Perl.
645 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
646 output them only if debugging. By default they are written stderr,
647 apparently, which is not desirable.
649 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
652 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
655 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
657 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
658 authenticated or an ident call has been made. Suppress the default
659 values for $authenticated_id and $authenticated_sender (but permit -oMai
660 and -oMas) when testing with -bh.
662 PH/24 Re-jigged the order of the tests in the default configuration so that the
663 tests for valid domains and recipients precede the DNS black list and CSA
664 tests, on the grounds that those ones are more expensive.
666 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
667 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
668 This bug exists in every version of Exim that I still have, right back to
671 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
672 However, an attempt to turn on case-sensitivity in a regex key by
673 including (?-i) didn't work because the subject string was already
674 lowercased, and the effects were non-intuitive. It turns out that a
675 one-line patch can be used to allow (?-i) to work as expected.
681 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
682 other effects) broke the use of negated acl sub-conditions.
684 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
687 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
688 "Deny" causes Exim to reject the incoming connection with a 554 error.
689 Unfortunately, if there is a major crisis, such as a disk failure,
690 tcp-wrappers gives "deny", whereas what one would like would be some
691 kind of temporary error. A kludge has been added to help with this.
692 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
693 554 error is used if errno is still zero or contains ENOENT (which occurs
694 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
697 PH/03 Add -lutil to the default FreeBSD LIBS setting.
699 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
700 errors. Otherwise a message that provokes a temporary error (when other
701 messages do not) can cause a whole host to time out.
703 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
704 addresses were routed directly to files or pipes from a redirect router.
705 File deliveries just didn't batch; pipe deliveries might have suffered
708 PH/06 A failure to get a lock for a hints database would erroneously always say
709 "Failed to get write lock", even when it was really a read lock.
711 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
712 of 0600. This has been changed to use the value of the lockfile_mode
713 option (which defaults to 0600).
715 PH/08 Applied small patch from the Sieve maintainer.
717 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
718 folder from quota calculations, a direct delivery into this folder messed
719 up the contents of the maildirsize file. This was because the regex was
720 used only to exclude .Trash (or whatever) when the size of the mailbox
721 was calculated. There was no check that a delivery was happening into an
722 excluded directory. This bug has been fixed by ignoring all quota
723 processing for deliveries into excluded directories.
725 PH/10 Added the maildirfolder_create_regex option to appendfile.
731 PH/01 The code for finding all the local interface addresses on a FreeBSD
732 system running IPv6 was broken. This may well have applied to all BSD
733 systems, as well as to others that have similar system calls. The broken
734 code found IPv4 interfaces correctly, but gave incorrect values for the
735 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
736 that it would not match correctly against @[] and not recognize the IPv6
739 PH/02 The ipliteral router was not recognizing addresses of the form user@
740 [ipv6:....] because it didn't know about the "ipv6:" prefix.
742 PH/03 Added disable_ipv6.
744 PH/04 Changed $reply_address to use the raw form of the headers instead of the
745 decoded form, because it is most often used to construct To: headers
746 lines in autoreplies, and the decoded form may well be syntactically
747 invalid. However, $reply_address has leading white space removed, and all
748 newlines turned into spaces so that the autoreply transport does not
751 PH/05 If group was specified without a user on a router, and no group or user
752 was specified on a transport, the group from the router was ignored.
754 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
755 for visible compile-time settings that can be used to change these
756 numbers, for those that want even more. Backwards compatibility with old
757 spool files has been maintained. However, going back to a previous Exim
758 release will lost any variables that are in spool files.
760 PH/07 Two small changes when running in the test harness: increase delay when
761 passing a TCP/IP connection to a new process, in case the original
762 process has to generate a bounce, and remove special handling of
763 127.0.0.2 (sic), which is no longer necessary.
765 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
766 be the same on different OS.
768 PH/09 Moved a debug statement in filter processing to avoid a race problem when
771 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
772 whether --show-vars was specified or not
774 JJ/02 exipick: Added support for new ACL variable spool format introduced
777 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
778 syntactically invalid From: or Reply-to: line, and a filter used this to
779 generate an autoreply, and therefore failed to obtain an address for the
780 autoreply, Exim could try to deliver to a non-existent relative file
781 name, causing unrelated and misleading errors. What now happens is that
782 it logs this as a hard delivery error, but does not attempt to create a
785 PH/11 The exinext utility has a -C option for testing purposes, but although
786 the given file was scanned by exinext itself; it wasn't being passed on
787 when Exim was called.
789 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
790 an end-of-file indication when reading a command response.
792 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
793 compiled. In many other places in Exim, IPv6 addresses are always
794 recognized, so I have changed this. It also means that IPv4 domain
795 literals of the form [IPV4:n.n.n.n] are now always recognized.
797 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
798 used if the router is not running as root, for example, when verifying at
799 ACL time, or when using -bh. The debugging output from this situation was
800 non-existent - all you got was a failure to exec. I have made two
803 (a) Failures to set uid/gid, the current directory, or a process leader
804 in a subprocess such as that created by queryprogram now generate
805 suitable debugging ouput when -d is set.
807 (b) The queryprogram router detects when it is not running as root,
808 outputs suitable debugging information if -d is set, and then runs
809 the subprocess without attempting to change uid/gid.
811 PH/15 Minor change to Makefile for building test_host (undocumented testing
814 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
815 additional section of a DNS packet that returns MX or SRV records.
816 Instead, it always explicitly searches for A/AAAA records. This avoids
817 major problems that occur when a DNS server includes only records of one
818 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
819 fixed another bug: if SRV records were looked up and the corresponding
820 address records were *not* found in the additional section, the port
821 values from the SRV records were lost.
823 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
824 using the correct key (the original address) when searching the retry
825 rules in order to find which one to use for generating the retry hint.
827 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
828 adding the default one. Similarly, if it contains a Reply-To: header, the
829 errors_reply_to option, if set, is not used.
831 PH/19 When calculating a retry time, Exim used to measure the "time since
832 failure" by looking at the "first failed" field in the retry record. Now
833 it does not use this if it is later than than the arrival time of the
834 message. Instead it uses the arrival time. This makes for better
835 behaviour in cases where some deliveries succeed, thus re-setting the
836 "first failed" field. An example is a quota failure for a huge message
837 when small messages continue to be delivered. Without this change, the
838 "time since failure" will always be short, possible causing more frequent
839 delivery attempts for the huge message than are intended.
840 [Note: This change was subsequently modified - see PH/04 for 4.62.]
842 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
843 $1, $2, $3) because the numerical variables can be reset during some
844 expansion items (e.g. "match"), thereby losing the authentication data.
846 PH/21 Make -bV show the size of off_t variables so that the test suite can
847 decide whether to run tests for quotas > 2G.
849 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
850 mailbox_size, and mailbox_filecount in the appendfile transport. If a
851 filecount value is greater than 2G or if a quota value is greater than 2G
852 on a system where the size of off_t is not greater than 4, a panic error
855 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
856 never match. The debug and -bh output now contains an explicit error
857 message indicating a malformed IPv4 address or mask.
859 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
860 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
863 PH/25 Do not write to syslog when running in the test harness. The only
864 occasion when this arises is a failure to open the main or panic logs
865 (for which there is an explicit test).
867 PH/26 Added the /no_tell option to "control=freeze".
869 PH/27 If a host name lookup failed very early in a connection, for example, if
870 the IP address matched host_lookup and the reverse lookup yielded a name
871 that did not have a forward lookup, an error message of the form "no IP
872 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
873 could be logged. Now it outputs the IP address instead of "NULL".
875 PH/28 An enabling patch from MH: add new function child_open_exim2() which
876 allows the sender and the authenticated sender to be set when
877 submitting a message from within Exim. Since child_open_exim() is
878 documented for local_scan(), the new function should be too.
880 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
881 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
882 results in an empty string is now treated as unset.
884 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
886 PH/31 Added sender_verify_fail logging option.
888 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
889 needed by RFC 822 but not by RFC 2822 was commented out. I have now
890 tidied the source and removed it altogether.
892 PH/33 When a queue run was abandoned because the load average was too high, a
893 log line was always written; now it is written only if the queue_run log
894 selector is set. In addition, the log line for abandonment now contains
895 information about the queue run such as the pid. This is always present
896 in "start" and "stop" lines but was omitted from the "abandon" line.
898 PH/34 Omit spaces between a header name and the colon in the error message that
899 is given when verify = headers_syntax fails (if there are lots of them,
900 the message gets confusing).
902 PH/35 Change the default for dns_check_names_pattern to allow slashes within
903 names, as there are now some PTR records that contain slashes. This check
904 is only to protect against broken name servers that fall over on strange
905 characters, so the fact that it applies to all lookups doesn't matter.
907 PH/36 Now that the new test suite is complete, we can remove some of the
908 special code in Exim that was needed for the old test suite. For example,
909 sorting DNS records because real resolvers return them in an arbitrary
910 order. The new test suite's fake resolver always returns records in the
913 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
914 bounces) except when queue_only is set, to avoid logging races between
915 the different processes.
917 PH/38 Panic-die if .include specifies a non-absolute path.
919 PH/39 A tweak to the "H" retry rule from its user.
921 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
922 a label. They prevented compilation on older perls.
924 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
925 a warning to be raised on newish perls.
927 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
928 on queue. Changes to match documented behaviour of showing count of
929 messages matching specified criteria.
931 PH/40 Changed the default ident timeout from 30s to 5s.
933 PH/41 Added support for the use of login_cap features, on those BSD systems
934 that have them, for controlling the resources used by pipe deliveries.
936 PH/42 The content-scanning code uses fopen() to create files in which to put
937 message data. Previously it was not paying any attention to the mode of
938 the files. Exim runs with umask(0) because the rest of the code creates
939 files with open(), and sets the required mode explicitly. Thus, these
940 files were ending up world-writeable. This was not a big issue, because,
941 being within the spool directory, they were not world-accessible. I have
942 created a function called modefopen, which takes an additional mode
943 argument. It sets umask(777), creates the file, chmods it to the required
944 mode, then resets the umask. All the relevant calls to fopen() in the
945 content scanning code have been changed to use this function.
947 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
948 to 24 hours. This avoids potential overflow problems when processing G
949 and H retry rules. I suspect nobody ever tinkers with this value.
951 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
953 PH/45 When the plaintext authenticator is running as a client, the server's
954 challenges are checked to ensure they are valid base64 strings. By
955 default, the authentication attempt is cancelled if an invalid string is
956 received. Setting client_ignore_invalid_base64 true ignores these errors.
957 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
958 they are received. Thus, the responses can be made to depend on the
959 challenges. If an invalid string is ignored, an empty string is placed in
962 PH/46 Messages that are created by the autoreply transport now contains a
963 References: header, in accordance with RFCs 2822 and 3834.
965 PH/47 Added authenticated_sender_force to the smtp transport.
967 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
969 PH/49 Installed latest patch from the Sieve maintainer.
971 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
972 was also set, the appendfile transport was unnecessarily scanning a
973 directory of message files (e.g. for maildir delivery) to find the count
974 of files (along with the size), even though it did not need this
975 information. It now does the scan only if it needs to find either the
976 size of the count of files.
978 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
980 PH/52 Two bugs concerned with error handling when the smtp transport is
983 (i) Exim was not creating retry information for temporary errors given
984 for individual recipients after the DATA command when the smtp transport
985 was used in LMTP mode. This meant that they could be retried too
986 frequently, and not timed out correctly.
988 (ii) Exim was setting the flag that allows error details to be returned
989 for LMTP errors on RCPT commands, but not for LMTP errors for individual
990 recipients that were returned after the DATA command.
992 PH/53 This is related to PH/52, but is more general: for any failing address,
993 when detailed error information was permitted to be returned to the
994 sender, but the error was temporary, then after the final timeout, only
995 "retry timeout exceeded" was returned. Now it returns the full error as
996 well as "retry timeout exceeded".
998 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
999 do this, and (what is worse) MTAs that accept it.
1001 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1002 will now be deprecated.
1004 PH/56 New os.c-cygwin from the Cygwin maintainer.
1006 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1007 formats (previously only available in exim formats via -bpr, -bpru,
1008 and -bpra. Now also available in native and exiqgrep formats)
1010 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1011 with very large, slow to parse queues
1013 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1015 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1017 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1018 responses to authentication challenges, though it was showing the
1019 challenges; (ii) I've removed the CR characters from the debug output for
1022 PH/58 Allow for the insertion of a newline as well as a space when a string
1023 is turned into more than one encoded-word during RFC 2047 encoding. The
1024 Sieve code now uses this.
1026 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1027 data_4xx, lost_connection, tls_required.
1029 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1030 message was being sent as an SMTP response.
1032 PH/61 Add -l and -k options to exicyclog.
1034 PH/62 When verifying, if an address was redirected to one new address, so that
1035 verification continued, and the new address failed or deferred after
1036 having set something in $address_data, the value of $address_data was not
1037 passed back to the ACL. This was different to the case when no
1038 redirection occurred. The value is now passed back in both cases.
1040 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1041 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1042 use login_cap.h, so on its own it isn't the distinguishing feature. The
1043 new name refers directly to the setclassresources() function.
1045 PH/65 Added configuration files for NetBSD3.
1047 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1049 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1050 is preferred over IPv4.
1052 PH/68 The bounce_return_message and bounce_return_body options were not being
1053 honoured for bounces generated during the reception of non-SMTP messages.
1054 In particular, this applied to messages rejected by the ACL. This bug has
1055 been fixed. However, if bounce_return_message is true and bounce_return_
1056 body is false, the headers that are returned for a non-SMTP message
1057 include only those that have been read before the error was detected.
1058 (In the case of an ACL rejection, they have all been read.)
1060 PH/69 The HTML version of the specification is now built in a directory called
1061 spec_html instead of spec.html, because the latter looks like a path with
1062 a MIME-type, and this confuses some software.
1064 PH/70 Catch two compiler warnings in sieve.c.
1066 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1067 function verify_get_ident() calls ip_connect() to connect a socket, but
1068 if the "connect()" function timed out, ip_connect() used to close the
1069 socket. However, verify_get_ident() also closes the socket later, and in
1070 between Exim writes to the log, which may get opened at this point. When
1071 the socket was closed in ip_connect(), the log could get the same file
1072 descriptor number as the socket. This naturally causes chaos. The fix is
1073 not to close the socket in ip_connect(); the socket should be closed by
1074 the function that creates it. There was only one place in the code where
1075 this was missing, in the iplookup router, which I don't think anybody now
1076 uses, but I've fixed it anyway.
1078 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1079 well as to direct DNS lookups. Otherwise the handling of names in host
1080 lists is inconsistent and therefore confusing.
1086 PH/01 Two changes to the default runtime configuration:
1088 (1) Move the checks for relay_from_hosts and authenticated clients from
1089 after to before the (commented out) DNS black list checks.
1091 (2) Add control=submission to the relay_from_hosts and authenticated
1092 clients checks, on the grounds that messages accepted by these
1093 statements are most likely to be submissions.
1095 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1097 (1) Generate an error if the third argument for the ${prvs expansion is
1100 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1103 (3) Reset the variables that are obtained from the first argument of
1104 ${prvscheck and used in the second argument before leaving the code,
1105 because their memory is reclaimed, so using them afterwards may do
1108 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1109 one (it's much easier than Tom thought :-).
1111 (5) Because of (4), we can now allow for the use of $prvscheck_result
1112 inside the third argument.
1114 PH/03 For some reason, the default setting of PATH when running a command from
1115 a pipe transport was just "/usr/bin". I have changed it to
1118 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1119 anything to be listed in the output from -bV.
1121 PH/05 When a filter generated an autoreply, the entire To: header line was
1122 quoted in the delivery log line, like this:
1124 => >A.N.Other <ano@some.domain> <original@ddress> ...
1126 This has been changed so that it extracts the operative address. There
1127 may be more than one such address. If so, they are comma-separated, like
1130 => >ano@some.domain,ona@other.domain <original@ddress> ...
1132 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1133 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1134 not being looked up in the rDNS to get a host name, Exim was showing the
1135 IP address twice in Received: lines, even though the IP addresses were
1136 identical. For example:
1138 Received: from [1.2.3.4] (helo=[1.2.3.4])
1140 However, if the real host name was known, it was omitting the HELO data
1141 if it matched the actual IP address. This has been tidied up so that it
1142 doesn't show the same IP address twice.
1144 PH/07 When both +timestamp and +memory debugging was on, the value given by
1145 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1146 called by the debug printing, thereby overwriting the timestamp buffer.
