1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.491 2007/03/13 11:06:48 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
10 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
13 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
16 PH/02 In an ACL statement such as
18 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
20 if a client was not listed at all, or was listed with a value other than
21 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
22 the condition was not true (as it should be), so access was not denied.
23 The bug was that the ! inversion was incorrectly passed on to the second
24 item. This has been fixed.
26 PH/03 Added additional dnslists conditions == and =& which are different from
27 = and & when the dns lookup returns more than one IP address.
29 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
30 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
32 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
33 FSYNC, which compiles an option called disable_fsync that allows for
34 bypassing fsync(). The documentation is heavily laced with warnings.
36 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
38 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
39 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
40 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
41 including adding "make clean"; (3) Added -fPIC when compiling the test
42 dynamically loaded module, to get rid of a warning.
44 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
45 message fails, move_frozen_messages = true and ignore_bounce_errors_after
46 = 0s. The bug is otherwise harmless.
48 PH/07 There was a bug in the dovecot authenticator such that the value of
49 $auth1 could be overwritten, and so not correctly preserved, after a
50 successful authentication. This usually meant that the value preserved by
51 the server_setid option was incorrect.
53 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
55 PH/09 Installed PCRE release 7.0.
57 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
58 run for batched SMTP input. It is now run at the start of every message
59 in the batch. While fixing this I discovered that the process information
60 (output by running exiwhat) was not always getting set for -bs and -bS
61 input. This is fixed, and it now also says "batched" for BSMTP.
63 PH/11 Added control=no_pipelining.
65 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
66 patch, slightly modified), and move the expansion of helo_data till after
67 the connection is made in the smtp transport (so it can use these
70 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
72 PH/14 Added log_selector = +pid.
74 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
76 PH/16 Add ${if forany and ${if forall.
78 PH/17 Added dsn_from option to vary the From: line in DSNs.
80 PH/18 Flush SMTP output before performing a callout, unless control =
81 no_callout_flush is set.
83 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
84 was true (the default) a successful delivery failed to delete the retry
85 item, thus causing premature timeout of the address. The bug is now
88 PH/20 Added hosts_avoid_pipelining to the smtp transport.
90 PH/21 Long custom messages for fakedefer and fakereject are now split up
91 into multiline reponses in the same way that messages for "deny" and
92 other ACL rejections are.
94 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
95 with slight modification.
97 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
98 draft, changing the syntax and factoring some duplicate code".
100 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
101 for deliveries of the second and subsequent messages over the same SMTP
104 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
105 ${reduce, with only minor "tidies".
107 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
109 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
110 expansion side effects.
112 PH/27 When a message times out after an over-quota error from an Exim-imposed
113 quota, the bounce message says "mailbox is full". This message was not
114 being given when it was a system quota that was exceeded. It now should
117 MH/03 Made $recipients available in local_scan(). local_scan() already has
118 better access to the recipient list through recipients_list[], but
119 $recipients can be useful in postmaster-provided expansion strings.
121 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
122 in the case of a MAIL command with additional options following the
123 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
124 were accidentally chopped off.
126 PH/29 SMTP synchronization checks are implemented when a command is read -
127 there is a check that no more input is waiting when there shouldn't be
128 any. However, for some commands, a delay in an ACL can mean that it is
129 some time before the response is written. In this time, more input might
130 arrive, invalidly. So now there are extra checks after an ACL has run for
131 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
132 pipelining has not been advertised.
134 PH/30 MH's patch to allow iscntrl() characters to be list separators.
136 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
137 returned in the SMTP response when smtp_return_error_details was false.
140 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
141 instead of the C I/O that was originally supplied, because problems were
144 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
145 Exim which did not show up earlier: it was assuming that a call to
146 SSL_CTX_set_info_callback() might give an error value. In fact, there is
147 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
148 was a macro that became an assignment, so it seemed to work. This has
149 changed to a proper function call with a void return, hence the compile
150 error. Exim's code has been fixed.
152 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
155 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
157 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
163 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
164 fixed by 4.65/MH/01 (is this a record?) are fixed:
166 (i) An empty string was always treated as zero by the numeric comparison
167 operators. This behaviour has been restored.
169 (ii) It is documented that the numeric comparison operators always treat
170 their arguments as decimal numbers. This was broken in that numbers
171 starting with 0 were being interpreted as octal.
173 While fixing these problems I realized that there was another issue that
174 hadn't been noticed. Values of message_size_limit (both the global option
175 and the transport option) were treated as octal if they started with 0.
176 The documentation was vague. These values are now always treated as
177 decimal, and I will make that clear in the documentation.
183 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
184 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
187 MH/01 Don't check that the operands of numeric comparison operators are
188 integers when their expansion is in "skipping" mode (fixes bug
189 introduced by 4.64-PH/07).
191 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
192 child addresses, Exim now panics and dies. Previously, because the count
193 is held in a short int, deliveries were likely to be lost. As such a
194 large number of recipients for a single message is ridiculous
195 (performance will be very, very poor), I have chosen to impose a limit
196 rather than extend the field.
202 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
203 leftover -K file (the existence of which was triggered by #402).
204 While we were at it, introduced process PID as part of the -K
205 filename. This should rule out race conditions when creating
208 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
209 processing considerably. Previous code took too long for large mails,
210 triggering a timeout which in turn triggers #401.
212 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
213 in the DK code in transports.c. sendfile() is not really portable,
214 hence the _LINUX specificness.
216 TF/01 In the add_headers option to the mail command in an Exim filter,
217 there was a bug that Exim would claim a syntax error in any
218 header after the first one which had an odd number of characters
221 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
222 callout verification, Exim cached a "reject" for the entire domain. This
223 is correct for most verifications, but it is not correct for a recipient
224 verification with use_sender or use_postmaster set, because in that case
225 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
226 case of MAIL FROM:<> rejection from other early rejections (e.g.
227 rejection of HELO). When verifying a recipient using a non-null MAIL
228 address, the cache is ignored if it shows MAIL FROM:<> rejection.
229 Whatever the result of the callout, the value of the domain cache is
230 left unchanged (for any other kind of callout, getting as far as trying
231 RCPT means that the domain itself is ok).
233 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
236 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
237 manifest itself as EPIPE rather than ECONNECT. When tidying away a
238 session, the daemon ignores ECONNECT errors and logs others; it now
239 ignores EPIPE as well.
241 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
242 (quoted-printable decoding).
244 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
245 later the small subsequent patch to fix an introduced bug.
247 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
249 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
251 PH/08 An error is now given if message_size_limit is specified negative.
253 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
254 to be given (somewhat) arbitrary names.
256 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
259 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
260 miscellaneous code fixes
262 PH/10 Added the log_reject_target ACL modifier to specify where to log
265 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
266 hostname. This is wrong, because it relates to the incoming message (and
267 probably the interface on which it is arriving) and not to the outgoing
268 callout (which could be using a different interface). This has been
269 changed to use the value of the helo_data option from the smtp transport
270 instead - this is what is used when a message is actually being sent. If
271 there is no remote transport (possible with a router that sets up host
272 addresses), $smtp_active_hostname is used.
274 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
275 tweaks were necessary in order to get it to work (see also 21 below):
276 (a) The code assumed that strncpy() returns a negative number on buffer
277 overflow, which isn't the case. Replaced with Exim's string_format()
279 (b) There were several signed/unsigned issues. I just did the minimum
280 hacking in of casts. There is scope for a larger refactoring.
281 (c) The code used strcasecmp() which is not a standard C function.
282 Replaced with Exim's strcmpic() function.
283 (d) The code set only $1; it now sets $auth1 as well.
284 (e) A simple test gave the error "authentication client didn't specify
285 service in request". It would seem that Dovecot has changed its
286 interface. Fortunately there's a specification; I followed it and
287 changed what the client sends and it appears to be working now.
289 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
292 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
293 address A is aliased to B and C, where B exists and C does not. Without
294 -v the output is "A verified" because verification stops after a
295 successful redirection if more than one address is generated. However,
296 with -v the child addresses are also verified. Exim was outputting "A
297 failed to verify" and then showing the successful verification for C,
298 with its parentage. It now outputs "B failed to verify", showing B's
299 parentage before showing the successful verification of C.
301 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
302 look up a TXT record in a specific list after matching in a combined
305 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
306 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
307 they consult the DNS. I had assumed they would set it the way they
308 wanted; and indeed my experiments on Linux seem to show that in some
309 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
310 To be on the safe side, however, I have now made the interface to
311 host_find_byname() similar to host_find_bydns(), with an argument
312 containing the DNS resolver options. The host_find_byname() function now
313 sets these options at its start, just as host_find_bydns() does. The smtp
314 transport options dns_qualify_single and dns_search_parents are passed to
315 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
316 of host_find_byname() use the default settings of RES_DEFNAMES
317 (qualify_single) but not RES_DNSRCH (search_parents).