1147 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1149 PH/08 When the original message was included in an autoreply transport, it
1150 always said "this is a copy of the message, including all the headers",
1151 even if body_only or headers_only was set. It now gives an appropriate
1154 PH/09 Applied a patch from the Sieve maintainer which:
1156 o fixes some comments
1157 o adds the (disabled) notify extension core
1158 o adds some debug output for the result of if/elsif tests
1159 o points to the current vacation draft in the documentation
1160 and documents the missing references header update
1164 o fixes a bug in processing the envelope test (when testing
1165 multiple envelope elements, the last element determinted the
1168 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1169 Electronic Mail") by including:
1171 Auto-submitted: auto-generated
1173 in the messages that it generates (bounce messages and others, such as
1174 warnings). In the case of bounce messages for non-SMTP mesages, there was
1175 also a typo: it was using "Auto_submitted" (underscore instead of
1176 hyphen). Since every message generated by Exim is necessarily in response
1177 to another message, thes have all been changed to:
1179 Auto-Submitted: auto-replied
1181 in accordance with these statements in the RFC:
1183 The auto-replied keyword:
1185 - SHOULD be used on messages sent in direct response to another
1186 message by an automatic process,
1188 - MUST NOT be used on manually-generated messages,
1190 - MAY be used on Delivery Status Notifications (DSNs) and Message
1191 Disposition Notifications (MDNs),
1193 - MUST NOT be used on messages generated by automatic or periodic
1194 processes, except for messages which are automatic responses to
1197 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1198 to the default Received: header definition.
1200 PH/12 Added log selector acl_warn_skipped (default on).
1202 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1203 variables because (a) they are in the wrong storage pool and (b) even if
1204 they were copied, it wouldn't work properly because of the caching.
1206 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1207 checking when decoding. Apparently there are clients that generate
1208 overlong encoded strings. Why am I not surprised?
1210 PH/15 If the first argument of "${if match_address" was not empty, but did not
1211 contain an "@" character, Exim crashed. Now it writes a panic log message
1212 and treats the condition as false.
1214 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1216 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1217 extension "envelope-auth". The code is finished and in agreement with
1218 other implementations, but there is no documentation so far and in fact,
1219 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1220 not changing the active code.
1222 Print executed "if" and "elsif" statements when debugging is used. This
1223 helps a great deal to understand what a filter does.
1225 Document more things not specified clearly in RFC3028. I had all this
1226 sorted out, when out of a sudden new issues came to my mind. Oops."
1228 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1231 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1232 canonical form (as documented). However, after a host name lookup from
1233 the IP address, check_host() was doing a simple string comparison with
1234 addresses acquired from the DNS when checking that the found name did
1235 have the original IP as one of its addresses. Since any found IPv6
1236 addresses are likely to be in abbreviated form, the comparison could
1237 fail. Luckily, there already exists a function for doing the comparison
1238 by converting both addresses to binary, so now that is used instead of
1239 the text comparison.
1241 PH/20 There was another similar case to PH/19, when a complete host name was
1242 given in a host list; looking up its IP address could give an abbreviated
1243 form, whereas the current host's name might or might not be abbreviated.
1244 The same fix has been applied.
1250 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1251 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1254 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1255 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1257 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1259 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1260 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1261 Also, for "Auto-Submitted", treat anything other than "no" as
1262 non-personal, in accordance with RFC 3834. (Previously it treated
1263 anything starting "auto-" as non-personal.)
1265 TF/01 The control=submission/name=... option had a problem with syntax
1266 errors if the name included a slash character. The /name= option
1267 now slurps the rest of the string, so it can include any characters
1268 but it must come last in the list of options (after /sender_retain
1271 PH/05 Some modifications to the interface to the fake nameserver for the new
1279 TK/01 Added the "success_on_redirect" address verification option. See
1280 NewStuff for rationale and an example.
1282 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1284 PH/02 Patch to exigrep to allow it to work on syslog lines.
1286 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1287 fread() to skip over the body file's header line, because in Cygwin the
1288 header line is locked and is inaccessible.
1290 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1291 co-exist for some time) to make it clear that it is the Exim ID that is
1292 referenced, not the Message-ID: header line.
1294 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1295 string_format() function, because snprintf() does not exist on all
1298 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1299 ${stat: expansion item.
1301 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1302 protocol synchronization error", to keep the pedants happy.
1304 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1305 well as for IRIX systems, when gcc is being used. See the host.c source
1308 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1310 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1313 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1314 transport and to the smtp transport in LMTP mode.
1316 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1318 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1319 run a filter in a subprocess. This could lead to confusion in subsequent
1320 lookups in the parent process. There should also be a search_tidyup() at
1321 the end of the subprocess.
1323 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1324 only if the host matched helo_try_verify_hosts, which caused the
1325 verification to occur when the EHLO/HELO command was issued. The ACL just
1326 tested the remembered result. Now, if a previous verification attempt has
1327 not happened, "verify = helo" does it there and then.
1329 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1331 TK/03 Fix log output including CR from clamd.
1333 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1334 exist provoked a memory error which could cause a segfault.
1336 PH/15 Installed PCRE 6.2
1338 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1340 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1341 of the problem. Specifically, suggested +O2 rather than +O1 for the
1344 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1346 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1347 record, the port was not correctly logged when the outgoing_port log
1348 selector was set (it logged the transort's default port).
1350 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1351 fallback_hosts, and "hosts" in the smtp transport.
1353 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1354 host errors such as "Connection refused".
1356 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1357 authentication with radiusclient 0.4.9:
1359 - Error returned from rc_read_config was caught wrongly
1360 - Username/password not passed on to radius server due to wrong length.
1362 The presumption is that some radiusclient API changes for 4.51/PH/17
1363 were not taken care of correctly. The code is still untested by me (my
1364 Linux distribution still has 0.3.2 of radiusclient), but it was
1365 contributed by a Radius user.
1367 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1368 expanding the "port" option of the smtp transport.
1370 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1371 while reading a MIME header. Thanks to Tom Hughes for a patch.
1373 PH/24 Include config.h inside local_scan.h so that configuration settings are
1376 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1377 that in an ACL for RCPT (for example), you can examine exactly what was
1380 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1381 commands, but it was not correctly comparing the address with the actual
1382 client host address. Thus, it would show the EHLO address in Received:
1383 header lines when this was not necessary.
1385 PH/27 Added the % operator to ${eval:}.
1387 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1388 it should be ignoring failures (because with -C, for example, it has lost
1389 privilege). It wasn't ignoring creation failures other than "already
1392 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1395 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1396 because an input line was too long, either on its own, or by virtue of
1397 too many continuations, the temporary file was not being removed, and the
1398 return code was incorrect.
1400 PH/31 Missing "BOOL" in function definition in filtertest.c.
1402 PH/32 Applied Sieve patches from the maintainer.
1404 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1406 PH/33 Added "verify = not_blind".
1408 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1409 Local/Makefile (with some defaults set). These are used in built scripts
1410 such as exicyclog, but they have never been used in the exim_install
1411 script (though there are many overriding facilities there). I have
1412 arranged that the exim_install script now takes note of these two
1415 PH/35 Installed configuration files for Dragonfly.
1417 PH/36 When a locally submitted message by a trusted user did not contain a
1418 From: header, and the sender address was obtained from -f or from an SMTP
1419 MAIL command, and the trusted user did not use -F to supply a sender
1420 name, $originator_name was incorrectly used when constructing a From:
1421 header. Furthermore, $originator_name was used for submission mode
1422 messages from external hosts without From: headers in a similar way,
1423 which is clearly wrong.
1425 PH/37 Added control=suppress_local_fixups.
1427 PH/38 When log_selector = +received_sender was set, and the addition of the
1428 sender made the log line's construction buffer exactly full, or one byte
1429 less than full, an overflow happened when the terminating "\n" was
1432 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1433 when the result of a list match is failure because a DNS lookup failed.
1435 PH/40 RM_COMMAND is now used in the building process.
1437 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1438 the "build-* directories that it finds.
1440 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1441 address in a domain literal was a prefix of an interface address.
1443 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1444 when verifying a sender address, unless rewrite_headers is false.
1446 PH/44 Wrote a long comment about why errors_to addresses are verified as
1447 recipients, not senders.
1449 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1450 the ratelimit ACL was added.
1452 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1454 PH/46 Added extra information about PostgreSQL errors to the error string.
1456 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1457 avoiding the need to install special zones in a real server. This is
1458 backwards compatible; if it can't find the fake resolver, it drops back.
1459 Thus, both old and new test suites can be run.
1461 TF/02 Added util/ratelimit.pl
1463 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1464 clock is set back in time.
1466 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1467 Candler <B.Candler@pobox.com>.
1469 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1470 OK for addresses that are the result of redirections.
1472 PH/48 A number of further additions for the benefit of the new test suite,
1473 including a fake gethostbyname() that interfaces to the fake DNS resolver
1476 TF/06 The fix for widen_domains has also been applied to qualify_single and
1477 search_parents which are the other dnslookup options that can cause
1480 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1483 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1485 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1486 service" if it encountered an MX record with an empty target hostname.
1487 The message is now "an MX or SRV record indicated no SMTP service".
1489 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1490 if the DNS of the sending site is misconfigured. This is quite a
1491 common situation. This change restores the behaviour of treating a
1492 helo verification defer as a failure.
1494 PH/51 If self=fail was set on a router, the bounce message did not include the
1495 actual error message.
1501 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1503 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1504 command itself ended in error, the underlying message about the transport
1505 filter timeout was being overwritten with the pipe command error. Now the
1506 underlying error message should be appended to the second error message.
1508 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1510 PH/02 Reduce the amount of output that "make" produces by default. Full output
1511 can still be requested.
1513 PH/03 The warning log line about a condition test deferring for a "warn" verb
1514 was being output only once per connection, rather than after each
1515 occurrence (because it was using the same function as for successful
1516 "warn" verbs). This seems wrong, so I have changed it.
1518 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1519 it should not have, which might have caused a crash in the right
1520 circumstances, but probably never did.
1522 PH/04 Installed a modified version of Tony Finch's patch to make submission
1523 mode fix the return path as well as the Sender: header line, and to
1524 add a /name= option so that you can make the user's friendly name appear
1527 TF/03 Added the control = fakedefer ACL modifier.
1529 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1530 Mark Lowes for thorough testing.
1532 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1534 TK/03 Merged latest SRS patch from Miles Wilton.
1536 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1537 with the definition in sysexits.h (which is #included earlier).
1538 Fortunately, Exim does not actually use EX_OK. The code used to try to
1539 preserve the sysexits.h value, by assumimg that macro definitions were
1540 scanned for macro replacements. I have been disabused of this notion,
1541 so now the code just undefines EX_OK before #including unistd.h.
1543 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1544 in the smtp transport. When a block could not be written in a single
1545 write() function, the timeout was being re-applied to each part-write.
1546 This seems wrong - if the receiver was accepting one byte at a time it
1547 would take for ever. The timeout is now adjusted when this happens. It
1548 doesn't have to be particularly precise.
1550 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1551 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1553 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1554 without a domain if the check to <postmaster@domain> fails.
1556 SC/01 Eximstats: added -xls and the ability to specify output files
1557 (patch written by Frank Heydlauf).
1559 SC/02 Eximstats: use FileHandles for outputing results.
1561 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1563 SC/04 Eximstats: fixed display of large numbers with -nvr option
1565 SC/05 Eximstats: fixed merging of reports with empty tables.
1567 SC/06 Eximstats: added the -include_original_destination flag
1569 SC/07 Eximstats: removed tabs and trailing whitespace.
1571 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1573 TK/06 MBOX spool code: Add real "From " MBOX separator line
1574 so the .eml file is really in mbox format (even though
1575 most programs do not really care). Patch from Alex Miller.
1577 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1578 The latter is generated from $received_to and is only set if the
1579 message has one envelope recipient. SA can use these headers,
1580 obviously out-of-the-box. Patch from Alex Miller.
1582 PH/08 The ${def test on a variable was returning false if the variable's
1583 value was "0", contrary to what the specification has always said!
1584 The result should be true unless the variable is empty.
1586 PH/09 The syntax error of a character other than { following "${if
1587 def:variable_name" (after optional whitespace) was not being diagnosed.
1588 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1589 accidental colon was present, for example, could give incorrect results.
1591 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1592 result is used (not including appendfile, where other changes are about
1595 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1596 This involved changing a lot of size variables from int to off_t. It
1597 should work with maildirs and everything.
1599 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1600 spamd dying while we are connected to it.
1602 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1605 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1606 function for BDB 4.3.
1608 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1610 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1611 the sources for PCRE have been re-arranged and more files are now
1614 PH/15 The code I had for printing potentially long long variables in PH/11
1615 above was not the best (it lost precision). The length of off_t variables
1616 is now inspected at build time, and an appropriate printing format (%ld
1617 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1618 to be "long long int" or "long int". This is needed for the internal
1619 formatting function string_vformat().
1621 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1622 the configuration file to be ":syslog", then the script "guesses" where
1623 the logs files are, rather than using the compiled in default. In our
1624 case the guess is not the same as the compiled default, so the script
1625 suddenly stopped working when I started to use syslog. The patch checks
1626 to see if log_file_path is "". If so, it attempts to read it from exim
1627 with no configuration file to get the compiled in version, before it
1628 falls back to the previous guessing code."
1630 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1631 implementing BATV in an Exim configuration. See NewStuff for the gory
1634 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1635 Makefile that are specific to HP-UX.
1637 PH/18 If the "use_postmaster" option was set for a recipient callout together
1638 with the "random" option, the postmaster address was used as the MAIL
1639 FROM address for the random test, but not for the subsequent recipient
1640 test. It is now used for both.
1642 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1643 patch removes a few documentation additions to RFC 3028, because the
1644 latest draft now contains them. It adds the new en;ascii-case comparator
1645 and a new error check for 8bit text in MIME parts. Comparator and
1646 require names are now matched exactly. I enabled the subaddress
1647 extension, but it is not well tested yet (read: it works for me)."
1649 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1650 rework some of the code of TK/09 above to avoid the hardwired use of
1651 "%lld" and "long long". Replaced the call to snprintf() with a call to
1654 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1655 records point to non-existent hosts", "retry timeout exceeded", and
1656 "retry time not reached for any host after a long failure period".
1658 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1659 experimental DomainKeys support:
1661 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1662 (2) On an error such as an illegally used "control", the wrong name for
1663 the control was given.
1665 These problems did NOT occur unless DomainKeys support was compiled.
1667 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1669 PH/24 Added ${if match_ip condition.
1671 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1672 databases so that it will be absolutely obvious if a crash occurs in the
1673 DB library. This is a regular occurrence (often caused by mis-matched
1676 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1677 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1678 picked up on a user's system that detects such things. There doesn't seem
1679 to be a gcc warning option for this - only an attribute that has to be
1680 put on the function's prototype. It seems that in Fedora Core 4 they have
1681 set this on a number of new functions. No doubt there will be more in due
1684 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1685 specify a transport. However, if an address that was verified by such a
1686 router was the subject of a callout, Exim crashed because it tried to
1687 read the rcpt_include_affixes from the non-existent transport. Now it
1688 just assumes that the setting of that option is false. This bug was
1689 introduced by 4.51/PH/31.
1691 PH/28 Changed -d+all to exclude +memory, because that information is very
1692 rarely of interest, but it makes the output a lot bigger. People tend to
1693 do -d+all out of habit.
1695 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1696 code in os-type was giving problems when libc.so lives in lib64, like on
1699 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1700 aren't the modern standard, and it seems that some systems' include files
1701 don't always have them. Exim was already checking for some of the newer
1702 ones like T_AAAA, and defining it itself. I've added checks for all the
1703 record types that Exim uses.
1705 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1706 not automatically generating a new one, as it is supposed to. This
1707 prevented TLS from working. If the file did exist, but contained invalid
1708 data, a new version was generated, as expected. It was only the case of a
1709 non-existent file that was broken.
1711 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1712 with a change in libdomainkeys > 0.64.
1714 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1715 from DNS. If the selector record carries the flag, it now has
1716 precedence over the domain-wide flag.
1718 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1720 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1721 the use of a port name that isn't defined in /etc/services) occurred, the
1722 message was deferred as in a normal delivery, and thus remained on the
1723 spool, instead of being failed because of the mua_wrapper setting. This
1724 is now fixed, and I tidied up some of the mua_wrapper messages at the
1727 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1728 the messages in a hash of arrays rather than using individual hashes.
1729 This is a bit cleaner and results in dramatic memory savings, albeit
1730 at a slight CPU cost.
1732 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1733 as requested by Marc Sherman.
1735 SC/10 Eximstats: added histograms for user specified patterns as requested
1738 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1740 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1741 fopen() in the content-scanning modules that did not already have it.
1747 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1748 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1750 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1752 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1754 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1755 to test Sieve filters that use "vacation".
1757 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1758 that changes the way the GnuTLS parameters are stored in the cache file.
1759 The new format can be generated externally. For backward compatibility,
1760 if the data in the cache doesn't make sense, Exim assumes it has read an
1761 old-format file, and it generates new data and writes a new file. This
1762 means that you can't go back to an older release without removing the
1765 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1766 work if there are any delivery delays because "one_time" forces the
1767 parent to be marked "delivered", so its unseen clone is never tried
1768 again. For this reason, Exim now forbids the simultaneous setting of
1771 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1772 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1773 entry if you want to know the details. The fix, however, neglected to
1774 consider the case when local delivery batching is involved. The test for
1775 "previously delivered" was not happening when checking to see if an
1776 address could be batched with a previous (undelivered) one; under
1777 certain circumstances this could lead to multiple deliveries to the same
1780 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1781 in its include files, and this causes problems building Exim.
1783 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1784 header_syntax) but Exim was just ignoring anything given after a slash.
1785 In particular, this caused confusion with an attempt to use "verify =
1786 reverse_host_lookup/defer_ok". An error is now given when options are
1787 supplied for verify items that do not have them. (Maybe reverse_host_
1788 lookup should have a defer_ok option, but that's a different point.)
1790 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1791 defined by RFC 821) to 2048, because there were problems with some AUTH
1792 commands, and RFC 1869 says the size should be increased for extended
1793 SMTP commands that take arguments.
1795 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1798 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1799 "unknown" error; now it says that the functionality isn't in the binary.
1801 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1802 an address' error message when a string expansion fails (syntax or
1803 whatever). Otherwise the password may appear in the log. Following change
1804 PH/42 below, there is no longer a chance of it appearing in a bounce
1807 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1809 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1810 of its addresses was ever tried. (Bugzilla bug #2.)
1812 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1813 the result incorrectly in the debug output. (It correctly added a newline
1814 to what was transported.)
1816 TF/01 Added $received_time.
1818 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1819 commented out examples of how to interface to a virus scanner and to
1820 SpamAssassin. Also added commented examples of av_scanner and
1821 spamd_address settings.
1823 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1824 and controls are allowed in which ACLs. There were a couple of minor
1825 errors. Some of the entries in the conditions table (which is a table of
1826 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1827 negation of where the condition IS allowed.
1829 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1831 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1832 header file does not have a version number, so I've had to invent a new
1833 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1834 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1835 radiusclient), but it was contributed by a Radius user.
1837 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1838 files or directories, for OpenSSL.
1840 PH/19 When an Exim process that is running as root has to create an Exim log
1841 file, it does so in a subprocess that runs as exim:exim so as to get the
1842 ownership right at creation (otherwise, other Exim processes might see
1843 the file with the wrong ownership). There was no test for failure of this
1844 fork() call, which would lead to the process getting stuck as it waited
1845 for a non-existent subprocess. Forks do occasionally fail when resources
1846 run out. I reviewed all the other calls to fork(); they all seem to check
1849 PH/20 When checking for unexpected SMTP input at connect time (before writing
1850 the banner), Exim was not dealing correctly with a non-positive return
1851 from the read() function. If the client had disconnected by this time,
1852 the result was a log entry for a synchronization error with an empty
1853 string after "input=" when read() returned zero. If read() returned -1
1854 (an event I could not check), uninitialized data bytes were printed.
1855 There were reports of junk text (parts of files, etc) appearing after
1858 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1860 PH/22 Added support for macro redefinition, and (re)definition in between
1861 driver and ACL definitions.
1863 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1864 forgetting to use the resulting value; it was using the unexpanded value.
1866 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1867 hadn't been configured. The fix is from Juergen Kreileder, who
1868 understands it better than I do:
1870 "Here's what I see happening with three configured cyrus_sasl
1871 authenticators configured (plain, login, cram-md5):
1873 On startup auth_cyrus_sasl_init() gets called for each of these.
1874 This means three calls to sasl_listmech() without a specified mech_list.
1875 => SASL tests which mechs of all available mechs actually work
1876 => three warnings about OTP not working
1877 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1879 With the patch, sasl_listmech() also gets called three times. But now
1880 SASL's mech_list option is set to the server_mech specified in the the
1881 authenticator. Or in other words, the answer from sasl_listmech()
1882 gets limited to just the mech you're testing for (which is different
1884 => the return list contains just 'plain' or 'login', 'cram-md5' or
1885 nothing depending on the value of ob->server_mech.
1887 I've just tested the patch: Authentication still works fine,
1888 unavailable mechs specified in the exim configuration are still
1889 caught, and the auth.log warnings about OTP are gone."
1891 PH/25 When debugging is enabled, the contents of the command line are added
1892 to the debugging output, even when log_selector=+arguments is not
1895 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1896 answer is "GNU", and only if the return is "GNU/something" is the answer
1899 PH/27 $acl_verify_message is now set immediately after the failure of a
1900 verification in an ACL, and so is available in subsequent modifiers. In
1901 particular, the message can be preserved by coding like this:
1903 warn !verify = sender
1904 set acl_m0 = $acl_verify_message
1906 Previously, $acl_verify_message was set only while expanding "message"
1907 and "log_message" when a very denied access.
1909 PH/28 Modified OS/os.c-Linux with
1911 -#ifndef OS_LOAD_AVERAGE
1912 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1914 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1915 nomenclature these days.)
1917 PH/29 Installed patch from the Sieve maintainer that adds the options
1918 sieve_useraddress and sieve_subaddress to the redirect router.
1920 PH/30 In these circumstances:
1921 . Two addresses routed to the same list of hosts;
1922 . First host does not offer TLS;
1923 . First host accepts first address;
1924 . First host gives temporary error to second address;
1925 . Second host offers TLS and a TLS session is established;
1926 . Second host accepts second address.
1927 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1928 and peerdn, if requested) that were in fact used only for the second
1931 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1932 attention to any local part prefix or suffix that was matched by the
1933 router that accepted the address. It now behaves in the same way as it
1934 does for delivery: the affixes are removed from the local part unless
1935 rcpt_include_affixes is set on the transport.
1937 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1938 timeout during the DATA phase of an incoming message.
1940 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1941 applied a patch sanctioned by the Sieve maintainer.
1943 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1944 the uid or gid is negative. A case of a negative gid caused this to be
1945 noticed. The fix allows for either to be negative.
1947 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1948 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1951 PH/36 The change PH/12 above was broken. Fixed it.
1953 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1954 the grounds that routing the same address twice would always produce the
1955 same answer. This might have been true once, but it is certainly no
1956 longer true now. Routing a child address may depend on the previous
1957 routing that produced that child. Some complicated redirection strategies
1958 went wrong when messages had multiple recipients, and made Exim's
1959 behaviour dependent on the order in which the addresses were given.
1961 I have moved the duplicate checking until after the routing is complete.
1962 Exim scans the addresses that are assigned to local and remote
1963 transports, and removes any duplicates. This means that more work will be
1964 done, as duplicates will always all be routed, but duplicates are
1965 presumably rare, so I don't expect this is of any significance.
1967 For deliveries to pipes, files, and autoreplies, the duplicate checking
1968 still happens during the routing process, since they are not going to be
1971 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1972 It corrects a timeout issue with spamd. This is Ian's comment: "The
1973 background is that sometimes spamd either never reads data from a
1974 connection it has accepted, or it never writes response data. The exiscan
1975 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1976 blindly assumes that writes won't block so it may never time out."
1978 PH/39 Allow G after quota size as well as K and M.
1980 PH/40 The value set for $authenticated_id in an authenticator may not contain
1981 binary zeroes or newlines because the value is written to log lines and
1982 to spool files. There was no check on this. Now the value is run through
1983 the string_printing() function so that such characters are converted to
1984 printable escape sequences.
1986 PH/41 $message_linecount is a new variable that contains the total number of
1987 lines in the message. Compare $body_linecount, which is the count for the
1990 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1991 bounce and delay warning messages, except in certain special cases, which
1994 (a) An SMTP error message from a remote host;
1995 (b) A message specified in a :fail: redirection;
1996 (c) A message specified in a "fail" command in a system filter;
1997 (d) A message specified in a FAIL return from the queryprogram router;
1998 (e) A message specified by the cannot_route_message router option.
2000 In these cases only, Exim does include the error details in bounce and
2001 warning messages. There are also a few cases where bland messages such
2002 as "unrouteable address" or "local delivery error" are given.
2004 PH/43 $value is now also set for the "else" part of a ${run expansion.
2006 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2007 being worked on, but at least Exim now implements the latest version to
2010 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2011 process to complete was treated as a delivery failure, a timeout while
2012 writing the message to the pipe was logged, but erroneously treated as a
2013 successful delivery. Such timeouts include transport filter timeouts. For
2014 consistency with the overall process timeout, these timeouts are now
2015 treated as errors, giving rise to delivery failures by default. However,
2016 there is now a new Boolean option for the pipe transport called
2017 timeout_defer, which, if set TRUE, converts the failures into defers for
2018 both kinds of timeout. A transport filter timeout is now identified in
2021 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2022 systems where "make" and "gmake" are different, calling "gmake" at top
2023 level broke things. I've arranged for the value of $(MAKE) to be passed
2024 from the Makefile to this script so that it can call the same version of
2028 A note about Exim versions 4.44 and 4.50
2029 ----------------------------------------
2031 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2032 changes of various kinds. As a consequence, a big documentation update was
2033 needed. This delayed the release for rather longer than seemed good, especially
2034 in the light of a couple of (minor) security issues. Therefore, the changes
2035 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2036 release. So 4.44 and 4.50 are in effect two different branches that both start
2039 I have left the 4.50 change log unchanged; it contains all the changes since
2040 4.43. The change log for 4.44 is below; many of its items are identical to
2041 those for 4.50. This seems to be the most sensible way to preserve the
2042 historical information.
2048 1. Minor wording change to the doc/README.SIEVE file.
2050 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2051 computation of the current number of files was incorrect.
2053 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2054 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2057 4. Give more explanation in the error message when the command for a transport
2058 filter fails to execute.
2060 5. There are several places where Exim runs a non-Exim command in a
2061 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2062 was being done only for the command run by the queryprogram router. It is
2063 now done for all such subprocesses. The other cases are: ${run, transport
2064 filters, and the commands run by the lmtp and pipe transports.
2066 6. Added CONFIGURE_GROUP build-time option.
2068 7. Some older OS have a limit of 256 on the maximum number of file
2069 descriptors. Exim was using setrlimit() to set 1000 as a large value
2070 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2071 systems. I've change it so that if it can't get 1000, it tries for 256.
2073 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2074 was an oversight, and furthermore, ever since the addition of extra
2075 controls (e.g. 4.43/32), the checks on when to allow different forms of
2076 "control" were broken. There should now be diagnostics for all cases when a
2077 control that does not make sense is encountered.
2079 9. Added the /retain_sender option to "control=submission".
2081 10. $recipients is now available in the predata ACL (oversight).
2083 11. Tidy the search cache before the fork to do a delivery from a message
2084 received from the command line. Otherwise the child will trigger a lookup
2085 failure and thereby defer the delivery if it tries to use (for example) a
2086 cached ldap connection that the parent has called unbind on.
2088 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2089 of $address_data from the recipient verification was clobbered by the
2090 sender verification.
2092 13. The value of address_data from a sender verification is now available in
2093 $sender_address_data in subsequent conditions in the ACL statement.
2095 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2097 15. Added a new option "connect=<time>" to callout options, to set a different
2100 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2101 was its contents. (It was OK if the option was not defined at all.)
2103 17. A "Completed" log line is now written for messages that are removed from
2104 the spool by the -Mrm option.
2106 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2107 information about exactly what failed.
2109 19. Added -dd to debug only the daemon process.
2111 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2112 handles timeouts, both on the server side and network timeouts. Renamed the
2113 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2115 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2116 It is now set to "smtps".
2118 22. $host_address is now set to the target address during the checking of
2119 ignore_target_hosts.
2121 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2122 being passed; this would have caused $sender_host_name to have been used if
2123 matching the list had actually called for a host name (not very likely,
2124 since this list is usually IP addresses). A host name is now passed as
2127 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2128 code that specifies a non-restarting handler (typically sigaction() in
2129 modern systems) in an attempt to fix a rare and obscure crash bug.
2131 25. Narrowed the window for a race in the daemon that could cause it to ignore
2132 SIGCHLD signals. This is not a major problem, because they are used only to
2133 wake it up if nothing else does.
2135 26. A malformed maildirsize file could cause Exim to calculate negative values
2136 for the mailbox size or file count. Odd effects could occur as a result.
2137 The maildirsize information is now recalculated if the size or filecount
2140 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2141 support for a long time. Removed HAVE_SYS_VFS_H.
2143 28. Installed the latest version of exipick from John Jetmore.
2145 29. In an address list, if the pattern was not a regular expression, an empty
2146 subject address (from a bounce message) matched only if the pattern was an
2147 empty string. Non-empty patterns were not even tested. This was the wrong
2148 because it is perfectly reasonable to use an empty address as part of a
2149 database query. An empty address is now tested by patterns that are
2150 lookups. However, all the other forms of pattern expect the subject to
2151 contain a local part and a domain, and therefore, for them, an empty
2152 address still always fails if the pattern is not itself empty.
2154 30. Exim went into a mad DNS loop when attempting to do a callout where the
2155 host was specified on an smtp transport, and looking it up yielded more
2156 than one IP address.
2158 31. Re-factored the code for checking spool and log partition space into a
2159 function that finds that data and another that does the check. The former
2160 is then used to implement four new variables: $spool_space, $log_space,
2161 $spool_inodes, and $log_inodes.
2163 32. The RFC2047 encoding function was originally intended for short strings
2164 such as real names; it was not keeping to the 75-character limit for
2165 encoded words that the RFC imposes. It now respects the limit, and
2166 generates multiple encoded words if necessary. To be on the safe side, I
2167 have increased the buffer size for the ${rfc2047: expansion operator from
2170 33. It is now permitted to omit both strings after an "if" condition; if the
2171 condition is true, the result is "true". As before, when the second string
2172 is omitted, a false condition yields an empty string. This makes it less
2173 cumbersome to write custom ACL and router conditions.
2175 34. Failure to deliver a bounce message always caused it to be frozen, even if
2176 there was an errors_to setting on the router. The errors_to setting is now
2179 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2180 canonical form (fully expanded) before being placed in
2181 $sender_host_address.
2183 36. The table in the code that translates DNS record types into text (T_A to
2184 "A" for instance) was missing entries for NS and CNAME. It is just possible
2185 that this could have caused confusion if both these types were looked up
2186 for the same domain, because the text type is used as part of Exim's
2187 per-process caching. But the chance of anyone hitting this buglet seems
2190 37. The dnsdb lookup has been extended in a number of ways.
2192 (1) There is a new type, "zns", which walks up the domain tree until it
2193 finds some nameserver records. It should be used with care.
2195 (2) There is a new type, "mxh", which is like "mx" except that it returns
2196 just the host names, not the priorities.
2198 (3) It is now possible to give a list of domains (or IP addresses) to be
2199 looked up. The behaviour when one of the lookups defers can be
2200 controlled by a keyword.
2202 (4) It is now possible to specify the separator character for use when
2203 multiple records are returned.
2205 38. The dnslists ACL condition has been extended: it is now possible to supply
2206 a list of IP addresses and/or domains to be looked up in a particular DNS
2209 39. Added log_selector=+queue_time_overall.
2211 40. When running the queue in the test harness, wait just a tad after forking a
2212 delivery process, to get repeatability of debugging output.
2214 41. Include certificate and key file names in error message when GnuTLS fails
2215 to set them up, because the GnuTLS error message doesn't include the name
2216 of the failing file when there is a problem reading it.
2218 42. Allow both -bf and -bF in the same test run.
2220 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2222 44. The "Exiscan patch" is now merged into the mainline Exim source.
2224 45. Sometimes the final signoff response after QUIT could fail to get
2225 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2226 < 0 before doing a fflush(). This bug looks as though it goes back to the
2227 introduction of TLS in release 3.20, but "sometimes" must have been rare
2228 because the tests only now provoked it.
2230 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2231 (this can affect the format of dates).
2233 47. exim_tidydb, when checking for the continued existence of a message for
2234 which it has found a message-specific retry record, was not finding
2235 messages that were in split spool directories. Consequently, it was
2236 deleting retry records that should have stayed in existence.
2238 48. Steve fixed some bugs in eximstats.
2240 49. The SPA authentication driver was not abandoning authentication and moving
2241 on to the next authenticator when an expansion was forced to fail,
2242 contradicting the general specification for all authenticators. Instead it
2243 was generating a temporary error. It now behaves as specified.