319 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
320 spool_read_header() do less string testing, by means of a preliminary
321 switch on the second character of optional "-foo" lines. (This is
322 overdue, caused by the large number of possibilities that now exist.
323 Originally there were few.) While I was there, I also converted the
324 str(n)cmp tests so they don't re-test the leading "-" and the first
325 character, in the hope this might squeeze out yet more improvement.
327 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
328 flag allowing group syntax was set by the header_syntax check but not
329 turned off, possible causing trouble later; (2) The flag was not being
330 set at all for the header_verify test, causing "group"-style headers to
331 be rejected. I have now set it in this case, and also caused header_
332 verify to ignore an empty address taken from a group. While doing this, I
333 came across some other cases where the code for allowing group syntax
334 while scanning a header line wasn't quite right (mostly, not resetting
335 the flag correctly in the right place). These bugs could have caused
336 trouble for malformed header lines. I hope it is now all correct.
338 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
339 with the "reply" argument non-NULL. The code, however (which originally
340 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
341 but it didn't always do it. This confused somebody who was copying the
342 code for some other use. I have removed all the tests.
344 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
345 feature that was used to support insecure browsers during the U.S. crypto
346 embargo. It requires special client support, and Exim is probably the
347 only MTA that supported it -- and would never use it because real RSA is
348 always available. This code has been removed, because it had the bad
349 effect of slowing Exim down by computing (never used) parameters for the
350 RSA_EXPORT functionality.
352 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
353 authenticator to fail if there's a tab character in the incoming data
354 (there should never be unless someone is messing about, as it's supposed
355 to be base64-encoded). Also added, on Timo's advice, the "secured" option
356 if the connection is using TLS or if the remote IP is the same as the
357 local IP, and the "valid-client-cert option" if a client certificate has
360 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
361 authenticators. This can be used for authorization after authentication
362 succeeds. (In the case of plaintext, it servers for both authentication
365 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
366 if any retry times were supplied.
368 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
369 connection, where there is no client IP address to check. In this
370 situation, the verify now always succeeds.
372 PH/25 Applied John Jetmore's -Mset patch.
374 PH/26 Added -bem to be like -Mset, but loading a message from a file.
376 PH/27 In a string expansion for a processed (not raw) header when multiple
377 headers of the same name were present, leading whitespace was being
378 removed from all of them, but trailing whitespace was being removed only
379 from the last one. Now trailing whitespace is removed from each header
380 before concatenation. Completely empty headers in a concatenation (as
383 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
384 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
386 PH/29 [Removed. This was a change that I later backed out, and forgot to
387 correct the ChangeLog entry (that I had efficiently created) before
388 committing the later change.]
390 PH/30 Exim was sometimes attempting to deliver messages that had suffered
391 address errors (4xx response to RCPT) over the same connection as other
392 messages routed to the same hosts. Such deliveries are always "forced",
393 so retry times are not inspected. This resulted in far too many retries
394 for the affected addresses. The effect occurred only when there were more
395 hosts than the hosts_max_try setting in the smtp transport when it had
396 the 4xx errors. Those hosts that it had tried were not added to the list
397 of hosts for which the message was waiting, so if all were tried, there
398 was no problem. Two fixes have been applied:
400 (i) If there are any address or message errors in an SMTP delivery, none
401 of the hosts (tried or untried) are now added to the list of hosts
402 for which the message is waiting, so the message should not be a
403 candidate for sending over the same connection that was used for a
404 successful delivery of some other message. This seems entirely
405 reasonable: after all the message is NOT "waiting for some host".
406 This is so "obvious" that I'm not sure why it wasn't done
407 previously. Hope I haven't missed anything, but it can't do any
408 harm, as the worst effect is to miss an optimization.
410 (ii) If, despite (i), such a delivery is accidentally attempted, the
411 routing retry time is respected, so at least it doesn't keep
412 hammering the server.
414 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
415 in ${readsocket because some servers need this prod.
417 PH/32 Added some extra debug output when updating a wait-xxx database.
419 PH/33 The hint "could be header name not terminated by colon", which has been
420 given for certain expansion errors for a long time, was not being given
421 for the ${if def:h_colon_omitted{... case.
423 PH/34 The spec says: "With one important exception, whenever a domain list is
424 being scanned, $domain contains the subject domain." There was at least
425 one case where this was not true.
427 PH/35 The error "getsockname() failed: connection reset by peer" was being
428 written to the panic log as well as the main log, but it isn't really
429 panic-worthy as it just means the connection died rather early on. I have
430 removed the panic log writing for the ECONNRESET error when getsockname()
433 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
434 runs only) independently of the message's sender address. This meant
435 that, if the 4xx error was in fact related to the sender, a different
436 message to the same recipient with a different sender could confuse
437 things. In particualar, this can happen when sending to a greylisting
438 server, but other circumstances could also provoke similar problems.
439 I have changed the default so that the retry time for these errors is now
440 based a combination of the sender and recipient addresses. This change
441 can be overridden by setting address_retry_include_sender=false in the
444 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
445 remote server are returned as part of bounce messages. This was not
446 happening for LMTP over a pipe (the lmtp transport), but now it is the
447 same for both kinds of LMTP.
449 PH/38 Despite being documented as not happening, Exim was rewriting addresses
450 in header lines that were in fact CNAMEs. This is no longer the case.
452 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
453 and queue runs started by the daemon processed all messages. This has
454 been fixed so that -R and -S can now usefully be given with -q<time>.
456 PH/40 Import PCRE release 6.7 (fixes some bugs).
458 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
460 PH/42 Give an error if -q is specified more than once.
462 PH/43 Renamed the variables $interface_address and $interface_port as
463 $received_ip_address and $received_port, to make it clear that these
464 values apply to message reception, and not to the outgoing interface when
465 a message is delivered. (The old names remain recognized, of course.)
467 PH/44 There was no timeout on the connect() call when using a Unix domain
468 socket in the ${readsocket expansion. There now is.
470 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
471 be meaningful with "accept".
473 SC/01 Eximstats V1.43
474 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
476 SC/02 Eximstats V1.44
477 Use a glob alias rather than an array ref in the generated
478 parser. This improves both readability and performance.
480 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
481 Collect SpamAssassin and rejection statistics.
482 Don't display local sender or destination tables unless
483 there is data to show.
484 Added average volumes into the top table text output.
486 SC/04 Eximstats V1.46
487 Collect data on the number of addresses (recipients)
488 as well as the number of messages.
490 SC/05 Eximstats V1.47
491 Added 'Message too big' to the list of mail rejection
492 reasons (thanks to Marco Gaiarin).
494 SC/06 Eximstats V1.48
495 Mainlog lines which have GMT offsets and are too short to
496 have a flag are now skipped.
498 SC/07 Eximstats V1.49 (Alain Williams)
499 Added the -emptyok flag.
501 SC/08 Eximstats V1.50
502 Fixes for obtaining the IP address from reject messages.
504 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
505 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
506 whitesspace changes from 4.64-PH/27
508 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
511 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
512 are found, allow negative numbers in numeric criteria)
514 JJ/06 exipick.20061117.2, added new $message_body_missing variable
516 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
517 to match changes made in 4.64-PH/43
519 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
521 PH/47 Put in an explicit test for a DNS lookup of an address record where the
522 "domain" is actually an IP address, and force a failure. This locks out
523 those revolvers/nameservers that support "A-for-A" lookups, in
524 contravention of the specifications.
526 PH/48 When a host name was looked up from an IP address, and the subsequent
527 forward lookup of the name timed out, the host name was left in
528 $sender_host_name, contrary to the specification.
530 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
531 restricted to single-key lookups, Exim was not diagnosing an error if
532 * or *@ was used with a query-style lookup.
534 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
536 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
537 long ago, but noone interested enough thought of it. Let's just say that
538 the "1.1" means that there are some new functions that weren't there at
539 some point in the past.
541 PH/51 Error processing for expansion failure of helo_data from an smtp
542 transport during callout processing was broken.
544 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
545 tested/used via the -bh/-bhc/-bs options.
547 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
548 bug, fixed in subsequent PCRE releases).
550 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
551 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
553 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
559 SC/01 Use a glob alias rather than an array ref in eximstats generated
560 parser. This improves both readability and performance.
562 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
563 Don't display local sender or destination tables in eximstats unless
564 there is data to show.
565 Added average volumes into the eximstats top table text output.
567 SC/03 Collect data on the number of addresses (recipients) as well
568 as the number of messages in eximstats.