2245 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2246 (the order specifies the preference for clients). The order is now AES256,
2247 AES128, 3DES, ARCFOUR128.
2249 51. Small patch to Sieve code - explicitly set From: when generating an
2252 52. Exim crashed if a remote delivery caused a very long error message to be
2253 recorded - for instance if somebody sent an entire SpamAssassin report back
2254 as a large number of 550 error lines. This bug was coincidentally fixed by
2255 increasing the size of one of Exim's internal buffers (big_buffer) that
2256 happened as part of the Exiscan merge. However, to be on the safe side, I
2257 have made the code more robust (and fixed the comments that describe what
2260 53. Now that there can be additional text after "Completed" in log lines (if
2261 the queue_time_overall log selector is set), a one-byte patch to exigrep
2262 was needed to allow it to recognize "Completed" as not the last thing in
2265 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2266 patch that reportedly fixes this has been added. I am not expert enough to
2267 create a test for it. This is what the patch creator wrote:
2269 "I found a little strange behaviour of ldap code when working with
2270 Windows 2003 AD Domain, where users was placed in more than one
2271 Organization Units. When I tried to give exim partial DN, the exit code
2272 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2273 But simultaneously result of request was absolutely normal ldap result,
2274 so I produce this patch..."
2276 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2277 REFERENCE, so I have modified the code to exclude the patch when that macro
2280 55. Some experimental protocols are using DNS PTR records for new purposes. The
2281 keys for these records are domain names, not reversed IP addresses. The
2282 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2283 leaves it alone. Component reversal etc. now happens only for IP addresses.
2286 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2288 57. Double the size of the debug message buffer (to 2048) so that more of very
2289 long debug lines gets shown.
2291 58. The exicyclog utility now does better if the number of log files to keep
2292 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2294 59. Two changes related to the smtp_active_hostname option:
2296 (1) $smtp_active_hostname is now available as a variable.
2297 (2) The default for smtp_banner uses $smtp_active_hostname instead
2298 of $primary_hostname.
2300 60. The host_aton() function is supposed to be passed a string that is known
2301 to be a valid IP address. However, in the case of IPv6 addresses, it was
2302 not checking this. This is a hostage to fortune. Exim now panics and dies
2303 if the condition is not met. A case was found where this could be provoked
2304 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2305 components; fortuitously, this particular loophole had already been fixed
2306 by change 4.50/55 above.
2308 If there are any other similar loopholes, the new check in host_aton()
2309 itself should stop them being exploited. The report I received stated that
2310 data on the command line could provoke the exploit when Exim was running as
2311 exim, but did not say which command line option was involved. All I could
2312 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2313 running as the user.
2316 61. There was a buffer overflow vulnerability in the SPA authentication code
2317 (which came originally from the Samba project). I have added a test to the
2318 spa_base64_to_bits() function which I hope fixes it.
2321 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2322 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2324 63. The daemon start-up calls getloadavg() while still root for those OS that
2325 need the first call to be done as root, but it missed one case: when
2326 deliver_queue_load_max is set with deliver_drop_privilege. This is
2327 necessary for the benefit of the queue runner, because there is no re-exec
2328 when deliver_drop_privilege is set.
2330 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2331 This has been fixed.
2333 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2334 was in use, was not putting the data itself into the right store pool;
2335 consequently, it could be overwritten for a subsequent message in the same
2336 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2339 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2341 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2342 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2343 the function were treating the return as a boolean value, which happened to
2344 work because 0=false and not-0=true, but is not correct code.
2346 68. The host_aton() function was not handling scoped IPv6 addresses (those
2347 with, for example, "%eth0" on the end) correctly.
2349 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2350 negated items (that is, ~something) in unsigned ints. Some compilers
2351 apparently mutter when there is no cast.
2353 70. If an address verification called from an ACL failed, and did not produce a
2354 user-specific message (i.e. there was only a "system" message), nothing was
2355 put in $acl_verify_message. In this situation, it now puts the system
2358 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2359 session; change 4.31/43 added the unwanted input to the log line - except
2360 that it did not do this in the start of session case. It now does.
2362 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2363 This is wrong and can cause the other end to generate a synchronization
2364 error if it is another Exim or anything else that does the synchronization
2365 check. A QUIT command is no longer sent after a timeout.
2367 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2368 during host lookups.
2370 74. The defer_ok option of callout verification was not working if it was used
2371 when verifying addresses in header lines, that is, for this case:
2373 verify = header_sender/callout=defer_ok
2375 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2376 those file descriptors could be used for SMTP connections. If anything
2377 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2378 could be sent to the SMTP client, causing chaos. The daemon now opens
2379 stdin, stdout, and stderr to /dev/null when it puts itself into the
2382 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2383 log by default. The user can override this with suitable Perl magic.
2385 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2386 the log message when discard triggers, was not working for the DATA ACL or
2387 for the non-SMTP ACL.
2389 78. Error message wording change in sieve.c.
2391 79. If smtp_accept_max_per_host was set, the number of connections could be
2392 restricted to fewer than expected, because the daemon was trying to set up
2393 a new connection before checking whether the processes handling previous
2394 connections had finished. The check for completed processes is now done
2395 earlier. On busy systems, this bug wouldn't be noticed because something
2396 else would have woken the daemon, and it would have reaped the completed
2399 80. If a message was submitted locally by a user whose login name contained one
2400 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2401 It caused a spool format error. I have fixed the spool reading code. A
2402 related problem was that the "from" clause in the Received: line became
2403 illegal because of the space(s). It is now covered by ${quote_local_part.
2405 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2408 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2410 83. Patch from Sieve maintainer to fix unterminated string problem in
2411 "vacation" handling.
2413 84. Some minor changes to the Linux configuration files to help with other
2414 OS variants using glibc.
2416 85. One more patch for Sieve to update vacation handling to latest spec.
2419 ----------------------------------------------------
2420 See the note above about the 4.44 and 4.50 releases.
2421 ----------------------------------------------------
2427 1. Change 4.43/35 introduced a bug that caused file counts to be
2428 incorrectly computed when quota_filecount was set in an appendfile
2431 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2432 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2435 3. Give more explanation in the error message when the command for a transport
2436 filter fails to execute.
2438 4. There are several places where Exim runs a non-Exim command in a
2439 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2440 was being done only for the command run by the queryprogram router. It is
2441 now done for all such subprocesses. The other cases are: ${run, transport
2442 filters, and the commands run by the lmtp and pipe transports.
2444 5. Some older OS have a limit of 256 on the maximum number of file
2445 descriptors. Exim was using setrlimit() to set 1000 as a large value
2446 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2447 systems. I've change it so that if it can't get 1000, it tries for 256.
2449 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2450 was an oversight, and furthermore, ever since the addition of extra
2451 controls (e.g. 4.43/32), the checks on when to allow different forms of
2452 "control" were broken. There should now be diagnostics for all cases when a
2453 control that does not make sense is encountered.
2455 7. $recipients is now available in the predata ACL (oversight).
2457 8. Tidy the search cache before the fork to do a delivery from a message
2458 received from the command line. Otherwise the child will trigger a lookup
2459 failure and thereby defer the delivery if it tries to use (for example) a
2460 cached ldap connection that the parent has called unbind on.
2462 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2463 of $address_data from the recipient verification was clobbered by the
2464 sender verification.
2466 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2467 was its contents. (It was OK if the option was not defined at all.)
2469 11. A "Completed" log line is now written for messages that are removed from
2470 the spool by the -Mrm option.
2472 12. $host_address is now set to the target address during the checking of
2473 ignore_target_hosts.
2475 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2476 being passed; this would have caused $sender_host_name to have been used if
2477 matching the list had actually called for a host name (not very likely,
2478 since this list is usually IP addresses). A host name is now passed as
2481 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2482 code that specifies a non-restarting handler (typically sigaction() in
2483 modern systems) in an attempt to fix a rare and obscure crash bug.
2485 15. Narrowed the window for a race in the daemon that could cause it to ignore
2486 SIGCHLD signals. This is not a major problem, because they are used only to
2487 wake it up if nothing else does.
2489 16. A malformed maildirsize file could cause Exim to calculate negative values
2490 for the mailbox size or file count. Odd effects could occur as a result.
2491 The maildirsize information is now recalculated if the size or filecount
2494 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2495 support for a long time. Removed HAVE_SYS_VFS_H.
2497 18. Updated exipick to current release from John Jetmore.
2499 19. Allow an empty sender to be matched against a lookup in an address list.
2500 Previously the only cases considered were a regular expression, or an
2503 20. Exim went into a mad DNS lookup loop when doing a callout where the
2504 host was specified on the transport, if the DNS lookup yielded more than
2507 21. The RFC2047 encoding function was originally intended for short strings
2508 such as real names; it was not keeping to the 75-character limit for
2509 encoded words that the RFC imposes. It now respects the limit, and
2510 generates multiple encoded words if necessary. To be on the safe side, I
2511 have increased the buffer size for the ${rfc2047: expansion operator from
2514 22. Failure to deliver a bounce message always caused it to be frozen, even if
2515 there was an errors_to setting on the router. The errors_to setting is now
2518 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2519 canonical form (fully expanded) before being placed in
2520 $sender_host_address.
2522 24. Updated eximstats to version 1.33
2524 25. Include certificate and key file names in error message when GnuTLS fails
2525 to set them up, because the GnuTLS error message doesn't include the name
2526 of the failing file when there is a problem reading it.
2528 26. Expand error message when OpenSSL has problems setting up cert/key files.
2531 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2532 (this can affect the format of dates).
2534 28. exim_tidydb, when checking for the continued existence of a message for
2535 which it has found a message-specific retry record, was not finding
2536 messages that were in split spool directories. Consequently, it was
2537 deleting retry records that should have stayed in existence.
2539 29. eximstats updated to version 1.35
2540 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2541 1.35 - bugfix such that pie charts by volume are generated correctly
2543 30. The SPA authentication driver was not abandoning authentication and moving
2544 on to the next authenticator when an expansion was forced to fail,
2545 contradicting the general specification for all authenticators. Instead it
2546 was generating a temporary error. It now behaves as specified.
2548 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2549 (the order specifies the preference for clients). The order is now AES256,
2550 AES128, 3DES, ARCFOUR128.
2552 31. Small patch to Sieve code - explicitly set From: when generating an
2555 32. Exim crashed if a remote delivery caused a very long error message to be
2556 recorded - for instance if somebody sent an entire SpamAssassin report back
2557 as a large number of 550 error lines. This bug was coincidentally fixed by
2558 increasing the size of one of Exim's internal buffers (big_buffer) that
2559 happened as part of the Exiscan merge. However, to be on the safe side, I
2560 have made the code more robust (and fixed the comments that describe what
2563 33. Some experimental protocols are using DNS PTR records for new purposes. The
2564 keys for these records are domain names, not reversed IP addresses. The
2565 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2566 leaves it alone. Component reversal etc. now happens only for IP addresses.
2569 34. The host_aton() function is supposed to be passed a string that is known
2570 to be a valid IP address. However, in the case of IPv6 addresses, it was
2571 not checking this. This is a hostage to fortune. Exim now panics and dies
2572 if the condition is not met. A case was found where this could be provoked
2573 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2574 components; fortuitously, this particular loophole had already been fixed
2575 by change 4.50/55 or 4.44/33 above.
2577 If there are any other similar loopholes, the new check in host_aton()
2578 itself should stop them being exploited. The report I received stated that
2579 data on the command line could provoke the exploit when Exim was running as
2580 exim, but did not say which command line option was involved. All I could
2581 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2582 running as the user.
2585 35. There was a buffer overflow vulnerability in the SPA authentication code
2586 (which came originally from the Samba project). I have added a test to the
2587 spa_base64_to_bits() function which I hope fixes it.
2590 36. The daemon start-up calls getloadavg() while still root for those OS that
2591 need the first call to be done as root, but it missed one case: when
2592 deliver_queue_load_max is set with deliver_drop_privilege. This is
2593 necessary for the benefit of the queue runner, because there is no re-exec
2594 when deliver_drop_privilege is set.
2596 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2597 was in use, was not putting the data itself into the right store pool;
2598 consequently, it could be overwritten for a subsequent message in the same
2599 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2602 38. Sometimes the final signoff response after QUIT could fail to get
2603 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2604 < 0 before doing a fflush(). This bug looks as though it goes back to the
2605 introduction of TLS in release 3.20, but "sometimes" must have been rare
2606 because the tests only now provoked it.
2612 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2613 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2614 Now it returns an integer. A number of calls were still expecting a T/F
2615 return. Fortuitously, in all cases, the tests worked in OK situations,
2616 which is the norm. However, things would have gone wrong on any write
2617 failures on the smtp file descriptor. This function is used when sending
2618 messages over SMTP and also when doing verify callouts.
2620 2. When Exim is called to do synchronous delivery of a locally submitted
2621 message (the -odf or -odi options), it no longer closes stderr before doing
2624 3. Implemented the mua_wrapper option.
2626 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2628 5. Implemented the functions header_remove(), header_testname(),
2629 header_add_at_position(), and receive_remove_recipient(), and exported them
2632 6. If an ACL "warn" statement specified the addition of headers, Exim already
2633 inserted X-ACL-Warn: at the start if there was no header name. However, it
2634 was not making this test for the second and subsequent header lines if
2635 there were newlines in the string. This meant that an invalid header could
2636 be inserted if Exim was badly configured.
2638 7. Allow an ACL "warn" statement to add header lines at the start or after all
2639 the Received: headers, as well as at the end.
2641 8. Added the rcpt_4xx retry error code.
2643 9. Added postmaster_mailfrom=xxx to callout verification option.
2645 10. Added mailfrom=xxxx to the callout verification option, for verify=
2648 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2649 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2651 12. Inserted some casts to stop certain compilers warning when using pointer
2652 differences as field lengths or precisions in printf-type calls (mostly
2653 affecting debugging statements).
2655 13. Added optional readline() support for -be (dynamically loaded).
2657 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2658 same clock tick as a message's arrival, so that its received time was the
2659 same as the "first fail" time on the retry record, and that message
2660 remained on the queue past the ultimate address timeout, every queue runner
2661 would try a delivery (because it was past the ultimate address timeout) but
2662 after another failure, the ultimate address timeout, which should have then
2663 bounced the address, did not kick in. This was a "< instead of <=" error;
2664 in most cases the first failure would have been in the next clock tick
2665 after the received time, and all would be well.
2667 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2668 being recognized when the domain list was tested by the match_domain
2669 condition in an expansion string.
2671 16. Added the ${str2b64: operator.
2673 17. Exim was always calling setrlimit() to set a large limit for the number of
2674 processes, without checking whether the existing limit was already
2675 adequate. (It did check for the limit on file descriptors.) Furthermore,
2676 errors from getrlimit() and setrlimit() were being ignored. Now they are
2677 logged to the main and panic logs, but Exim does carry on, to try to do its
2678 job under whatever limits there are.
2680 18. Imported PCRE 5.0.
2682 19. Trivial typo in log message " temporarily refused connection" (the leading
2685 20. If the log selector return_path_on_delivery was set and an address was
2686 redirected to /dev/null, the delivery process crashed because it assumed
2687 that a return path would always be set for a "successful" delivery. In this
2688 case, the whole delivery is bypassed as an optimization, and therefore no
2691 21. Internal re-arrangement: the function for sending a challenge and reading
2692 a response while authentication was assuming a zero-terminated challenge
2693 string. It's now changed to take a pointer and a length, to allow for
2694 binary data in such strings.
2696 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2698 23. Exim was not respecting finduser_retries when seeking the login of the
2699 uid under which it was called; it was always trying 10 times. (The default
2700 setting of finduser_retries is zero.) Also, it was sleeping after the final
2701 failure, which is pointless.
2703 24. Implemented tls_on_connect_ports.
2705 25. Implemented acl_smtp_predata.
2707 26. If the domain in control=submission is set empty, Exim assumes that the
2708 authenticated id is a complete email address when it generates From: or
2709 Sender: header lines.
2711 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2712 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2713 chown and chgrp in /bin and hostname in /usr/bin.
2715 28. Exim was keeping the "process log" file open after each use, just as it
2716 does for the main log. This opens the possibility of it remaining open for
2717 long periods when the USR1 signal hits a daemon. Occasional processlog
2718 errors were reported, that could have been caused by this. Anyway, it seems
2719 much more sensible not to leave this file open at all, so that is what now
2722 29. The long-running daemon process does not normally write to the log once it
2723 has entered its main loop, and it closes the log before doing so. This is
2724 so that log files can straightforwardly be renamed and moved. However,
2725 there are a couple of unusual error situations where the daemon does write
2726 log entries, and I had neglected to close the log afterwards.
2728 30. The text of an SMTP error response that was received during a remote
2729 delivery was being truncated at 512 bytes. This is too short for some of
2730 the long messages that one sometimes sees. I've increased the limit to
2733 31. It is now possible to make retry rules that apply only when a message has a
2734 specific sender, in particular, an empty sender.
2736 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2737 it possible to be selective about when SMTP synchronization is enforced.
2739 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2741 32. Implemented hosts_connection_nolog.
2743 33. Added an ACL for QUIT.
2745 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2748 35. Added mailbox_size and mailbox_filecount to appendfile.
2750 36. Added control = no_multiline_responses to ACLs.
2752 37. There was a bug in the logic of the code that waits for the clock to tick
2753 in the case where the clock went backwards by a substantial amount such
2754 that the microsecond fraction of "now" was more than the microsecond
2755 fraction of "then" (but the whole seconds number was less).
2757 38. Added support for the libradius Radius client library this is found on
2758 FreeBSD (previously only the radiusclient library was supported).
2764 1. When certain lookups returned multiple values in the form name=value, the
2765 quoting of the values was not always being done properly. Specifically:
2766 (a) If the value started with a double quote, but contained no whitespace,
2768 (b) If the value contained whitespace other than a space character (i.e.
2769 tabs or newlines or carriage returns) it was not quoted.
2770 This fix has been applied to the mysql and pgsql lookups by writing a
2771 separate quoting function and calling it from the lookup code. The fix
2772 should probably also be applied to nisplus, ibase and oracle lookups, but
2773 since I cannot test any of those, I have not disturbed their existing code.
2775 2. A hit in the callout cache for a specific address caused a log line with no
2776 reason for rejecting RCPT. Now it says "Previous (cached) callout
2777 verification failure".
2779 3. There was an off-by-one bug in the queryprogram router. An over-long
2780 return line was truncated at 256 instead of 255 characters, thereby
2781 overflowing its buffer with the terminating zero. As well as fixing this, I
2782 have increased the buffer size to 1024 (and made a note to document this).
2784 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2785 when Exim is waiting for an SMTP response from a remote server, Exim
2786 restarts its select() call on the socket, thereby resetting its timeout.
2787 This is not a problem when such interrupts are rare. Somebody set up a cron
2788 job to run exiwhat every 2 minutes, which is less than the normal select()
2789 timeout (5 or 10 minutes). This meant that the select() timeout never
2790 kicked in because it was always reset. I have fixed this by comparing the
2791 time when an interrupt arrives with the time at the start of the first call
2792 to select(). If more time than the timeout has elapsed, the interrupt is
2793 treated as a timeout.
2795 5. Some internal re-factoring in preparation for the addition of Sieve
2796 extensions (by MH). In particular, the "personal" test is moved to a
2797 separate function, and given an option for scanning Cc: and Bcc: (which is
2798 not set for Exim filters).
2800 6. When Exim created an email address using the login of the caller as the
2801 local part (e.g. when creating a From: or Sender: header line), it was not
2802 quoting the local part when it contained special characters such as @.
2804 7. Installed new OpenBSD configuration files.
2806 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2807 try to make them clearer.
2809 9. Callout options, other than the timeout value, were being ignored when
2810 verifying sender addresses in header lines. For example, when using
2812 verify = header_sender/callout=no_cache
2814 the cache was (incorrectly) being used.
2816 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2817 only the Cygwin environment.
2819 11. When return_path_on_delivery was set as a log selector, if different remote
2820 addresses in the same message used different return paths and parallel
2821 remote delivery occurred, the wrong values would sometimes be logged.
2822 (Whenever a remote delivery process finished, the return path value from
2823 the most recently started remote delivery process was logged.)
2825 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2826 lines when AUTH and/or TLS are in use. This is the "received protocol"
2827 field. Exim used to use "asmtp" for authenticated SMTP, without any
2828 indication (in the protocol name) for TLS use. Now it follows the RFC and
2829 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2830 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2831 names appear in log lines as well as in Received: header lines.
2833 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2834 extensions, and comparison tests, and to fix some bugs.
2836 14. Changes to the "personal" filter test:
2838 (1) The test was buggy in that it was just doing the equivalent of
2839 "contains" tests on header lines. For example, if a user's address was
2840 anne@some.where, the "personal" test would incorrectly be true for
2842 To: susanne@some.where
2844 This test is now done by extracting each address from the header in turn,
2845 and checking the entire address. Other tests that are part of "personal"
2846 are now done using regular expressions (for example, to check local parts
2847 of addresses in From: header lines).
2849 (2) The list of non-personal local parts in From: addresses has been
2850 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2851 taken from the Sieve specification recommendations.
2853 (3) If the message contains any header line starting with "List-" it is
2854 treated as non-personal.
2856 (4) The test for "circular" in the Subject: header line has been removed
2857 because it now seems ill-conceived.
2859 15. Minor typos in src/EDITME comments corrected.
2861 16. Installed latest exipick from John Jetmore.
2863 17. If headers_add on a router specified a text string that was too long for
2864 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2865 of string_sprintf() is now avoided.
2867 18. $message_body_size was not set (it was always zero) when running the DATA
2868 ACL and the local_scan() function.
2870 19. For the "mail" command in an Exim filter, no default was being set for
2871 the once_repeat time, causing a random time value to be used if "once" was
2872 specified. (If the value happened to be <= 0, no repeat happened.) The
2873 default is now 0s, meaning "never repeat". The "vacation" command was OK
2874 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2875 (I found it when inspecting the code).
2877 20. There is now an overall timeout for performing a callout verification. It
2878 defaults to 4 times the callout timeout, which applies to individual SMTP
2879 commands during the callout. The overall timeout applies when there is more
2880 than one host that can be tried. The timeout is checked before trying the
2881 next host. This prevents very long delays if there are a large number of
2882 hosts and all are timing out (e.g. when the network connections are timing
2883 out). The value of the overall timeout can be changed by specifying an
2884 additional sub-option for "callout", called "maxwait". For example:
2886 verify = sender/callout=5s,maxwait=20s
2888 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2889 to the end before writing, but this should make it even safer).
2891 22. Exim was forgetting that it had advertised PIPELINING for the second and
2892 subsequent messages on an SMTP connection. It was also not resetting its
2893 memory on STARTTLS and an internal HELO.
2895 23. When Exim logs an SMTP synchronization error within a session, it now
2896 records whether PIPELINING has been advertised or not.
2898 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2899 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2900 rather than long int.
2902 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2904 26. Added the never_mail option to autoreply.
2910 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2911 crash if the getsockname() call failed; this can happen if a connection is
2912 closed very soon after it is established. The problem was simply in the
2913 order in which certain operations were done, causing Exim to try to write
2914 to the SMTP stream before it had set up the file descriptor. The bug has
2915 been fixed by making things happen in the correct order.
2921 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2922 before the connection was closed, thus losing the rejection response.
2924 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2925 some early Solaris releases, but causes trouble in current releases where
2926 socklen_t is defined.
2928 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2931 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2934 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2935 that releasing the top of it at the end releases what was used for sub-
2936 expansions (unless the block got too big). However, discard this block if
2937 the first thing is a variable or header, so that we can use its block when
2938 it is dynamic (useful for very large $message_headers, for example).
2940 6. Lookups now cache *every* query, not just the most recent. A new, separate
2941 store pool is used for this. It can be recovered when all lookup caches are
2942 flushed. Lookups now release memory at the end of their result strings.
2943 This has involved some general refactoring of the lookup sources.
2945 7. Some code has been added to the store_xxx() functions to reduce the amount
2946 of flapping under certain conditions.
2948 8. log_incoming_interface used to affect only the <= reception log lines. Now
2949 it causes the local interface and port to be added to several more SMTP log
2950 lines, for example "SMTP connection from", and rejection lines.
2952 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2954 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2956 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2957 could be overwritten at the end of the current message (or the start of a
2958 new message if it was set in a HELO ACL). The value is now preserved for
2959 the duration of the SMTP connection.
2961 12. If a transport had a headers_rewrite setting, and a matching header line
2962 contained an unqualified address, that address was qualified, even if it
2963 did not match any rewriting rules. The underlying bug was that the values
2964 of the flags that permit the existence of unqualified sender and recipient
2965 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2966 non-local messages, and by -bnq for local messages) were not being
2967 preserved with the message after it was received.
2969 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2970 "next input=" as part of the text comprising the host identity instead of
2971 the correct text. The code was using the same buffer for two different
2972 strings. However, depending on which order the printing function evaluated
2973 its arguments, the bug did not always show up. Under Linux, for example, my
2974 test suite worked just fine.
2976 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2977 doesn't work with some older versions of Perl. It has been changed to "my",
2978 which in any case is probably the better facility to use.
2980 15. A really picky compiler found some instances of statements for creating
2981 error messages that either had too many or two few arguments for the format
2984 16. The size of the buffer for calls to the DNS resolver has been increased
2985 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2986 for addresses that have a lot of PTR records. This alleviates a problem; it
2987 does not fully solve it.
2989 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2990 buffer now truncates the list and logs the incident, which is the same
2991 action as happens when Exim is looking up a host name and its aliases.
2992 Previously in this situation something unpredictable would happen;
2993 sometimes it was "internal error: store_reset failed".
2995 18. If a server dropped the connection unexpectedly when an Exim client was
2996 using GnuTLS and trying to read a response, the client delivery process
2997 crashed while trying to generate an error log message.
2999 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3000 string, for example:
3002 warn message = H1: something\nH2: something
3004 the text was added as a single header line from Exim's point of view
3005 though it ended up OK in the delivered message. However, searching for the
3006 second and subsequent header lines using $h_h2: did not work. This has been
3007 fixed. Similarly, if a system filter added multiple headers in this way,
3008 the routers could not see them.
3010 20. Expanded the error message when iplsearch is called with an invalid key to
3011 suggest using net-iplsearch in a host list.
3013 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3014 ACLs are no longer actually imposed (and a message to that effect is
3017 22. If a "gecos" field in a passwd entry contained escaped characters, in
3018 particular, if it contained a \" sequence, Exim got it wrong when building
3019 a From: or a Sender: header from that name. A second bug also caused
3020 incorrect handling when an unquoted " was present following a character
3021 that needed quoting.
3023 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3024 was not being matched caselessly.
3026 24. Arranged for all hyphens in the exim.8 source to be escaped with
3029 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3030 properly. Recipient callout cache records were still being keyed to include
3031 the sender, even when use_sender was set false. This led to far more
3032 callouts that were necessary. The sender is no longer included in the key
3033 when use_sender is false.
3035 26. Added "control = submission" modifier to ACLs.
3037 27. Added the ${base62d: operator to decode base 62 numbers.
3039 28. dnsdb lookups can now access SRV records.
3041 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3042 the configuration file.
3044 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3045 (-v) mode. This makes the output for a verbose queue run more intelligible.
3047 31. Added a use_postmaster feature to recipient callouts.
3049 32. Added the $body_zerocount variable, containing the number of binary zero
3050 bytes in the message body.
3052 33. The time of last modification of the "new" subdirectory is now used as the
3053 "mailbox time last read" when there is a quota error for a maildir
3056 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3058 35. Added +ignore_unknown as a special item in host lists.
3060 36. Code for decoding IPv6 addresses in host lists is now included, even if
3061 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3062 address was recognized as an IP address, but was then not correctly decoded
3063 into binary, causing unexpected and incorrect effects when compared with
3070 1. Very minor rewording of debugging text in manualroute to say "list of
3071 hosts" instead of "hostlist".
3073 2. If verify=header_syntax was set, and a header line with an unqualified
3074 address (no domain) and a large number of spaces between the end of the
3075 name and the colon was received, the reception process suffered a buffer
3076 overflow, and (when I tested it) crashed. This was caused by some obsolete
3077 code that should have been removed. The fix is to remove it!
3079 3. When running in the test harness, delay a bit after writing a bounce
3080 message to get a bit more predictability in the log output.
3082 4. Added a call to search_tidyup() just before forking a reception process. In
3083 theory, someone could use a lookup in the expansion of smtp_accept_max_
3084 per_host which, without the tidyup, could leave open a database connection.
3086 5. Added the variables $recipient_data and $sender_data which get set from a
3087 lookup success in an ACL "recipients" or "senders" condition, or a router
3088 "senders" option, similar to $domain_data and $local_part_data.
3090 6. Moved the writing of debug_print from before to after the "senders" test
3093 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3094 problems for message scanning, either using a data ACL, or using
3095 local_scan() because the Received: header was not generated till after they
3096 were called (in order to set the time as the time of reception completion).
3097 I have revised the way this works. The header is now generated after the
3098 body is received, but before the ACL or local_scan() are called. After they
3099 are run, the timestamp in the header is updated.
3105 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3106 before starting a queue runner without re-exec. This happened only when
3107 deliver_drop_privilege was set or when the Exim user was set to root. The
3108 effect of the bug was that timeouts during subsequent deliveries caused
3109 crashes instead of being properly handled. The handler is now left at its
3110 default (and expected) setting.
3112 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3113 message, again when deliver_drop_privilege is set or Exim is run as root.
3114 The bug described in (1) was not present in this case, but the tidying up
3115 of the other signals was missing. I have made the two cases consistent.
3117 3. The ignore_target_hosts setting on a manualroute router was being ignored
3118 for hosts that were looked up using the /MX notation.
3120 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3123 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3124 operated on the sender address. After changing the $sender_address to <>
3125 for the sender address verify, Exim was re-instated it as the original
3126 (before rewriting) address, but remembering that it had rewritten it, so it
3127 wasn't rewriting it again. This bug also had the effect of breaking the
3128 sender address verification caching when the sender address was rewritten.
3130 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3131 This has been changed so that if the ip literal address matches
3132 ignore_target_hosts, the router declines.
3134 7. Added expansion conditions match_domain, match_address, and match_local_
3135 part (NOT match_host).
3137 8. The placeholder for the Received: header didn't have a length field set.
3139 9. Added code to Exim itself and to exim_lock to test for a specific race
3140 condition that could lead to file corruption when using MBX delivery. The
3141 issue is with the lockfile that is created in /tmp. If this file is removed
3142 after a process has opened it but before that process has acquired a lock,
3143 there is the potential for a second process to recreate the file and also
3144 acquire a lock. This could lead to two Exim processes writing to the file
3145 at the same time. The added code performs the same test as UW imapd; it
3146 checks after acquiring the lock that its file descriptor still refers to
3147 the same named file.
3149 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3150 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3153 11. Added the smtp_active_hostname option. If used, this will typically be made
3154 to depend on the incoming interface address. Because $interface_address is
3155 not set up until the daemon has forked a reception process, error responses
3156 that can happen earlier (such as "too many connections") no longer contain
3159 12. If an expansion in a condition on a "warn" statement fails because a lookup
3160 defers, the "warn" statement is abandoned, and the next ACL statement is
3161 processed. Previously this caused the whole ACL to be aborted.
3163 13. Added the iplsearch lookup type.
3165 14. Added ident_timeout as a log selector.
3167 15. Added tls_certificate_verified as a log selector.
3169 16. Added a global option tls_require_ciphers (compare the smtp transport
3170 option of the same name). This controls incoming TLS connections.
3172 17. I finally figured out how to make tls_require_ciphers do a similar thing
3173 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3174 before starting the TLS session.
3176 18. Tabs are now shown as \t in -bP output.
3178 19. If the log selector return_path_on_delivery was set, Exim crashed when
3179 bouncing a message because it had too many Received: header lines.
3181 20. If two routers both had headers_remove settings, and the first one included
3182 a superfluous trailing colon, the final name in the first list and the
3183 first name in the second list were incorrectly joined into one item (with a
3184 colon in the middle).
3190 1. Added -C and -D options to the exinext utility, mainly to make it easier
3191 to include in the automated testing, but these could be helpful when
3192 multiple configurations are in use.
3194 2. The exinext utility was not formatting the output nicely when there was
3195 an alternate port involved in the retry record key, nor when there was a
3196 message id as well (for retries that were specific to a specific message
3197 and a specific host). It was also confused by IPv6 addresses, because of
3198 the additional colons they contain. I have fixed the IPv4 problem, and
3199 patched it up to do a reasonable job for IPv6.
3201 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3202 delivery, the log line now contains "pipelined" if PIPELINING was used.
3204 4. An SMTP transport process used to panic and die if the bind() call to set
3205 an explicit outgoing interface failed. This has been changed; it is now
3206 treated in the same way as a connect() failure.
3208 5. A reference to $sender_host_name in the part of a conditional expansion
3209 that was being skipped was still causing a DNS lookup. This no longer
3212 6. The def: expansion condition was not recognizing references to header lines
3213 that used bh_ and bheader_.