570 TF/01 Correct an error in the documentation for the redirect router. Exim
571 does (usually) call initgroups() when daemonizing.
573 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
574 with consistent privilege compared to when running as a daemon.
576 TF/03 Note in the spec that $authenticated_id is not set for local
577 submissions from trusted users.
579 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
580 Thanks to Dean Brooks <dean@iglou.com> for the patch.
582 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
583 by adding some example configuration directives to the default
584 configuration file. A little bit of work is required to uncomment the
585 directives and define how usernames and passwords are checked, but
586 there is now a framework to start from.
588 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
589 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
590 without this. I don't know how relevant this is to other LDAP libraries.
592 PH/02 Add the verb name to the "unknown ACL verb" error.
594 PH/03 Magnus Holmgren's patch for filter_prepend_home.
596 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
598 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
599 directory not expanded when it should be if an expanded home directory
600 was set for the address (which is overridden by the transport).
602 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
605 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
606 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
607 because it is too late at that time, and has no effect.
609 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
610 security issue with \' (bugzilla #107). I could not use the
611 PQescapeStringConn() function, because it needs a PGconn value as one of
614 PH/08 When testing addresses using -bt, indicate those final addresses that
615 are duplicates that would not cause an additional delivery. At least one
616 person was confused, thinking that -bt output corresponded to deliveries.
617 (Suppressing duplicates isn't a good idea as you lose the information
618 about possibly different redirections that led to the duplicates.)
620 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
621 systems where poll() doesn't work, in particular OS X.
623 PH/10 Added more information to debugging output for retry time not reached.
625 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
626 operations in malware.c.
628 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
631 PH/13 If write_rejectlog was set false when logging was sent to syslog with
632 syslog_duplication set false, log lines that would normally be written
633 both the the main log and to the reject log were not written to syslog at
636 PH/14 In the default configuration, change the use of "message" in ACL warn
637 statements to "add_header".
639 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
640 not followed by a command (e.g. "seen endif").
642 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
643 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
646 PH/17 Added extra conditions to the default value of delay_warning_condition
650 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
651 { match{$h_precedence:}{(?i)bulk|list|junk} } \
652 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
655 The Auto-Submitted: and various List- headers are standardised, whereas I
656 don't think Precedence: ever was.
658 PH/18 Refactored debugging code in route_finduser() to show more information,
659 in particular, the error code if getpwnam() issues one.
661 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
662 This is apparently needed in addition to the PH/07 change above to avoid
663 any possible encoding problems.
665 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
666 but not after initializing Perl.
668 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
669 output them only if debugging. By default they are written stderr,
670 apparently, which is not desirable.
672 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
675 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
678 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
680 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
681 authenticated or an ident call has been made. Suppress the default
682 values for $authenticated_id and $authenticated_sender (but permit -oMai
683 and -oMas) when testing with -bh.
685 PH/24 Re-jigged the order of the tests in the default configuration so that the
686 tests for valid domains and recipients precede the DNS black list and CSA
687 tests, on the grounds that those ones are more expensive.
689 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
690 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
691 This bug exists in every version of Exim that I still have, right back to
694 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
695 However, an attempt to turn on case-sensitivity in a regex key by
696 including (?-i) didn't work because the subject string was already
697 lowercased, and the effects were non-intuitive. It turns out that a
698 one-line patch can be used to allow (?-i) to work as expected.
704 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
705 other effects) broke the use of negated acl sub-conditions.
707 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
710 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
711 "Deny" causes Exim to reject the incoming connection with a 554 error.
712 Unfortunately, if there is a major crisis, such as a disk failure,
713 tcp-wrappers gives "deny", whereas what one would like would be some
714 kind of temporary error. A kludge has been added to help with this.
715 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
716 554 error is used if errno is still zero or contains ENOENT (which occurs
717 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
720 PH/03 Add -lutil to the default FreeBSD LIBS setting.
722 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
723 errors. Otherwise a message that provokes a temporary error (when other
724 messages do not) can cause a whole host to time out.
726 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
727 addresses were routed directly to files or pipes from a redirect router.
728 File deliveries just didn't batch; pipe deliveries might have suffered
731 PH/06 A failure to get a lock for a hints database would erroneously always say
732 "Failed to get write lock", even when it was really a read lock.
734 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
735 of 0600. This has been changed to use the value of the lockfile_mode
736 option (which defaults to 0600).
738 PH/08 Applied small patch from the Sieve maintainer.
740 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
741 folder from quota calculations, a direct delivery into this folder messed
742 up the contents of the maildirsize file. This was because the regex was
743 used only to exclude .Trash (or whatever) when the size of the mailbox
744 was calculated. There was no check that a delivery was happening into an
745 excluded directory. This bug has been fixed by ignoring all quota
746 processing for deliveries into excluded directories.
748 PH/10 Added the maildirfolder_create_regex option to appendfile.
754 PH/01 The code for finding all the local interface addresses on a FreeBSD
755 system running IPv6 was broken. This may well have applied to all BSD
756 systems, as well as to others that have similar system calls. The broken
757 code found IPv4 interfaces correctly, but gave incorrect values for the
758 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
759 that it would not match correctly against @[] and not recognize the IPv6
762 PH/02 The ipliteral router was not recognizing addresses of the form user@
763 [ipv6:....] because it didn't know about the "ipv6:" prefix.
765 PH/03 Added disable_ipv6.
767 PH/04 Changed $reply_address to use the raw form of the headers instead of the
768 decoded form, because it is most often used to construct To: headers
769 lines in autoreplies, and the decoded form may well be syntactically
770 invalid. However, $reply_address has leading white space removed, and all
771 newlines turned into spaces so that the autoreply transport does not
774 PH/05 If group was specified without a user on a router, and no group or user
775 was specified on a transport, the group from the router was ignored.
777 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
778 for visible compile-time settings that can be used to change these
779 numbers, for those that want even more. Backwards compatibility with old
780 spool files has been maintained. However, going back to a previous Exim
781 release will lost any variables that are in spool files.
783 PH/07 Two small changes when running in the test harness: increase delay when
784 passing a TCP/IP connection to a new process, in case the original
785 process has to generate a bounce, and remove special handling of
786 127.0.0.2 (sic), which is no longer necessary.
788 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
789 be the same on different OS.
791 PH/09 Moved a debug statement in filter processing to avoid a race problem when
794 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
795 whether --show-vars was specified or not
797 JJ/02 exipick: Added support for new ACL variable spool format introduced
800 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
801 syntactically invalid From: or Reply-to: line, and a filter used this to
802 generate an autoreply, and therefore failed to obtain an address for the
803 autoreply, Exim could try to deliver to a non-existent relative file
804 name, causing unrelated and misleading errors. What now happens is that
805 it logs this as a hard delivery error, but does not attempt to create a
808 PH/11 The exinext utility has a -C option for testing purposes, but although
809 the given file was scanned by exinext itself; it wasn't being passed on
810 when Exim was called.
812 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
813 an end-of-file indication when reading a command response.
815 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
816 compiled. In many other places in Exim, IPv6 addresses are always
817 recognized, so I have changed this. It also means that IPv4 domain
818 literals of the form [IPV4:n.n.n.n] are now always recognized.
820 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
821 used if the router is not running as root, for example, when verifying at
822 ACL time, or when using -bh. The debugging output from this situation was
823 non-existent - all you got was a failure to exec. I have made two
826 (a) Failures to set uid/gid, the current directory, or a process leader
827 in a subprocess such as that created by queryprogram now generate
828 suitable debugging ouput when -d is set.
830 (b) The queryprogram router detects when it is not running as root,
831 outputs suitable debugging information if -d is set, and then runs
832 the subprocess without attempting to change uid/gid.
834 PH/15 Minor change to Makefile for building test_host (undocumented testing
837 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
838 additional section of a DNS packet that returns MX or SRV records.
839 Instead, it always explicitly searches for A/AAAA records. This avoids
840 major problems that occur when a DNS server includes only records of one
841 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
842 fixed another bug: if SRV records were looked up and the corresponding
843 address records were *not* found in the additional section, the port
844 values from the SRV records were lost.
846 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
847 using the correct key (the original address) when searching the retry
848 rules in order to find which one to use for generating the retry hint.
850 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
851 adding the default one. Similarly, if it contains a Reply-To: header, the
852 errors_reply_to option, if set, is not used.
854 PH/19 When calculating a retry time, Exim used to measure the "time since
855 failure" by looking at the "first failed" field in the retry record. Now
856 it does not use this if it is later than than the arrival time of the
857 message. Instead it uses the arrival time. This makes for better
858 behaviour in cases where some deliveries succeed, thus re-setting the
859 "first failed" field. An example is a quota failure for a huge message
860 when small messages continue to be delivered. Without this change, the
861 "time since failure" will always be short, possible causing more frequent
862 delivery attempts for the huge message than are intended.