3215 7. Added the _cache feature to named lists.
3217 8. The code for checking quota_filecount in the appendfile transport was
3218 allowing one more file than it should have been.
3220 9. For compatibility with Sendmail, the command line option
3228 and sets the incoming protocol and host name (for trusted callers). The
3229 host name and its colon can be omitted when only the protocol is to be set.
3230 Note the Exim already has two private options, -pd and -ps, that refer to
3231 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3232 "s", but I don't think that's a major issue.
3234 10. A number of refactoring changes to the code, none of which should affect
3237 (a) The number of logging options was getting close to filling up the
3238 32-bit word that was used as a bit map. I have split them into two classes:
3239 those that are passed in the argument to log_write(), and those that are
3240 only ever tested independently outside of that function. These are now in
3241 separate 32-bit words, so there is plenty of room for expansion again.
3242 There is no change in the user interface or the logging behaviour.
3244 (b) When building, for example, log lines, the code previously used a
3245 macro that called string_cat() twice, in order to add two strings. This is
3246 not really sufficiently general. Furthermore, there was one instance where
3247 it was actually wrong because one of the argument was used twice, and in
3248 one call a function was used. (As it happened, calling the function twice
3249 did not affect the overall behaviour.) The macro has been replaced by a
3250 function that can join an arbitrary number of extra strings onto a growing
3253 (c) The code for expansion conditions now uses a table and a binary chop
3254 instead of a serial search (which was left over from when there were very
3255 few conditions). Also, it now recognizes conditions like "pam" even when
3256 the relevant support is not compiled in: a suitably worded error message is
3257 given if an attempt is made to use such a condition.
3259 11. Added ${time_interval:xxxxx}.
3261 12. A bug was causing one of the ddress fields not to be passed back correctly
3262 from remote delivery subprocesses. The field in question was not being
3263 subsequently used, so this caused to problems in practice.
3265 13. Added new log selectors queue_time and deliver_time.
3267 14. Might have fixed a bug in maildirsizefile handling that threw up
3268 "unexpected character" debug warnings, and recalculated the data
3269 unnecessarily. In any case, I expanded the warning message to give more
3272 15. Added the message "Restricted characters in address" to the statements in
3273 the default ACL that block characters like @ and % in local parts.
3275 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3276 Three changes have been made:
3278 (a) There was a serious bug; a negative response to MAIL caused the whole
3279 recipient domain to be cached as invalid, thereby blocking all messages
3280 to all local parts at the same domain, from all senders. This bug has
3281 been fixed. The domain is no longer cached after a negative response to
3282 MAIL if the sender used is not empty.
3284 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3287 (c) A new callout option, "use_sender" has been added for people who want
3288 the modified behaviour.
3294 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3297 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3298 indeed breaks things for older releases.
3300 3. Added additional logging to the case where there is a problem reading data
3301 from a filter that is running in a subprocess using a pipe, in order to
3302 try to track down a specific problem.
3304 4. Testing facility fudge: when running in the test harness and attempting
3305 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3306 getting "No route to host". Convert this to a timeout.
3308 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3311 6. Some OS don't have socklen_t but use size_t instead. This affects the
3312 fifth argument of getsockopt() amongst other things. This is now
3313 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3314 can be set for individual OS. I have set it for SunOS5, OSF1, and
3315 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3316 some earlier ones do not.
3318 7. Change 4.30/15 was not doing the test caselessly.
3320 8. The standard form for an IPv6 address literal was being rejected by address
3321 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3322 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3323 this, as well as the form without the "IPv6" on the front (but only when
3324 address literals are enabled, of course).
3326 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3328 10. Exim crashed if a message with an empty sender address specified by -f
3329 encountered a router with an errors_to setting. This could be provoked only
3330 by a command such as
3334 where an empty string was supplied; "<>" did not hit this bug.
3336 11. Installed PCRE release 4.5.
3338 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3339 remained set. It is now erased.
3341 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3342 times from message ids (which are base 36 rather than the normal 62).
3344 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3345 were being counted as actual protocol errors, and logged if the log
3346 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3347 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3348 and DATA following a set of rejected RCPTs do not count as protocol errors.
3349 In other words, Exim assumes they were pipelined, though this may not
3350 actually be the case. Of course, in all cases the client gets an
3351 appropriate error code.
3353 15. If a lookup fails in an ACL condition, a message about the failure may
3354 be available; it is used if testing the ACL cannot continue, because most
3355 such messages specify what the cause of the deferral is. However, some
3356 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3357 that caused an old message to be retained and used if a later statement
3358 caused a defer, replacing the real cause of the deferral.
3360 16. If an IP address had so many PTR records that the DNS lookup buffer
3361 was not large enough to hold them, Exim could crash while trying to process
3362 the truncated data. It now detects and logs this case.
3364 17. Further to 4.21/58, another change has been made: if (and only if) the
3365 first line of a message (the first header line) ends with CRLF, a bare LF
3366 in a subsequent header line has a space inserted after it, so as not to
3367 terminate the header.
3369 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3370 unnecessarily, used atoi() instead of strtol(), and didn't check the
3371 termination when getting file sizes from file names by regex.
3373 19. Completely re-implemented the support for maildirsize files, in the light
3374 of a number of problems with the previous contributed implementation
3375 (4.30/29). In particular:
3377 . If the quota is zero, the maildirsize file is maintained, but no quota is
3380 . If the maildir directory does not exist, it is created before any attempt
3381 to write a maildirsize file.
3383 . The quota value in the file is just a cache; if the quota is changed in
3384 the transport, the new value overrides.
3386 . A regular expression is available for excluding directories from the
3389 20. The autoreply transport checks the characters in options that define the
3390 message's headers; it allows continued headers, but it was checking with
3391 isspace() after an embedded newline instead of explicitly looking for a
3394 21. If all the "regular" hosts to which an address was routed had passed their
3395 expiry times, and had not reached their retry times, the address was
3396 bounced, even if fallback hosts were defined. Now Exim should go on to try
3399 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3400 equivalent code in the SMTP transport. Some hosts send humungous responses
3401 to HELO/EHLO, more than 1024 it seems.
3403 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3404 gives compiler warnings in some environments. I've now done it "properly",
3407 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3408 (because of problems with the built-in one) was declared to return uschar *
3409 instead of char *, causing compiler failure.
3411 25. Fixed a file descriptor leak when processing alias/forward files.
3413 26. Fixed a minor format string issue in dbfn.c.
3415 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3417 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3418 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3421 29. When a sender address is verified, it is cached, to save repeating the test
3422 when there is more than one recipient in a message. However, when the
3423 verification involves a callout, it is possible for different callout
3424 options to be set for different recipients. It is too complicated to keep
3425 track of this in the cache, so now Exim always runs a verification when a
3426 callout is required, relying on the callout cache for the optimization.
3427 The overhead is duplication of the address routing, but this should not be
3430 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3431 to be verified with callout=postmaster, and the main callout worked but the
3432 postmaster check failed, the verification correctly failed. However, if a
3433 subsequent RCPT command asked for sender verification *without* the
3434 postmaster check, incorrect caching caused this verification also to fail,
3437 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3438 it was not caching the DNS options (qualify_single, search_parents) that
3439 were used when the lookup failed. A subsequent lookup with different
3440 options therefore always gave the same answer, though there were cases
3441 where it should not have. (Example: a "domains = !$mx_any" option on a
3442 dnslookup router: the "domains" option is always processed without any
3443 widening, but the router might have qualify_single set.) Now Exim uses the
3444 cached value only when the same options are set.
3446 32. Added John Jetmore's "exipick" utility to the distribution.
3448 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3449 than a timeout (e.g. a certificate is required, and is not provided), an
3450 Exim server now closes the connection immediately. Previously it waited for
3451 the client to close - but if the client is SSL, it seems that they each
3452 wait for each other, leading to a delay before one of them times out.
3454 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3455 maintained 0.8.x compatibility because I don't think many are using it, and
3456 it is clearly obsolete.
3458 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3461 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3462 client certificate was expired. A simple patch fixes this, though I don't
3463 understand the full logic of why the verify callback is called multiple
3466 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3467 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3468 which causes problems with some clients (such as the Certicom SSL Plus
3469 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3470 disables the coutermeasure allowing Eudora to connect."
3472 38. Exim was not checking that a write() to a log file succeeded. This could
3473 lead to Bad Things if a log got too big, in particular if it hit a file
3474 size limit. Exim now panics and dies if it cannot write to a log file, just
3475 as it does if it cannot open a log file.
3477 39. Modified OS/Makefile-Linux so that it now contains
3479 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3481 The two -D definitions ensure that Exim is compiled with large file
3482 support, which makes it possible to handle log files that are bigger than
3485 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3486 instance) a domain was checked against a named list that involved a lookup,
3487 causing $domain_data to be set, then another domain was checked against the
3488 same list, then the first domain was re-checked, the value of $domain_data
3489 after the final check could be wrong. In particular, if the second check
3490 failed, it could be set empty. This bug probably also applied to
3493 41. The strip_trailing_dot option was not being applied to the address given
3494 with the -f command-line option.
3496 42. The code for reading a message's header from the spool was incrementing
3497 $received_count, but never initializing it. This meant that the value was
3498 incorrect (doubled) while delivering a message in the same process in which
3499 it was received. In the most common configuration of Exim, this never
3500 happens - a fresh exec is done - but it can happen when
3501 deliver_drop_privilege is set.
3503 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3504 it now includes up to 150 characters of the unexpected data in the log
3507 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3508 and building data strings. The size of both of these buffers was 10 000
3509 bytes - far larger than anybody would *ever* want, thought I. Needless to
3510 say, somebody hit the limit. I have increased the maximum line length to
3511 20 000 and the maximum data length of concatenated lines to 100 000. I have
3512 also fixed two bugs, because there was no checking on these buffers. Tsk,
3513 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3514 buffer is too small.
3516 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3517 lsearch lookups. Now it does.
3519 46. When parsing a route_list item in a manualroute router, a fixed-length
3520 buffer was used for the list of hosts. I made this 1024 bytes long,
3521 thinking that nobody would ever have a list of hosts that long. Wrong.
3522 Somebody had a whole pile of complicated expansion conditions, and the
3523 string was silently truncated, leading to an expansion error. It turns out
3524 that it is easier to change to an unlimited length (owing to other changes
3525 that have happened since this code was originally written) than to build
3526 structure for giving a limitation error. The length of the item that
3527 expands into the list of hosts is now unlimited.
3529 47. The lsearch lookup could not handle data where the length of text line was
3530 more than 4095 characters. Such lines were truncated, leading to shortened
3531 data being returned. It should now handle lines of any length.
3533 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3534 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3537 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3538 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3539 no longer happens in comment lines. A list of replacements is now placed
3540 at the head of all of the source files, except those whose only change is
3541 to replace PERL_COMMAND in the very first #! line.
3543 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3544 messages on the queue, with a bottom-up merge sort, using code contributed
3545 by Michael Haardt. This should make operations like -bp somewhat faster on
3546 large queues. It won't affect queue runners, except when queue_run_in_order
3549 51. Installed eximstats 1.31 in the distribution.
3551 52. Added support for SRV lookups to the dnslookup router.
3553 53. If an ACL referred to $message_body or $message_body_end, the value was not
3554 reset for any messages that followed in the same SMTP session.
3556 54. The store-handling optimization for building very long strings was not
3557 differentiating between the different store pools. I don't think this
3558 actually made any difference in practice, but I've tidied it.
3560 55. While running the routers to verify a sender address, $sender_address
3561 was still set to the sender address. This is wrong, because when routing to
3562 send a bounce to the sender, it would be empty. Therefore, I have changed
3563 it so that, while verifying a sender address, $sender_address is set to <>.
3564 (There is no change to what happens when verifying a recipient address.)
3566 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3567 target A or AAAA records (if not already returned) without resetting the
3568 qualify_single or search_parents options of the DNS resolver. These are
3569 inappropriate in this case because the targets of MX and SRV records must
3570 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3571 target that, when qualified, matched something in the local domain. These
3572 two options are now turned off when doing these lookups.
3574 57. It seems that at least some releases of Reiserfs (which does not have the
3575 concept of a fixed number of inodes) returns zero and not -1 for the
3576 number of available inodes. This interacted badly with check_spool_inodes,
3577 which assumed that -1 was the "no such thing" setting. What I have done is
3578 to check that the total number of inodes is greater than zero before doing
3579 the test of how many are available.
3581 58. When a "warn" ACL statement has a log_message modifier, the message is
3582 remembered, and not repeated. This is to avoid a lot of repetition when a
3583 message has many recipients that cause the same warning to be written.
3584 Howewer, Exim was preserving the list of already written lines for an
3585 entire SMTP session, which doesn't seem right. The memory is now reset if a
3586 new message is started.
3588 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3589 debugging output unless log_rewrite was also set.
3591 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3592 of (int)(handle) when we know that handle contains (void *)(-1).
3594 61. The Exim daemon panic-logs an error return when it closes the incoming
3595 connection. However "connection reset by peer" seems to be common, and
3596 isn't really an error worthy of noting specially, so that particular error
3599 62. When Exim is trying to find all the local interfaces, it used to panic and
3600 die if the ioctl to get the interface flags failed. However, it seems that
3601 on at least one OS (Solaris 9) it is possible to have an interface that is
3602 included in the list of interfaces, but for which you get a failure error
3603 for this call. This happens when the interface is not "plumbed" into a
3604 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3605 failure of the "get flags" call assumes that the interface is down.
3607 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3608 makes life easier for people who are doing arithmetic on fields extracted
3609 from dates, where you often get leading zeros that should not be
3610 interpreted as octal.
3612 64. Added qualify_domain to the redirect router, to override the global
3615 65. If a pathologically long header line contained very many addresses (the
3616 report of this problem mentioned 10 000) and each of them was rewritten,
3617 Exim could use up a very large amount of memory. (It kept on making new
3618 copies of the header line as it rewrote, and never released the old ones.)
3619 At the expense of a bit more processing, the header rewriting function has
3620 been changed so that it no longer eats memory in this way.
3622 66. The generation of the Received: header has been moved from the time that a
3623 message starts to be received, to the time that it finishes. The timestamp
3624 in the Received: header should now be very close to that of the <= log
3625 line. There are two side-effects of this change:
3627 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3628 logged header lines no longer include the local Received: line, because
3629 it has not yet been created. The same applies to a copy of the message
3630 that is returned to a non-SMTP sender when a message is rejected.
3632 (b) When a filter file is tested using -bf, no additional Received: header
3633 is added to the test message. After some thought, I decided that this
3636 This change does not affect the value of $received_for. It is still set
3637 after address rewriting, but before local_scan() is called.
3639 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3641 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3642 gave an unhelpful panic error message, and a defer error. I have managed to
3643 change this behaviour so that it now rejects any supplied certificate,
3644 which seems right, as the list of acceptable certificates is empty.
3646 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3647 gave an unhelpful defer error. I have not managed to make this reject any
3648 supplied certificates, but the error message it gives is "no certificate
3649 supplied", which is not helpful.
3651 70. exigrep's output now also includes lines that are not associated with any
3652 message, but which match the given pattern. Implemented by a patch from
3653 Martin Sluka, which also tidied up the Perl a bit.
3655 71. Recipient callout verification, like sender verification, was using <> in
3656 the MAIL FROM command. This isn't really the right thing, since the actual
3657 sender may affect whether the remote host accepts the recipient or not. I
3658 have changed it to use the actual sender in the callout; this means that
3659 the cache record is now keyed on a recipient/sender pair, not just the
3660 recipient address. There doesn't seem to be a real danger of callout loops,
3661 since a callout by the remote host to check the sender would use <>.
3662 [SEE ABOVE: changed after hitting problems.]
3664 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3665 temporary errors. However, in the case of such a code being given after
3666 the end of a data transmission (i.e. after ".") Exim was failing to write
3667 a retry record for the message. (Yes, there was some broken host that was
3668 actually sending 8xx at this point.)
3670 73. An unknown lookup type in a host list could cause Exim to panic-die when
3671 the list was checked. (An example that provoked this was putting <; in the
3672 middle of a list instead of at the start.) If this happened during a DATA
3673 ACL check, a -D file could be left lying around. This kind of configuration
3674 error no longer causes Exim to die; instead it causes a defer errror. The
3675 incident is still logged to the main and panic logs.
3677 74. Buglet left over from Exim 3 conversion. The message "too many messages
3678 in one connection" was written to the rejectlog but not the mainlog, except
3679 when address rewriting (yes!) was being logged.
3681 75. Added write_rejectlog option.
3683 76. When a system filter was run not as root (that is, when system_filter_user
3684 was set), the values of the $n variables were not being returned to the
3685 main process; thus, they were not subsequently available in the $sn
3688 77. Added +return_path_on_delivery log selector.
3690 78. A connection timeout was being treated differently from recipients deferred
3691 when testing hosts_max_try with a message that was older than the host's
3692 retry timeout. (The host should not be counted, thus allowing all hosts to
3693 be tried at least once before bouncing.) This may have been the cause of an
3694 occasionally reported bug whereby a message would remain on the queue
3695 longer than the retry timeout, but would be bounced if a delivery was
3696 forced. I say "may" because I never totally pinned down the problem;
3697 setting up timeout/retry tests is difficult. See also the next item.
3699 79. The ultimate address timeout was not being applied to errors that involved
3700 a combination of host plus message (for example, a timeout on a MAIL
3701 command). When an address resolved to a number of possible hosts, and they
3702 were not all tried for each delivery (e.g. because of hosts_max_try), a
3703 message could remain on the queue longer than the retry timeout.
3705 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3708 81. Fixed an obscure SMTP outgoing bug which required at least the following
3709 conditions: (a) there was another message waiting for the same server;
3710 (b) the server returned 5xx to all RCPT commands in the first message so
3711 that the message was not completed; (c) the server dropped the connection
3712 or gave a negative response to the RSET that Exim sends to abort the
3713 transaction. The observed case was a dropped connection after DATA that had
3714 been sent in pipelining mode. That is, the server had advertised PIPELINING
3715 but was not implementing it correctly. The effect of the bug was incorrect
3716 behaviour, such as trying another host, and this could lead to a crash.
3722 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3723 and daemon.c were passed as pointers to ints; they should have been
3724 pointers to socklen_t variables (which are typically unsigned ints).
3726 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3729 3. Fixed a really odd bug that affected only the testing scheme; patching a
3730 certain fixed string in the binary changed the value of another string that
3731 happened to be identical to the end of the original first string.
3733 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3734 name", it returns that address as the IP address. On some operating
3735 systems (e.g. Solaris), it also passes back the IP address string as the
3736 "host name". However, on others (e.g. Linux), it passes back an empty
3737 string. Exim wasn't checking for this, and was changing the host name to an
3738 empty string, assuming it had been canonicized.
3740 5. Although rare, it is permitted to have more than one PTR record for a given
3741 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3742 all the names associated with an address, because they do in Solaris.
3743 However, it seems that they do not in Linux for data that comes from the
3744 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3745 I found this out when I moved to a new Linux workstation and tried to run
3746 the Exim test suite.
3748 To get round this problem I have changed the code so that it now does its
3749 own call to the DNS to look up PTR records when searching for a host name.
3750 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3751 addresses that are only in /etc/hosts are still found.
3753 This behaviour is, however, controlled by an option called host_lookup_
3754 order, which defaults to "bydns:byaddr". If people want to use the other
3755 order, or indeed, just use one or the other means of lookup, they can
3756 specify it in this variable.
3758 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3759 some operating systems, this comes back from gethostbyaddr() as an empty
3760 string, and this is what Exim used to test for. However, it seems that in
3761 other systems, "." is yielded. Exim now tests for this case too.
3763 7. The values of check_spool_space and check_log_space are now held internally
3764 as a number of kilobytes instead of an absolute number of bytes. If a
3765 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3766 kilobyte. This means that much larger values can be stored.
3768 8. Exim monitor: an attempt to get the action menu when not actually pointing
3769 at a message produces an empty menu entitled "No message selected". This
3770 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3771 no entries in it ("Shell widget menu has zero width and/or height"). So I
3772 have added a single, blank menu entry in this case.
3774 9. Added ${quote_local_part.
3776 10. MIME decoding is now applied to the contents of Subject: header lines when
3779 11. Now that a reference to $sender_host_address automatically causes a reverse
3780 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3781 host lookup before query-style lookups in lists that might use this
3782 variable. This has therefore been abolished, and the "net-" prefix is no
3783 longer necessary for query-style lookups.
3785 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3786 have been a typo for LFLAGS, so it has been changed.
3788 13. The install script calls Exim with "-C /dev/null" in order to find the
3789 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3790 to be output. Howeve, since Exim outputs its version number before the
3791 error, it didn't break the script. It just looked ugly. I fixed this by
3792 always allowing "-C /dev/null" if the caller is root.
3794 14. Ignore overlarge ACL variable number when reading spool file - insurance
3795 against a later release with more variables having written the file.
3797 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3798 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3799 this, as well as the form without the "IPv6" on the front.
3801 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3802 OS/Makefile-Darwin file.
3804 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3805 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3807 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3808 variable. However, some OS use ints for inodes. I've added cast to long int
3809 to get rid of the compiler warning.
3811 19. I had forgotten to lock out "/../" in configuration file names when
3812 ALT_CONFIG_PREFIX was set.
3814 20. Routers used for verification do not need to specify transports. However,
3815 if such a router generated a host list, and callout was configured, Exim
3816 crashed, because it could not find a port number from the (non-existent)
3817 transport. It now assumes port 25 in this circumstance.
3819 21. Added the -t option to exigrep.
3821 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3822 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3823 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3824 them is set, LOOKUP_LSEARCH is forced.
3826 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3828 24. Added sender and host information to the "rejected by local_scan()" log
3829 line; previously there was no indication of these.
3831 25. Added .include_if_exists.
3833 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3834 Linux. It turns out that not all file systems support this. Apparently some
3835 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3836 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3837 not supported on the file descriptor, is now ignored when Exim is trying to
3838 sync a directory. This applies only to Linux.
3840 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3842 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3843 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3846 29. Added support for maildirsize files from supplied patch (modified a bit).
3848 30. The use of :fail: followed by an empty string could lead Exim to respond to
3849 sender verification failures with (e.g.):
3851 550 Verification failed for <xxx>
3852 550 Sender verify failed
3854 where the first response line was missing the '-' that indicates it is not
3855 the final line of the response.
3857 31. The loop for finding the name of the user that called Exim had a hardwired
3858 limit of 10; it now uses the value of finduser_retries, which is used for
3859 all other user lookups.
3861 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3864 33. Exim was neglecting to zero errno before one call of strtol() when
3865 expanding a string and expecting an integer value. On some systems this
3866 resulted in spurious "integer overflow" errors. Also, it was casting the
3867 result into an int without checking.
3869 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3870 did not work. The code looks as if it has *never* worked, though it appears
3871 to have been documented since at least releast 1.62. I have made it work.
3873 35. The "timeout_DNS" error in retry rules, also documented since at least
3874 1.62, also never worked. As it isn't clear exactly what this means, and
3875 clearly it isn't a major issue, I have abolished the feature by treating it
3876 as "timeout", and writing a warning to the main and panic logs.
3878 36. The display of retry rules for -brt wasn't always showing the error code
3881 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3882 timeout_connect_A, timeout_connect_MX.
3884 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3885 to the empty sender.
3887 39. The daemon was not analysing the content of -oX till after it had closed
3888 stderr and disconnected from the controlling terminal. This meant that any
3889 syntax errors were only noted on the panic log, and the return code from
3890 the command was 0. By re-arranging the code a little, I've made the
3891 decoding happen first, so such errors now appear on stderr, and the return
3892 code is 1. However, the actual setting up of the sockets still happens in
3893 the disconnected process, so errors there are still only recorded on the
3896 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3897 connections (as happens on some IP stacks) was logged at start up time as
3898 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3899 it from "IPv6 and IPv4", which means that two separate sockets are being
3902 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3903 says whether AF_INET or AF_INET6 was passed as an argument.
3905 42. Exiwhat output was messed up when time zones were included in log
3908 43. Exiwhat now gives more information about the daemon's listening ports,
3909 and whether -tls-on-connect was used.
3911 44. The "port" option of the smtp transport is now expanded.
3913 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3914 silently ignored. Now an error message is written to the main and panic
3917 46. There's a new ACL modifier called "logwrite" which writes to a log file
3918 as soon as it is encountered.
3920 47. Added $local_user_uid and $local_user_gid at routing time.
3922 48. Exim crashed when trying to verify a sender address that was being
3925 49. Exim was recognizing only a space character after ".include". It now also
3926 recognizes a tab character.
3928 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3929 extracting the relevant information from the specification. The man page no
3930 longer contains scrambled data for the -d option, and I've added a section
3931 at the front about calling Exim under different names.
3933 51. Added "extra_headers" argument to the "mail" command in filter files.
3935 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3938 53. Installed eximstats 1.29.
3940 54. Added transport_filter_timeout as a generic transport option.
3942 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3943 Cc: header lines. This was required by RFC 822, but it not required by RFC
3946 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3947 incoming messages that did not have them. Now it does so only if the
3948 message originates locally, that is, if there is no associated remote host
3949 address. When Resent- header lines are present, this applies to the Resent-
3950 lines rather than the non-Resent- lines.
3952 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3953 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3955 58. Messages for configuration errors now include the name of the main
3956 configuration file - useful now that there may be more than one file in a
3957 list (.included file names were always shown).
3959 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3960 for those rare installations that do not start the daemon as root or run it
3961 setuid root. I've cut out the call to initgroups() if the daemon is not
3964 60. The Exim user and group can now be bound into the binary as text strings
3965 that are looked up at the start of Exim's processing.
3967 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3969 62. Added $mailstore_basename variable.
3971 63. Installed patch to sieve.c from Michael Haardt.
3973 64. When Exim failed to open the panic log after failing to open the main log,
3974 the original message it was trying to log was written to stderr and debug
3975 output, but if they were not available (the usual case in production), it
3976 was lost. Now it is written to syslog before the two lines that record the
3977 failures to open the logs.
3979 65. Users' Exim filters run in subprocesses under the user's uid. It is
3980 possible for a "deliver" command or an alias in a "personal" command to
3981 provoke an address rewrite. If logging of address rewriting is configured,
3982 this fails because the process is not running as root or exim. There may be
3983 a better way of dealing with this, but for the moment (because 4.30 needs
3984 to be released), I have disabled address rewrite logging when running a
3985 filter in a non-root, non-exim process.
3991 1. The buildconfig auxiliary program wasn't quoting the value set for
3992 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3993 not defined. This bug was masked in 4.22 by the effect that was fixed in
3996 2. Some messages that were rejected after a message id was allocated were
3997 shown as "incomplete" by exigrep. It no longer does this for messages that
3998 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4000 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4001 have allow_domain_literals set, the ID did not get logged in the <= line.
4002 Domain literals are now always recognized in Message-ID: header lines.
4004 4. The first argument for a ${extract expansion item is the key name or field
4005 number. Leading and trailing spaces in this item were not being ignored,
4006 causing some misleading effects.
4008 5. When deliver_drop_privilege was set, single queue runner processes started
4009 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4010 same command in the process it spins off) were not dropping privilege.
4012 6. When the daemon running as "exim" started a queue runner, it always
4013 re-executed Exim in the spun-off process. This is a waste of effort when
4014 deliver_drop_privilege is set. The new process now just calls the
4015 queue-runner function directly.
4021 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4024 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4025 ignored. Though the use of -oP was forcing the writing of a pid file, it
4026 was always written to the default place.
4028 3. If the message "no IP address found for host xxxx" is generated during
4029 incoming verification, it is now followed by identification of the incoming
4030 connection (so you can more easily find what provoked it).
4032 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4034 5. Added some features to "harden" Exim a bit more against certain attacks:
4036 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4037 be put in Local/Makefile. This is like the never_users runtime option,
4038 but it cannot be overridden. The default setting is "root".
4040 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4041 prefix string with which any file named in a -C command line option
4044 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4045 is retained for -C and -D only if the caller of Exim is root. Without
4046 it, the exim user may also use -C and -D and retain privilege.
4048 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4049 command line option is disabled.
4051 6. Macro names set by the -D option must start with an upper case letter, just
4052 like macro names defined in the configuration file.
4054 7. Added "dereference=" facility to LDAP.
4056 8. Two instances of the typo "uknown" in the source files are fixed.
4058 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4059 the Configure-Makefile script screwed up while processing it.
4061 10. Incorporated PCRE 4.4.
4063 11. The SMTP synchronization check was not operating right at the start of an
4064 SMTP session. For example, it could not catch a HELO sent before the client
4065 waited for the greeting. There is now a check for outstanding input at the
4066 point when the greeting is written. Because of the duplex, asynchronous
4067 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4068 way, but not yet received, when the check is performed.
4070 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4071 on TCP/IP sockets, because this apparently causes some broken clients to
4074 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4075 unchanged) from the Cygwin maintainer.
4077 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4078 was supported instead of testing for mbx. Effectively a typo.
4080 15. The spa authenticator server code was not checking that the input it
4081 received was valid base64.
4083 16. The debug output line for the "set" modifier in ACLs was not showing the
4084 name of the variable that was being set.
4086 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4088 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4089 lookup on its own. Something else was needed to trigger the lookup. For
4090 example, a match in host_lookup or the need for a host name in a host list.
4091 Now, if $sender_host_name is referenced and the host name has not yet been
4092 looked up, a lookup is performed. If the lookup fails, the variable remains
4093 empty, and $host_lookup_failed is set to "1".
4095 19. Added "eqi" as a case-independent comparison operator.
4097 20. The saslauthd authentication condition could segfault if neither service
4098 nor realm was specified.
4100 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4101 error message that was logged was misleading, and incoming SMTP
4102 connections were dropped. The message is now more accurate, and temporary
4103 errors are given to SMTP connections.
4105 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4106 (except RSET) with a 421 error, until QUIT is received. However, it was
4107 failing to send a response to QUIT.
4109 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4110 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4111 ACL is now run after the helo_try_verify_hosts code.
4113 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4114 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4115 case-independent, so other case variants are also recognized). Apparently
4116 some systems use these upper case variants.
4118 25. If more than two messages were waiting for the same host, and a transport
4119 filter was specified for the transport, Exim sent two messages over the
4120 same TCP/IP connection, and then failed with "socket operation on non-
4121 socket" when it tried to send the third.
4123 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4125 27. The extern definition of crypt16() in expand.c was not being excluded when
4126 the OS had its own crypt16() function.
4128 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4129 as a preferred synonym for return_size_limit, both as an option and as an
4132 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4134 30. Changed the default configuration ACL to relax the local part checking rule
4135 for addresses that are not in any local domains. For these addresses,
4136 slashes and pipe symbols are allowed within local parts, but the sequence
4137 /../ is explicitly forbidden.
4139 31. SPA server authentication was not clearing the challenge buffer before
4142 32. log_message in a "warn" ACL statement was writing to the reject log as
4143 well as to the main log, which contradicts the documentation and doesn't
4144 seem right (because no rejection is happening). So I have stopped it.
4146 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4147 However, I am unable to do any testing of this.
4149 34. Fixed an infelicity in the appendfile transport. When checking directories
4150 for a mailbox, to see if any needed to be created, it was accidentally
4151 using path names with one or more superfluous leading slashes; tracing
4152 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4154 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4155 contents are added to the log line that is written for every discarded
4156 recipient. (Previously a log_message setting was ignored.)
4158 36. The ${quote: operator now quotes the string if it is empty.
4160 37. The install script runs exim in order to find its version number. If for
4161 some reason other than non-existence or emptiness, which it checks, it
4162 could not run './exim', it was installing it with an empty version number,
4163 i.e. as "exim-". This error state is now caught, and the installation is
4166 38. An argument was missing from the function that creates an error message
4167 when Exim fails to connect to the socket for saslauthd authentication.
4168 This could cause Exim to crash, or give a corrupted message.
4170 39. Added isip, isip4, and isip6 to ${if conditions.
4172 40. The ACL variables $acl_xx are now saved with the message, and can be
4173 accessed later in routers, transports, and filters.
4175 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4176 strings in the file are not string-expanded.
4178 42. If a MAIL command specified a SIZE value that was too large to fit into an
4179 int variable, the check against message_size_limit failed. Such values are
4180 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4181 day this will have to be increased, but I don't think I want to be around
4182 when emails are that large.
4189 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4190 iconv() is not standard in FreeBSD.
4192 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4193 IPv6 enabled. The observed symptom was a segmentation fault on return from
4194 the function os_common_find_running_interfaces() in src/os.c.
4196 3. In the check_special_case() function in daemon.c I had used "errno" as an
4197 argument name, which causes warnings on some systems. This was basically a
4198 typo, since it was named "eno" in the comments!
4200 4. The code that waits for the clock to tick (at a resolution of some fraction
4201 of a second) so as to ensure message-id uniqueness was always waiting for
4202 at least one whole tick, when it could have waited for less. [This is
4203 almost certainly not relevant at current processor speeds, where it is
4204 unlikely to ever wait at all. But we try to future-proof.]