863 [Note: This change was subsequently modified - see PH/04 for 4.62.]
865 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
866 $1, $2, $3) because the numerical variables can be reset during some
867 expansion items (e.g. "match"), thereby losing the authentication data.
869 PH/21 Make -bV show the size of off_t variables so that the test suite can
870 decide whether to run tests for quotas > 2G.
872 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
873 mailbox_size, and mailbox_filecount in the appendfile transport. If a
874 filecount value is greater than 2G or if a quota value is greater than 2G
875 on a system where the size of off_t is not greater than 4, a panic error
878 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
879 never match. The debug and -bh output now contains an explicit error
880 message indicating a malformed IPv4 address or mask.
882 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
883 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
886 PH/25 Do not write to syslog when running in the test harness. The only
887 occasion when this arises is a failure to open the main or panic logs
888 (for which there is an explicit test).
890 PH/26 Added the /no_tell option to "control=freeze".
892 PH/27 If a host name lookup failed very early in a connection, for example, if
893 the IP address matched host_lookup and the reverse lookup yielded a name
894 that did not have a forward lookup, an error message of the form "no IP
895 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
896 could be logged. Now it outputs the IP address instead of "NULL".
898 PH/28 An enabling patch from MH: add new function child_open_exim2() which
899 allows the sender and the authenticated sender to be set when
900 submitting a message from within Exim. Since child_open_exim() is
901 documented for local_scan(), the new function should be too.
903 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
904 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
905 results in an empty string is now treated as unset.
907 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
909 PH/31 Added sender_verify_fail logging option.
911 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
912 needed by RFC 822 but not by RFC 2822 was commented out. I have now
913 tidied the source and removed it altogether.
915 PH/33 When a queue run was abandoned because the load average was too high, a
916 log line was always written; now it is written only if the queue_run log
917 selector is set. In addition, the log line for abandonment now contains
918 information about the queue run such as the pid. This is always present
919 in "start" and "stop" lines but was omitted from the "abandon" line.
921 PH/34 Omit spaces between a header name and the colon in the error message that
922 is given when verify = headers_syntax fails (if there are lots of them,
923 the message gets confusing).
925 PH/35 Change the default for dns_check_names_pattern to allow slashes within
926 names, as there are now some PTR records that contain slashes. This check
927 is only to protect against broken name servers that fall over on strange
928 characters, so the fact that it applies to all lookups doesn't matter.
930 PH/36 Now that the new test suite is complete, we can remove some of the
931 special code in Exim that was needed for the old test suite. For example,
932 sorting DNS records because real resolvers return them in an arbitrary
933 order. The new test suite's fake resolver always returns records in the
936 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
937 bounces) except when queue_only is set, to avoid logging races between
938 the different processes.
940 PH/38 Panic-die if .include specifies a non-absolute path.
942 PH/39 A tweak to the "H" retry rule from its user.
944 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
945 a label. They prevented compilation on older perls.
947 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
948 a warning to be raised on newish perls.
950 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
951 on queue. Changes to match documented behaviour of showing count of
952 messages matching specified criteria.
954 PH/40 Changed the default ident timeout from 30s to 5s.
956 PH/41 Added support for the use of login_cap features, on those BSD systems
957 that have them, for controlling the resources used by pipe deliveries.
959 PH/42 The content-scanning code uses fopen() to create files in which to put
960 message data. Previously it was not paying any attention to the mode of
961 the files. Exim runs with umask(0) because the rest of the code creates
962 files with open(), and sets the required mode explicitly. Thus, these
963 files were ending up world-writeable. This was not a big issue, because,
964 being within the spool directory, they were not world-accessible. I have
965 created a function called modefopen, which takes an additional mode
966 argument. It sets umask(777), creates the file, chmods it to the required
967 mode, then resets the umask. All the relevant calls to fopen() in the
968 content scanning code have been changed to use this function.
970 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
971 to 24 hours. This avoids potential overflow problems when processing G
972 and H retry rules. I suspect nobody ever tinkers with this value.
974 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
976 PH/45 When the plaintext authenticator is running as a client, the server's
977 challenges are checked to ensure they are valid base64 strings. By
978 default, the authentication attempt is cancelled if an invalid string is
979 received. Setting client_ignore_invalid_base64 true ignores these errors.
980 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
981 they are received. Thus, the responses can be made to depend on the
982 challenges. If an invalid string is ignored, an empty string is placed in
985 PH/46 Messages that are created by the autoreply transport now contains a
986 References: header, in accordance with RFCs 2822 and 3834.
988 PH/47 Added authenticated_sender_force to the smtp transport.
990 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
992 PH/49 Installed latest patch from the Sieve maintainer.
994 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
995 was also set, the appendfile transport was unnecessarily scanning a
996 directory of message files (e.g. for maildir delivery) to find the count
997 of files (along with the size), even though it did not need this
998 information. It now does the scan only if it needs to find either the
999 size of the count of files.
1001 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1003 PH/52 Two bugs concerned with error handling when the smtp transport is
1006 (i) Exim was not creating retry information for temporary errors given
1007 for individual recipients after the DATA command when the smtp transport
1008 was used in LMTP mode. This meant that they could be retried too
1009 frequently, and not timed out correctly.
1011 (ii) Exim was setting the flag that allows error details to be returned
1012 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1013 recipients that were returned after the DATA command.
1015 PH/53 This is related to PH/52, but is more general: for any failing address,
1016 when detailed error information was permitted to be returned to the
1017 sender, but the error was temporary, then after the final timeout, only
1018 "retry timeout exceeded" was returned. Now it returns the full error as
1019 well as "retry timeout exceeded".
1021 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1022 do this, and (what is worse) MTAs that accept it.
1024 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1025 will now be deprecated.
1027 PH/56 New os.c-cygwin from the Cygwin maintainer.
1029 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1030 formats (previously only available in exim formats via -bpr, -bpru,
1031 and -bpra. Now also available in native and exiqgrep formats)
1033 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1034 with very large, slow to parse queues
1036 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1038 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1040 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1041 responses to authentication challenges, though it was showing the
1042 challenges; (ii) I've removed the CR characters from the debug output for
1045 PH/58 Allow for the insertion of a newline as well as a space when a string
1046 is turned into more than one encoded-word during RFC 2047 encoding. The
1047 Sieve code now uses this.
1049 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1050 data_4xx, lost_connection, tls_required.
1052 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1053 message was being sent as an SMTP response.
1055 PH/61 Add -l and -k options to exicyclog.
1057 PH/62 When verifying, if an address was redirected to one new address, so that
1058 verification continued, and the new address failed or deferred after
1059 having set something in $address_data, the value of $address_data was not
1060 passed back to the ACL. This was different to the case when no
1061 redirection occurred. The value is now passed back in both cases.
1063 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1064 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1065 use login_cap.h, so on its own it isn't the distinguishing feature. The
1066 new name refers directly to the setclassresources() function.
1068 PH/65 Added configuration files for NetBSD3.
1070 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1072 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1073 is preferred over IPv4.
1075 PH/68 The bounce_return_message and bounce_return_body options were not being
1076 honoured for bounces generated during the reception of non-SMTP messages.
1077 In particular, this applied to messages rejected by the ACL. This bug has
1078 been fixed. However, if bounce_return_message is true and bounce_return_
1079 body is false, the headers that are returned for a non-SMTP message
1080 include only those that have been read before the error was detected.
1081 (In the case of an ACL rejection, they have all been read.)
1083 PH/69 The HTML version of the specification is now built in a directory called
1084 spec_html instead of spec.html, because the latter looks like a path with
1085 a MIME-type, and this confuses some software.
1087 PH/70 Catch two compiler warnings in sieve.c.
1089 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1090 function verify_get_ident() calls ip_connect() to connect a socket, but
1091 if the "connect()" function timed out, ip_connect() used to close the
1092 socket. However, verify_get_ident() also closes the socket later, and in
1093 between Exim writes to the log, which may get opened at this point. When
1094 the socket was closed in ip_connect(), the log could get the same file
1095 descriptor number as the socket. This naturally causes chaos. The fix is
1096 not to close the socket in ip_connect(); the socket should be closed by
1097 the function that creates it. There was only one place in the code where
1098 this was missing, in the iplookup router, which I don't think anybody now
1099 uses, but I've fixed it anyway.
1101 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1102 well as to direct DNS lookups. Otherwise the handling of names in host
1103 lists is inconsistent and therefore confusing.
1109 PH/01 Two changes to the default runtime configuration:
1111 (1) Move the checks for relay_from_hosts and authenticated clients from
1112 after to before the (commented out) DNS black list checks.
1114 (2) Add control=submission to the relay_from_hosts and authenticated
1115 clients checks, on the grounds that messages accepted by these
1116 statements are most likely to be submissions.
1118 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1120 (1) Generate an error if the third argument for the ${prvs expansion is
1123 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1126 (3) Reset the variables that are obtained from the first argument of
1127 ${prvscheck and used in the second argument before leaving the code,
1128 because their memory is reclaimed, so using them afterwards may do
1131 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1132 one (it's much easier than Tom thought :-).
1134 (5) Because of (4), we can now allow for the use of $prvscheck_result
1135 inside the third argument.
1137 PH/03 For some reason, the default setting of PATH when running a command from
1138 a pipe transport was just "/usr/bin". I have changed it to
1141 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1142 anything to be listed in the output from -bV.
1144 PH/05 When a filter generated an autoreply, the entire To: header line was
1145 quoted in the delivery log line, like this:
1147 => >A.N.Other <ano@some.domain> <original@ddress> ...
1149 This has been changed so that it extracts the operative address. There
1150 may be more than one such address. If so, they are comma-separated, like
1153 => >ano@some.domain,ona@other.domain <original@ddress> ...
1155 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1156 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1157 not being looked up in the rDNS to get a host name, Exim was showing the
1158 IP address twice in Received: lines, even though the IP addresses were
1159 identical. For example:
1161 Received: from [1.2.3.4] (helo=[1.2.3.4])
1163 However, if the real host name was known, it was omitting the HELO data
1164 if it matched the actual IP address. This has been tidied up so that it
1165 doesn't show the same IP address twice.
1167 PH/07 When both +timestamp and +memory debugging was on, the value given by
1168 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1169 called by the debug printing, thereby overwriting the timestamp buffer.
1170 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1172 PH/08 When the original message was included in an autoreply transport, it
1173 always said "this is a copy of the message, including all the headers",
1174 even if body_only or headers_only was set. It now gives an appropriate
1177 PH/09 Applied a patch from the Sieve maintainer which:
1179 o fixes some comments
1180 o adds the (disabled) notify extension core
1181 o adds some debug output for the result of if/elsif tests
1182 o points to the current vacation draft in the documentation
1183 and documents the missing references header update
1187 o fixes a bug in processing the envelope test (when testing
1188 multiple envelope elements, the last element determinted the
1191 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1192 Electronic Mail") by including:
1194 Auto-submitted: auto-generated
1196 in the messages that it generates (bounce messages and others, such as
1197 warnings). In the case of bounce messages for non-SMTP mesages, there was
1198 also a typo: it was using "Auto_submitted" (underscore instead of
1199 hyphen). Since every message generated by Exim is necessarily in response
1200 to another message, thes have all been changed to:
1202 Auto-Submitted: auto-replied
1204 in accordance with these statements in the RFC:
1206 The auto-replied keyword:
1208 - SHOULD be used on messages sent in direct response to another
1209 message by an automatic process,
1211 - MUST NOT be used on manually-generated messages,
1213 - MAY be used on Delivery Status Notifications (DSNs) and Message
1214 Disposition Notifications (MDNs),
1216 - MUST NOT be used on messages generated by automatic or periodic
1217 processes, except for messages which are automatic responses to
1220 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1221 to the default Received: header definition.
1223 PH/12 Added log selector acl_warn_skipped (default on).
1225 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1226 variables because (a) they are in the wrong storage pool and (b) even if
1227 they were copied, it wouldn't work properly because of the caching.
1229 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1230 checking when decoding. Apparently there are clients that generate
1231 overlong encoded strings. Why am I not surprised?
1233 PH/15 If the first argument of "${if match_address" was not empty, but did not
1234 contain an "@" character, Exim crashed. Now it writes a panic log message
1235 and treats the condition as false.
1237 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1239 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1240 extension "envelope-auth". The code is finished and in agreement with
1241 other implementations, but there is no documentation so far and in fact,
1242 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1243 not changing the active code.
1245 Print executed "if" and "elsif" statements when debugging is used. This
1246 helps a great deal to understand what a filter does.
1248 Document more things not specified clearly in RFC3028. I had all this
1249 sorted out, when out of a sudden new issues came to my mind. Oops."
1251 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1254 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1255 canonical form (as documented). However, after a host name lookup from
1256 the IP address, check_host() was doing a simple string comparison with
1257 addresses acquired from the DNS when checking that the found name did
1258 have the original IP as one of its addresses. Since any found IPv6
1259 addresses are likely to be in abbreviated form, the comparison could
1260 fail. Luckily, there already exists a function for doing the comparison
1261 by converting both addresses to binary, so now that is used instead of
1262 the text comparison.
1264 PH/20 There was another similar case to PH/19, when a complete host name was
1265 given in a host list; looking up its IP address could give an abbreviated
1266 form, whereas the current host's name might or might not be abbreviated.
1267 The same fix has been applied.
1273 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1274 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1277 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1278 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1280 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1282 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1283 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1284 Also, for "Auto-Submitted", treat anything other than "no" as
1285 non-personal, in accordance with RFC 3834. (Previously it treated
1286 anything starting "auto-" as non-personal.)
1288 TF/01 The control=submission/name=... option had a problem with syntax
1289 errors if the name included a slash character. The /name= option
1290 now slurps the rest of the string, so it can include any characters
1291 but it must come last in the list of options (after /sender_retain
1294 PH/05 Some modifications to the interface to the fake nameserver for the new
1302 TK/01 Added the "success_on_redirect" address verification option. See
1303 NewStuff for rationale and an example.
1305 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1307 PH/02 Patch to exigrep to allow it to work on syslog lines.
1309 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1310 fread() to skip over the body file's header line, because in Cygwin the
1311 header line is locked and is inaccessible.
1313 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1314 co-exist for some time) to make it clear that it is the Exim ID that is
1315 referenced, not the Message-ID: header line.
1317 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1318 string_format() function, because snprintf() does not exist on all
1321 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1322 ${stat: expansion item.
1324 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1325 protocol synchronization error", to keep the pedants happy.
1327 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1328 well as for IRIX systems, when gcc is being used. See the host.c source
1331 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1333 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1336 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1337 transport and to the smtp transport in LMTP mode.
1339 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1341 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1342 run a filter in a subprocess. This could lead to confusion in subsequent
1343 lookups in the parent process. There should also be a search_tidyup() at
1344 the end of the subprocess.
1346 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1347 only if the host matched helo_try_verify_hosts, which caused the
1348 verification to occur when the EHLO/HELO command was issued. The ACL just
1349 tested the remembered result. Now, if a previous verification attempt has
1350 not happened, "verify = helo" does it there and then.
1352 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1354 TK/03 Fix log output including CR from clamd.
1356 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1357 exist provoked a memory error which could cause a segfault.
1359 PH/15 Installed PCRE 6.2
1361 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1363 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1364 of the problem. Specifically, suggested +O2 rather than +O1 for the
1367 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1369 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1370 record, the port was not correctly logged when the outgoing_port log
1371 selector was set (it logged the transort's default port).
1373 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1374 fallback_hosts, and "hosts" in the smtp transport.
1376 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1377 host errors such as "Connection refused".
1379 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1380 authentication with radiusclient 0.4.9:
1382 - Error returned from rc_read_config was caught wrongly
1383 - Username/password not passed on to radius server due to wrong length.
1385 The presumption is that some radiusclient API changes for 4.51/PH/17
1386 were not taken care of correctly. The code is still untested by me (my
1387 Linux distribution still has 0.3.2 of radiusclient), but it was
1388 contributed by a Radius user.
1390 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1391 expanding the "port" option of the smtp transport.
1393 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1394 while reading a MIME header. Thanks to Tom Hughes for a patch.
1396 PH/24 Include config.h inside local_scan.h so that configuration settings are
1399 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1400 that in an ACL for RCPT (for example), you can examine exactly what was
1403 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1404 commands, but it was not correctly comparing the address with the actual
1405 client host address. Thus, it would show the EHLO address in Received:
1406 header lines when this was not necessary.
1408 PH/27 Added the % operator to ${eval:}.
1410 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1411 it should be ignoring failures (because with -C, for example, it has lost
1412 privilege). It wasn't ignoring creation failures other than "already
1415 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1418 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1419 because an input line was too long, either on its own, or by virtue of
1420 too many continuations, the temporary file was not being removed, and the
1421 return code was incorrect.
1423 PH/31 Missing "BOOL" in function definition in filtertest.c.
1425 PH/32 Applied Sieve patches from the maintainer.
1427 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1429 PH/33 Added "verify = not_blind".
1431 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1432 Local/Makefile (with some defaults set). These are used in built scripts
1433 such as exicyclog, but they have never been used in the exim_install
1434 script (though there are many overriding facilities there). I have
1435 arranged that the exim_install script now takes note of these two
1438 PH/35 Installed configuration files for Dragonfly.
1440 PH/36 When a locally submitted message by a trusted user did not contain a
1441 From: header, and the sender address was obtained from -f or from an SMTP
1442 MAIL command, and the trusted user did not use -F to supply a sender
1443 name, $originator_name was incorrectly used when constructing a From:
1444 header. Furthermore, $originator_name was used for submission mode
1445 messages from external hosts without From: headers in a similar way,
1446 which is clearly wrong.
1448 PH/37 Added control=suppress_local_fixups.
1450 PH/38 When log_selector = +received_sender was set, and the addition of the
1451 sender made the log line's construction buffer exactly full, or one byte
1452 less than full, an overflow happened when the terminating "\n" was
1455 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1456 when the result of a list match is failure because a DNS lookup failed.
1458 PH/40 RM_COMMAND is now used in the building process.
1460 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1461 the "build-* directories that it finds.
1463 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1464 address in a domain literal was a prefix of an interface address.
1466 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1467 when verifying a sender address, unless rewrite_headers is false.
1469 PH/44 Wrote a long comment about why errors_to addresses are verified as
1470 recipients, not senders.
1472 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1473 the ratelimit ACL was added.
1475 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1477 PH/46 Added extra information about PostgreSQL errors to the error string.
1479 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1480 avoiding the need to install special zones in a real server. This is
1481 backwards compatible; if it can't find the fake resolver, it drops back.
1482 Thus, both old and new test suites can be run.
1484 TF/02 Added util/ratelimit.pl
1486 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1487 clock is set back in time.
1489 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1490 Candler <B.Candler@pobox.com>.
1492 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1493 OK for addresses that are the result of redirections.
1495 PH/48 A number of further additions for the benefit of the new test suite,
1496 including a fake gethostbyname() that interfaces to the fake DNS resolver
1499 TF/06 The fix for widen_domains has also been applied to qualify_single and
1500 search_parents which are the other dnslookup options that can cause
1503 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1506 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1508 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1509 service" if it encountered an MX record with an empty target hostname.
1510 The message is now "an MX or SRV record indicated no SMTP service".
1512 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1513 if the DNS of the sending site is misconfigured. This is quite a
1514 common situation. This change restores the behaviour of treating a
1515 helo verification defer as a failure.
1517 PH/51 If self=fail was set on a router, the bounce message did not include the
1518 actual error message.
1524 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1526 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1527 command itself ended in error, the underlying message about the transport
1528 filter timeout was being overwritten with the pipe command error. Now the
1529 underlying error message should be appended to the second error message.
1531 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1533 PH/02 Reduce the amount of output that "make" produces by default. Full output
1534 can still be requested.
1536 PH/03 The warning log line about a condition test deferring for a "warn" verb
1537 was being output only once per connection, rather than after each
1538 occurrence (because it was using the same function as for successful
1539 "warn" verbs). This seems wrong, so I have changed it.
1541 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1542 it should not have, which might have caused a crash in the right
1543 circumstances, but probably never did.
1545 PH/04 Installed a modified version of Tony Finch's patch to make submission
1546 mode fix the return path as well as the Sender: header line, and to
1547 add a /name= option so that you can make the user's friendly name appear
1550 TF/03 Added the control = fakedefer ACL modifier.
1552 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1553 Mark Lowes for thorough testing.
1555 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1557 TK/03 Merged latest SRS patch from Miles Wilton.
1559 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1560 with the definition in sysexits.h (which is #included earlier).
1561 Fortunately, Exim does not actually use EX_OK. The code used to try to
1562 preserve the sysexits.h value, by assumimg that macro definitions were
1563 scanned for macro replacements. I have been disabused of this notion,
1564 so now the code just undefines EX_OK before #including unistd.h.
1566 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1567 in the smtp transport. When a block could not be written in a single
1568 write() function, the timeout was being re-applied to each part-write.
1569 This seems wrong - if the receiver was accepting one byte at a time it
1570 would take for ever. The timeout is now adjusted when this happens. It
1571 doesn't have to be particularly precise.
1573 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1574 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1576 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1577 without a domain if the check to <postmaster@domain> fails.
1579 SC/01 Eximstats: added -xls and the ability to specify output files
1580 (patch written by Frank Heydlauf).
1582 SC/02 Eximstats: use FileHandles for outputing results.
1584 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1586 SC/04 Eximstats: fixed display of large numbers with -nvr option
1588 SC/05 Eximstats: fixed merging of reports with empty tables.
1590 SC/06 Eximstats: added the -include_original_destination flag
1592 SC/07 Eximstats: removed tabs and trailing whitespace.
1594 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1596 TK/06 MBOX spool code: Add real "From " MBOX separator line
1597 so the .eml file is really in mbox format (even though
1598 most programs do not really care). Patch from Alex Miller.
1600 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1601 The latter is generated from $received_to and is only set if the
1602 message has one envelope recipient. SA can use these headers,
1603 obviously out-of-the-box. Patch from Alex Miller.
1605 PH/08 The ${def test on a variable was returning false if the variable's
1606 value was "0", contrary to what the specification has always said!
1607 The result should be true unless the variable is empty.
1609 PH/09 The syntax error of a character other than { following "${if
1610 def:variable_name" (after optional whitespace) was not being diagnosed.
1611 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1612 accidental colon was present, for example, could give incorrect results.
1614 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1615 result is used (not including appendfile, where other changes are about
1618 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1619 This involved changing a lot of size variables from int to off_t. It
1620 should work with maildirs and everything.
1622 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1623 spamd dying while we are connected to it.
1625 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1628 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1629 function for BDB 4.3.
1631 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1633 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1634 the sources for PCRE have been re-arranged and more files are now
1637 PH/15 The code I had for printing potentially long long variables in PH/11
1638 above was not the best (it lost precision). The length of off_t variables
1639 is now inspected at build time, and an appropriate printing format (%ld
1640 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1641 to be "long long int" or "long int". This is needed for the internal
1642 formatting function string_vformat().
1644 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1645 the configuration file to be ":syslog", then the script "guesses" where
1646 the logs files are, rather than using the compiled in default. In our
1647 case the guess is not the same as the compiled default, so the script
1648 suddenly stopped working when I started to use syslog. The patch checks
1649 to see if log_file_path is "". If so, it attempts to read it from exim
1650 with no configuration file to get the compiled in version, before it
1651 falls back to the previous guessing code."
1653 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1654 implementing BATV in an Exim configuration. See NewStuff for the gory
1657 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1658 Makefile that are specific to HP-UX.
1660 PH/18 If the "use_postmaster" option was set for a recipient callout together
1661 with the "random" option, the postmaster address was used as the MAIL
1662 FROM address for the random test, but not for the subsequent recipient
1663 test. It is now used for both.
1665 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1666 patch removes a few documentation additions to RFC 3028, because the
1667 latest draft now contains them. It adds the new en;ascii-case comparator
1668 and a new error check for 8bit text in MIME parts. Comparator and
1669 require names are now matched exactly. I enabled the subaddress
1670 extension, but it is not well tested yet (read: it works for me)."
1672 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1673 rework some of the code of TK/09 above to avoid the hardwired use of
1674 "%lld" and "long long". Replaced the call to snprintf() with a call to
1677 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1678 records point to non-existent hosts", "retry timeout exceeded", and
1679 "retry time not reached for any host after a long failure period".
1681 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1682 experimental DomainKeys support:
1684 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1685 (2) On an error such as an illegally used "control", the wrong name for
1686 the control was given.
1688 These problems did NOT occur unless DomainKeys support was compiled.
1690 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1692 PH/24 Added ${if match_ip condition.
1694 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1695 databases so that it will be absolutely obvious if a crash occurs in the
1696 DB library. This is a regular occurrence (often caused by mis-matched
1699 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1700 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1701 picked up on a user's system that detects such things. There doesn't seem
1702 to be a gcc warning option for this - only an attribute that has to be
1703 put on the function's prototype. It seems that in Fedora Core 4 they have
1704 set this on a number of new functions. No doubt there will be more in due
1707 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1708 specify a transport. However, if an address that was verified by such a
1709 router was the subject of a callout, Exim crashed because it tried to
1710 read the rcpt_include_affixes from the non-existent transport. Now it
1711 just assumes that the setting of that option is false. This bug was
1712 introduced by 4.51/PH/31.
1714 PH/28 Changed -d+all to exclude +memory, because that information is very
1715 rarely of interest, but it makes the output a lot bigger. People tend to
1716 do -d+all out of habit.
1718 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1719 code in os-type was giving problems when libc.so lives in lib64, like on
1722 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1723 aren't the modern standard, and it seems that some systems' include files
1724 don't always have them. Exim was already checking for some of the newer
1725 ones like T_AAAA, and defining it itself. I've added checks for all the
1726 record types that Exim uses.
1728 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1729 not automatically generating a new one, as it is supposed to. This
1730 prevented TLS from working. If the file did exist, but contained invalid
1731 data, a new version was generated, as expected. It was only the case of a
1732 non-existent file that was broken.
1734 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1735 with a change in libdomainkeys > 0.64.
1737 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1738 from DNS. If the selector record carries the flag, it now has
1739 precedence over the domain-wide flag.
1741 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1743 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1744 the use of a port name that isn't defined in /etc/services) occurred, the
1745 message was deferred as in a normal delivery, and thus remained on the
1746 spool, instead of being failed because of the mua_wrapper setting. This
1747 is now fixed, and I tidied up some of the mua_wrapper messages at the
1750 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1751 the messages in a hash of arrays rather than using individual hashes.
1752 This is a bit cleaner and results in dramatic memory savings, albeit
1753 at a slight CPU cost.
1755 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1756 as requested by Marc Sherman.
1758 SC/10 Eximstats: added histograms for user specified patterns as requested
1761 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1763 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1764 fopen() in the content-scanning modules that did not already have it.
1770 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1771 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1773 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1775 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1777 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1778 to test Sieve filters that use "vacation".
1780 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1781 that changes the way the GnuTLS parameters are stored in the cache file.
1782 The new format can be generated externally. For backward compatibility,
1783 if the data in the cache doesn't make sense, Exim assumes it has read an
1784 old-format file, and it generates new data and writes a new file. This
1785 means that you can't go back to an older release without removing the
1788 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1789 work if there are any delivery delays because "one_time" forces the
1790 parent to be marked "delivered", so its unseen clone is never tried
1791 again. For this reason, Exim now forbids the simultaneous setting of
1794 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1795 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1796 entry if you want to know the details. The fix, however, neglected to
1797 consider the case when local delivery batching is involved. The test for
1798 "previously delivered" was not happening when checking to see if an
1799 address could be batched with a previous (undelivered) one; under
1800 certain circumstances this could lead to multiple deliveries to the same
1803 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1804 in its include files, and this causes problems building Exim.
1806 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1807 header_syntax) but Exim was just ignoring anything given after a slash.
1808 In particular, this caused confusion with an attempt to use "verify =
1809 reverse_host_lookup/defer_ok". An error is now given when options are
1810 supplied for verify items that do not have them. (Maybe reverse_host_
1811 lookup should have a defer_ok option, but that's a different point.)
1813 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1814 defined by RFC 821) to 2048, because there were problems with some AUTH
1815 commands, and RFC 1869 says the size should be increased for extended
1816 SMTP commands that take arguments.
1818 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1821 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1822 "unknown" error; now it says that the functionality isn't in the binary.
1824 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1825 an address' error message when a string expansion fails (syntax or
1826 whatever). Otherwise the password may appear in the log. Following change
1827 PH/42 below, there is no longer a chance of it appearing in a bounce
1830 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1832 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1833 of its addresses was ever tried. (Bugzilla bug #2.)
1835 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1836 the result incorrectly in the debug output. (It correctly added a newline
1837 to what was transported.)
1839 TF/01 Added $received_time.
1841 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1842 commented out examples of how to interface to a virus scanner and to
1843 SpamAssassin. Also added commented examples of av_scanner and
1844 spamd_address settings.
1846 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1847 and controls are allowed in which ACLs. There were a couple of minor
1848 errors. Some of the entries in the conditions table (which is a table of
1849 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1850 negation of where the condition IS allowed.
1852 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1854 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1855 header file does not have a version number, so I've had to invent a new
1856 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1857 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1858 radiusclient), but it was contributed by a Radius user.
1860 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1861 files or directories, for OpenSSL.
1863 PH/19 When an Exim process that is running as root has to create an Exim log
1864 file, it does so in a subprocess that runs as exim:exim so as to get the
1865 ownership right at creation (otherwise, other Exim processes might see
1866 the file with the wrong ownership). There was no test for failure of this
1867 fork() call, which would lead to the process getting stuck as it waited
1868 for a non-existent subprocess. Forks do occasionally fail when resources
1869 run out. I reviewed all the other calls to fork(); they all seem to check
1872 PH/20 When checking for unexpected SMTP input at connect time (before writing
1873 the banner), Exim was not dealing correctly with a non-positive return
1874 from the read() function. If the client had disconnected by this time,
1875 the result was a log entry for a synchronization error with an empty
1876 string after "input=" when read() returned zero. If read() returned -1
1877 (an event I could not check), uninitialized data bytes were printed.
1878 There were reports of junk text (parts of files, etc) appearing after
1881 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1883 PH/22 Added support for macro redefinition, and (re)definition in between
1884 driver and ACL definitions.
1886 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1887 forgetting to use the resulting value; it was using the unexpanded value.
1889 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1890 hadn't been configured. The fix is from Juergen Kreileder, who
1891 understands it better than I do:
1893 "Here's what I see happening with three configured cyrus_sasl
1894 authenticators configured (plain, login, cram-md5):
1896 On startup auth_cyrus_sasl_init() gets called for each of these.
1897 This means three calls to sasl_listmech() without a specified mech_list.
1898 => SASL tests which mechs of all available mechs actually work
1899 => three warnings about OTP not working
1900 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1902 With the patch, sasl_listmech() also gets called three times. But now
1903 SASL's mech_list option is set to the server_mech specified in the the
1904 authenticator. Or in other words, the answer from sasl_listmech()
1905 gets limited to just the mech you're testing for (which is different
1907 => the return list contains just 'plain' or 'login', 'cram-md5' or
1908 nothing depending on the value of ob->server_mech.
1910 I've just tested the patch: Authentication still works fine,
1911 unavailable mechs specified in the exim configuration are still
1912 caught, and the auth.log warnings about OTP are gone."
1914 PH/25 When debugging is enabled, the contents of the command line are added
1915 to the debugging output, even when log_selector=+arguments is not
1918 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1919 answer is "GNU", and only if the return is "GNU/something" is the answer
1922 PH/27 $acl_verify_message is now set immediately after the failure of a
1923 verification in an ACL, and so is available in subsequent modifiers. In
1924 particular, the message can be preserved by coding like this:
1926 warn !verify = sender
1927 set acl_m0 = $acl_verify_message
1929 Previously, $acl_verify_message was set only while expanding "message"
1930 and "log_message" when a very denied access.
1932 PH/28 Modified OS/os.c-Linux with
1934 -#ifndef OS_LOAD_AVERAGE
1935 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1937 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1938 nomenclature these days.)
1940 PH/29 Installed patch from the Sieve maintainer that adds the options
1941 sieve_useraddress and sieve_subaddress to the redirect router.
1943 PH/30 In these circumstances:
1944 . Two addresses routed to the same list of hosts;
1945 . First host does not offer TLS;
1946 . First host accepts first address;
1947 . First host gives temporary error to second address;
1948 . Second host offers TLS and a TLS session is established;
1949 . Second host accepts second address.
1950 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1951 and peerdn, if requested) that were in fact used only for the second
1954 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1955 attention to any local part prefix or suffix that was matched by the
1956 router that accepted the address. It now behaves in the same way as it
1957 does for delivery: the affixes are removed from the local part unless
1958 rcpt_include_affixes is set on the transport.
1960 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1961 timeout during the DATA phase of an incoming message.
1963 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1964 applied a patch sanctioned by the Sieve maintainer.
1966 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1967 the uid or gid is negative. A case of a negative gid caused this to be
1968 noticed. The fix allows for either to be negative.
1970 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1971 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1974 PH/36 The change PH/12 above was broken. Fixed it.
1976 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1977 the grounds that routing the same address twice would always produce the
1978 same answer. This might have been true once, but it is certainly no
1979 longer true now. Routing a child address may depend on the previous
1980 routing that produced that child. Some complicated redirection strategies
1981 went wrong when messages had multiple recipients, and made Exim's
1982 behaviour dependent on the order in which the addresses were given.
1984 I have moved the duplicate checking until after the routing is complete.
1985 Exim scans the addresses that are assigned to local and remote
1986 transports, and removes any duplicates. This means that more work will be
1987 done, as duplicates will always all be routed, but duplicates are
1988 presumably rare, so I don't expect this is of any significance.
1990 For deliveries to pipes, files, and autoreplies, the duplicate checking
1991 still happens during the routing process, since they are not going to be
1994 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1995 It corrects a timeout issue with spamd. This is Ian's comment: "The
1996 background is that sometimes spamd either never reads data from a
1997 connection it has accepted, or it never writes response data. The exiscan
1998 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1999 blindly assumes that writes won't block so it may never time out."
2001 PH/39 Allow G after quota size as well as K and M.
2003 PH/40 The value set for $authenticated_id in an authenticator may not contain
2004 binary zeroes or newlines because the value is written to log lines and
2005 to spool files. There was no check on this. Now the value is run through
2006 the string_printing() function so that such characters are converted to
2007 printable escape sequences.
2009 PH/41 $message_linecount is a new variable that contains the total number of
2010 lines in the message. Compare $body_linecount, which is the count for the
2013 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2014 bounce and delay warning messages, except in certain special cases, which
2017 (a) An SMTP error message from a remote host;
2018 (b) A message specified in a :fail: redirection;
2019 (c) A message specified in a "fail" command in a system filter;
2020 (d) A message specified in a FAIL return from the queryprogram router;
2021 (e) A message specified by the cannot_route_message router option.
2023 In these cases only, Exim does include the error details in bounce and
2024 warning messages. There are also a few cases where bland messages such
2025 as "unrouteable address" or "local delivery error" are given.
2027 PH/43 $value is now also set for the "else" part of a ${run expansion.
2029 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2030 being worked on, but at least Exim now implements the latest version to
2033 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2034 process to complete was treated as a delivery failure, a timeout while
2035 writing the message to the pipe was logged, but erroneously treated as a
2036 successful delivery. Such timeouts include transport filter timeouts. For
2037 consistency with the overall process timeout, these timeouts are now
2038 treated as errors, giving rise to delivery failures by default. However,
2039 there is now a new Boolean option for the pipe transport called
2040 timeout_defer, which, if set TRUE, converts the failures into defers for
2041 both kinds of timeout. A transport filter timeout is now identified in
2044 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2045 systems where "make" and "gmake" are different, calling "gmake" at top
2046 level broke things. I've arranged for the value of $(MAKE) to be passed
2047 from the Makefile to this script so that it can call the same version of
2051 A note about Exim versions 4.44 and 4.50
2052 ----------------------------------------
2054 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2055 changes of various kinds. As a consequence, a big documentation update was
2056 needed. This delayed the release for rather longer than seemed good, especially
2057 in the light of a couple of (minor) security issues. Therefore, the changes
2058 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2059 release. So 4.44 and 4.50 are in effect two different branches that both start
2062 I have left the 4.50 change log unchanged; it contains all the changes since
2063 4.43. The change log for 4.44 is below; many of its items are identical to
2064 those for 4.50. This seems to be the most sensible way to preserve the
2065 historical information.
2071 1. Minor wording change to the doc/README.SIEVE file.
2073 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2074 computation of the current number of files was incorrect.
2076 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2077 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2080 4. Give more explanation in the error message when the command for a transport
2081 filter fails to execute.
2083 5. There are several places where Exim runs a non-Exim command in a
2084 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2085 was being done only for the command run by the queryprogram router. It is
2086 now done for all such subprocesses. The other cases are: ${run, transport
2087 filters, and the commands run by the lmtp and pipe transports.
2089 6. Added CONFIGURE_GROUP build-time option.
2091 7. Some older OS have a limit of 256 on the maximum number of file
2092 descriptors. Exim was using setrlimit() to set 1000 as a large value
2093 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2094 systems. I've change it so that if it can't get 1000, it tries for 256.
2096 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2097 was an oversight, and furthermore, ever since the addition of extra
2098 controls (e.g. 4.43/32), the checks on when to allow different forms of
2099 "control" were broken. There should now be diagnostics for all cases when a
2100 control that does not make sense is encountered.
2102 9. Added the /retain_sender option to "control=submission".
2104 10. $recipients is now available in the predata ACL (oversight).
2106 11. Tidy the search cache before the fork to do a delivery from a message
2107 received from the command line. Otherwise the child will trigger a lookup
2108 failure and thereby defer the delivery if it tries to use (for example) a
2109 cached ldap connection that the parent has called unbind on.
2111 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2112 of $address_data from the recipient verification was clobbered by the
2113 sender verification.
2115 13. The value of address_data from a sender verification is now available in
2116 $sender_address_data in subsequent conditions in the ACL statement.
2118 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2120 15. Added a new option "connect=<time>" to callout options, to set a different
2123 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2124 was its contents. (It was OK if the option was not defined at all.)
2126 17. A "Completed" log line is now written for messages that are removed from
2127 the spool by the -Mrm option.
2129 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2130 information about exactly what failed.
2132 19. Added -dd to debug only the daemon process.
2134 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2135 handles timeouts, both on the server side and network timeouts. Renamed the
2136 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2138 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2139 It is now set to "smtps".
2141 22. $host_address is now set to the target address during the checking of
2142 ignore_target_hosts.
2144 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2145 being passed; this would have caused $sender_host_name to have been used if
2146 matching the list had actually called for a host name (not very likely,
2147 since this list is usually IP addresses). A host name is now passed as
2150 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2151 code that specifies a non-restarting handler (typically sigaction() in
2152 modern systems) in an attempt to fix a rare and obscure crash bug.
2154 25. Narrowed the window for a race in the daemon that could cause it to ignore
2155 SIGCHLD signals. This is not a major problem, because they are used only to
2156 wake it up if nothing else does.
2158 26. A malformed maildirsize file could cause Exim to calculate negative values
2159 for the mailbox size or file count. Odd effects could occur as a result.
2160 The maildirsize information is now recalculated if the size or filecount
2163 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2164 support for a long time. Removed HAVE_SYS_VFS_H.
2166 28. Installed the latest version of exipick from John Jetmore.
2168 29. In an address list, if the pattern was not a regular expression, an empty
2169 subject address (from a bounce message) matched only if the pattern was an
2170 empty string. Non-empty patterns were not even tested. This was the wrong
2171 because it is perfectly reasonable to use an empty address as part of a
2172 database query. An empty address is now tested by patterns that are
2173 lookups. However, all the other forms of pattern expect the subject to
2174 contain a local part and a domain, and therefore, for them, an empty
2175 address still always fails if the pattern is not itself empty.
2177 30. Exim went into a mad DNS loop when attempting to do a callout where the
2178 host was specified on an smtp transport, and looking it up yielded more
2179 than one IP address.
2181 31. Re-factored the code for checking spool and log partition space into a
2182 function that finds that data and another that does the check. The former
2183 is then used to implement four new variables: $spool_space, $log_space,
2184 $spool_inodes, and $log_inodes.
2186 32. The RFC2047 encoding function was originally intended for short strings
2187 such as real names; it was not keeping to the 75-character limit for
2188 encoded words that the RFC imposes. It now respects the limit, and
2189 generates multiple encoded words if necessary. To be on the safe side, I
2190 have increased the buffer size for the ${rfc2047: expansion operator from
2193 33. It is now permitted to omit both strings after an "if" condition; if the
2194 condition is true, the result is "true". As before, when the second string
2195 is omitted, a false condition yields an empty string. This makes it less
2196 cumbersome to write custom ACL and router conditions.
2198 34. Failure to deliver a bounce message always caused it to be frozen, even if
2199 there was an errors_to setting on the router. The errors_to setting is now
2202 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2203 canonical form (fully expanded) before being placed in
2204 $sender_host_address.
2206 36. The table in the code that translates DNS record types into text (T_A to
2207 "A" for instance) was missing entries for NS and CNAME. It is just possible
2208 that this could have caused confusion if both these types were looked up
2209 for the same domain, because the text type is used as part of Exim's
2210 per-process caching. But the chance of anyone hitting this buglet seems
2213 37. The dnsdb lookup has been extended in a number of ways.
2215 (1) There is a new type, "zns", which walks up the domain tree until it
2216 finds some nameserver records. It should be used with care.
2218 (2) There is a new type, "mxh", which is like "mx" except that it returns
2219 just the host names, not the priorities.
2221 (3) It is now possible to give a list of domains (or IP addresses) to be
2222 looked up. The behaviour when one of the lookups defers can be
2223 controlled by a keyword.
2225 (4) It is now possible to specify the separator character for use when
2226 multiple records are returned.
2228 38. The dnslists ACL condition has been extended: it is now possible to supply
2229 a list of IP addresses and/or domains