4206 5. The function that sleeps for a time interval that includes fractions of a
4207 second contained a race. It did not block SIGALRM between setting the
4208 timer, and suspending (a couple of lines later). If the interval was short
4209 and the sigsuspend() was delayed until after it had expired, the suspension
4210 never ended. On busy systems this could lead to processes getting stuck for
4213 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4214 process, before it forks any delivery processes. The open lookup caching
4215 mechanism meant that the open file or database connection was passed into
4216 the delivery process. The problem was that delivery processes always tidy
4217 up cached lookup data. This could cause a problem for the next delivery
4218 process started by the queue runner, because the external queue runner
4219 process does not know about the closure. So the next delivery process
4220 still has data in the lookup cache. In the case of a file lookup, there was
4221 no problem because closing a file descriptor in a subprocess doesn't affect
4222 the parent. However, if the lookup was caching a connection to a database,
4223 the connection was closed, and the second delivery process was likely to
4224 see errors such as "PGSQL: query failed: server closed the connection
4225 unexpectedly". The problem has been fixed by closing all cached lookups
4226 in a queue runner before running a delivery process.
4228 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4229 seem to have the "const" qualifier which it has on other OS. I've
4232 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4233 *of the same type* (client or server) with the same public name that an
4234 error should be diagnosed.
4236 9. When Exim looked up a host name for an IP address, but failed to find the
4237 original IP address when looking up the host name (a safety check), it
4238 output the message "<ip address> does not match any IP for NULL", which was
4239 confusing, to say the least. The bug was that the host name should have
4240 appeared instead of "NULL".
4242 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4243 user that is built into the binary, and the -C or -D options is used, root
4244 privilege is dropped before the configuration file is read. In addition,
4245 logging is switched to stderr instead of the normal log files. If the
4246 configuration then re-defines the Exim user, the unprivileged environment
4247 is probably not what is expected, so Exim logs a panic warning message (but
4250 However, if deliver_drop_privilege is set, the unprivileged state may well
4251 be exactly what is intended, so the warning has been cut out in that case,
4252 and Exim is allowed to try to write to its normal log files.
4258 1. smtp_return_error_details was not giving details for temporary sender
4259 or receiver verification errors.
4261 2. Diagnose a configuration error if two authenticators have the same public
4264 3. Exim used not to create the message log file for a message until the first
4265 delivery attempt. This could be confusing when incoming messages were held
4266 for policy or load reasons. The message log file is now created at the time
4267 the message is received, and an initial "Received" line is written to it.
4269 4. The automatically generated man page for command line options had a minor
4270 bug that caused no ill effects; however, a more serious problem was that
4271 the procedure for building the man page automatically didn't always
4272 operate. Consequently, release 4.20 contains an out-of-date version. This
4273 shouldn't happen again.
4275 5. When building Exim with embedded Perl support, the script that builds the
4276 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4277 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4279 6. The freeze_tell option was not being used for messages that were frozen on
4280 arrival, either by an ACL or by local_scan().
4282 7. Added the smtp_incomplete_transaction log selector.
4284 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4285 was accepting AUTH without a new EHLO.
4287 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4288 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4291 10. Logging of TCP/IP connections (when configured) now happens in the main
4292 daemon process instead of the child process, so that the TCP/IP connection
4293 count is more accurate (but it can never be perfect).
4295 11. The use of "drop" in a nested ACL was not being handled correctly in the
4296 outer ACL. Now, if condition failure induced by the nested "drop" causes
4297 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4298 or "require"), the connection is dropped.
4300 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4301 that yield "discard" can now be used with an "accept" or a "discard" verb,
4302 but an error is generated for any others (because I can't see a useful way
4303 to define what should happen).
4305 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4306 are now processed in the same way as lines in the Exim configuration file.
4307 In particular, continuation lines are supported.
4309 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4311 15. Added -ti meaning -t -i.
4313 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4314 domains, and warn by logging if others are found.
4316 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4317 returned from GIFCONF when Exim is trying to find the list of interfaces on
4318 a host. The code in os.c has been modified to copy each ifreq to an aligned
4319 structure in all cases.
4321 Also, in some cases, the returned ifreq's were being copied to a 'struct
4322 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4323 means the last couple of bytes of an IPv6 address could be chopped if the
4324 ifreq contained only a normal sockaddr (14 bytes storage).
4326 18. Named domain lists were not supported in the hosts_treat_as_local option.
4327 An entry such as +xxxx was not recognized, and was treated as a literal
4330 19. Ensure that header lines added by a DATA ACL are included in the reject log
4331 if the ACL subsequently rejects the message.
4333 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4334 MD5 (which is deprecated).
4336 21. When testing a filter file using -bf, Exim was writing a message when it
4337 took the sender from a "From " line in the message, but it was not doing so
4338 when it took $return_path from a Return-Path: header line. It now does.
4340 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4341 with a valid header line field name (a series of printing characters
4342 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4344 23. Changed "disc" in the source to "disk" to conform to the documentation and
4345 the book and for uniformity.
4347 24. Ignore Sendmail's -Ooption=value command line item.
4349 25. When execve() failed while trying to run a command in a pipe transport,
4350 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4351 could be confused with a return value of 69 from the command itself. This
4352 has been changed to 127, the value the shell returns if it is asked to run
4353 a non-existent command. The wording for the related log line suggests a
4354 non-existent command as the problem.
4356 26. If received_header_text expands to an empty string, do not add a Received:
4357 header line to the message. (Well, it adds a token one on the spool, but
4358 marks it "old" so that it doesn't get used or transmitted.)
4360 27. Installed eximstats 1.28 (addition of -nt option).
4362 28. There was no check for failure on the call to getsockname() in the daemon
4363 code. This can fail if there is a shortage of resources on the system, with
4364 ENOMEM, for example. A temporary error is now given on failure.
4366 29. Contrary to the C standard, it seems that in some environments, the
4367 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4368 program. Exim now does this explicitly; it affects the formatting of
4369 timestamps using strftime().
4371 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4372 complaints. I've now initialized all the variables, to avoid this.
4374 32. Header lines added by a system filter were not being "seen" during
4375 transport-time rewrites.
4377 33. The info_callback() function passed to OpenSSL is set up with type void
4378 (*)(SSL *, int, int), as described somewhere. However, when calling the
4379 function (actually a macro) that sets it up, the type void(*)() is
4380 expected. I've put in a cast to prevent warnings from picky compilers.
4382 34. If a DNS black list lookup found a CNAME record, but there were no A
4383 records associated with the domain it pointed at, Exim crashed.
4385 35. If a DNS black list lookup returned more than one A record, Exim ignored
4386 all but the first. It now scans all returned addresses if a particular IP
4387 value is being sought. In this situation, the contents of the
4388 $dnslist_value variable are a list of all the addresses, separated by a
4391 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4392 to accept a host name and all its aliases if the forward lookup for any of
4393 them yielded the IP address of the incoming connection. Now it accepts only
4394 those names whose forward lookup yields the correct IP address. Any other
4395 names are discarded. This closes a loophole whereby a rogue DNS
4396 administrator could create reverse DNS records to break through a
4397 wildcarded host restriction in an ACL.
4399 37. If a user filter or a system filter that ran in a subprocess used any of
4400 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4401 the wrong values were passed to the pipe command ($thisaddress had the
4402 value of $0, $0 had the value of $1, etc). This bug was introduced by
4403 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4405 38. Improved the line breaking for long SMTP error messages from ACLs.
4406 Previously, if there was no break point between 40 and 75 characters, Exim
4407 left the rest of the message alone. Two changes have been made: (a) I've
4408 reduced the minimum length to 35 characters; (b) if it can't find a break
4409 point between 35 and 75 characters, it looks ahead and uses the first one
4410 that it finds. This may give the occasional overlong line, but at least the
4411 remaining text gets split now.
4413 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4414 file descriptors might be low, and that setting 1000 would always raise it.
4415 It turns out that in some environments, the limit is already over 1000 and
4416 that lowering it causes trouble. So now Exim takes care not to decrease it.
4418 40. When delivering a message, the value of $return_path is set to $sender_
4419 address at the start of routing (routers may change the value). By an
4420 oversight, this default was not being set up when an address was tested by
4421 -bt or -bv, which affected the outcome if any router or filter referred to
4424 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4425 log only when "message" or "log_message" are set. However, if one of the
4426 conditions was an address verification, or a call to a nested ACL, the
4427 messages generated by the underlying test were being passed through. This
4428 no longer happens. The underlying message is available in $acl_verify_
4429 message for both "message" and "log_message" expansions, so it can be
4430 passed through if needed.
4432 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4433 new expansion $bh_ to give the encoded byte string without charset
4434 translation. Translation happens only if iconv() is available; HAVE_ICONV
4435 indicates this at build time. HEADERS_CHARSET gives the charset to
4436 translate to; headers_charset can change it in the configuration, and
4437 "headers charset" can change it in an individual filter file.
4439 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4440 that creates RFC 2047 encoded "words" labels them as that charset instead
4441 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4442 expansion operator; (ii) when Exim creates a From: line for a local
4443 message; (iii) when a header line is rewritten to include a "phrase" part.
4445 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4446 buggy, causing it to skip the first lines of messages whose message ID
4447 ended in 'D'. This would not have bitten before Exim release 4.14, because
4448 message IDs were unlikely to end in 'D' before then. The effect was to have
4449 incorrect size information for certain domains.
4451 45. #include "config.h" was missing at the start of the crypt16.c module. This
4452 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4455 46. If there was a timeout during a "random" callout check, Exim treated it as
4456 a failure of the random address, and carried on sending RSET and the real
4457 address. If the delay was just some slowness somewhere, the response to the
4458 original RCPT would be taken as a response to RSET and so on, causing
4459 mayhem of various kinds.
4461 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4462 when I implemented it. It didn't allow for the fact that some option values
4463 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4464 the right test for positive values.
4466 48. Domain names in DNS records are case-independent. Exim always looks them up
4467 in lower case. Some resolvers return domain names in exactly the case they
4468 appear in the zone file, that is, they may contain uppercase letters. Not
4469 all resolvers do this - some return always lower case. Exim was treating a
4470 change of case by a resolver as a change of domain, similar to a widening
4471 of a domain abbreviation. This triggered its re-routing code and so it was
4472 trying to route what was effectively the same domain again. This normally
4473 caused routing to fail (because the router wouldn't handle the domain
4474 twice). Now Exim checks for this case specially, and just changes the
4475 casing of the domain that it ultimately uses when it transmits the message
4478 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4481 50. If a filter generated a file delivery with a non-absolute name (possible if
4482 no home directory exists for the router), the forbid_file option was not
4485 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4486 the existing equality matching.
4488 52. Exim was using ints instead of ino_t variables in some places where it was
4489 dealing with inode numbers.
4491 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4492 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4493 called TMPDIR, and if it finds it is different, it changes its value.
4495 54. The smtp_printf() function is now made available to local_scan() so
4496 additional output lines can be written before returning. There is also an
4497 smtp_fflush() function to enable the detection of a dropped connection.
4498 The variables smtp_input and smtp_batched_input are exported to
4501 55. Changed the default runtime configuration: the message "Unknown user"
4502 has been removed from the ACL, and instead placed on the localuser router,
4503 using the cannot_route_message feature. This means that any verification
4504 failures that generate their own messages won't get overridden. Similarly,
4505 the "Unrouteable address" message that was in the ACL for unverifiable
4506 relay addresses has also been removed.
4508 56. Added hosts_avoid_esmtp to the smtp transport.
4510 57. The exicyclog script was not checking for the esoteric option
4511 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4512 will work only if exicyclog is run under the appropriate euid.
4514 58. Following a discussion on the list, the rules by which Exim recognises line
4515 endings on incoming messages have been changed. The -dropcr and drop_cr
4516 options are now no-ops, retained only for backwards compatibility. The
4517 following line terminators are recognized: LF CRLF CR. However, special
4518 processing applies to CR:
4520 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4521 nor a local message in the state where . is a terminator.
4523 (ii) If a bare CR is encountered in a header line, an extra space is added
4524 after the line terminator so as not to end the header. The reasoning
4525 behind this is that bare CRs in header lines are most likely either
4526 to be mistakes, or people trying to play silly games.
4528 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4529 was being incorrectly given as 18 bytes larger than it should have been.
4530 This is a VOB (very old bug).
4532 60. This may never have affected anything current, but just in case it has:
4533 When the local host is found other than at the start of a list of hosts,
4534 the local host, those with the same MX, and any that follow, are discarded.
4535 When the list in question was part of a longer list of hosts, the following
4536 hosts (not currently being processed) were also being discarded. This no
4537 longer happens. I'm not sure if this situation could ever has previously
4540 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4543 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4544 header. This is something that is recommended in a new Internet Draft, and
4545 is something that is documented as being done by Sendmail. There are two
4546 possible values. For messages generated by the autoreply transport, Exim
4549 Auto-Submitted: auto-replied
4551 whereas for all other generated messages (e.g. bounces) it adds
4553 Auto-Submitted: auto-generated
4555 63. The "personal" condition in filters now includes a test for the
4556 Auto-Submitted: header. If it contains the string "auto-" the message it
4557 not considered personal.
4559 64. Added rcpt_include_affixes as a generic transport option.
4561 65. Added queue_only_override (default true).
4563 66. Added the syslog_duplication option.
4565 67. If what should have been the first header line of a message consisted of
4566 a space followed by a colon, Exim was mis-interpreting it as a header line.
4567 It isn't of course - it is syntactically invalid and should therefore be
4568 treated as the start of the message body. The misbehaviour could have
4569 caused a number of strange effects, including loss of data in subsequent
4570 header lines, and spool format errors.
4572 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4573 client host had authenticated. This control can now be exercised by an ACL
4574 for more flexibility.
4576 69. By default, callouts do not happen when testing with -bh. There is now a
4577 variant, -bhc, which does actually run the callout code, including
4578 consulting and updating the callout cache.
4580 70. Added support for saslauthd authentication, courtesy of Alexander
4583 71. If statvfs() failed on the spool or log directories while checking their
4584 size for availability, Exim confusingly gave the error "space shortage".
4585 Furthermore, in debugging mode it crashed with a floating point exception.
4586 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4587 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4588 serious problem, Exim now writes to the main and panic logs when this
4589 happens, with details of the failure. It then refuses to accept the
4590 incoming message, giving the message "spool directory problem" or "log
4591 directory problem" with a 421 code for SMTP messages.
4593 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4594 0, 1, and 2 exist, because some OS complain for execs without them (see
4595 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4596 descriptors. However, the code omitted to check that the open succeeded,
4597 causing mysterious errors if for some reason the permissions on /dev/null
4598 got screwed. Now Exim writes a message to the main and panic logs, and
4599 bombs out if it can't open /dev/null.
4601 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4602 interact so that it is all more flexible. It is supposed to remain
4603 backwards compatible. Also added extra_local_interfaces.
4605 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4606 to bomb out with an assertion failure - to the client this appears as a
4607 connection drop. This problem occurs in the part of the code that was taken
4608 from the Samba project. Fortunately, the assertion is in a very simple
4609 function, so I have fixed this by reproducing the function inline in the
4610 one place where it is called, and arranging for authentication to fail
4611 instead of killing the process with assert().
4613 75. The SPA client code was not working when the server requested OEM rather
4614 than Unicode encoding.
4616 76. Added code to make require_files with a specific uid setting more usable in
4617 the case where statting the file as root fails - usually a non-root-mounted
4618 NFS file system. When this happens and the failure is EACCES, Exim now
4619 forks a subprocess and does the per-uid checking as the relevant uid.
4621 77. Added process_log_path.
4623 78. If log_file_path was not explicitly set, a setting of check_log_space or
4624 check_log_inodes was ignored.
4626 79. If a space check for the spool or log partitions fails, the incident is now
4627 logged. Of course, in the latter case the data may get lost...
4629 80. Added the %p formatting code to string_format() so that it can be used to
4630 print addresses in debug_print(). Adjusted all the address printing in the
4631 debugging in store.c to use %p rather than %d.
4633 81. There was a concern that a line of code in smtp_in.c could overflow a
4634 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4635 initially expressed, the concern was not well-founded, because trailing
4636 spaces are removed early. However, if the trailing spaces were followed by
4637 a NULL, they did not get removed, so the overflow was possible. Two fixes
4640 (a) I re-wrote the offending code in a cleaner fashion.
4641 (b) If an incoming SMTP command contains a NULL character, it is rejected
4644 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4645 runs initgroups(), so that if the Exim user is in any additional groups,
4646 they will be used during message reception.
4652 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob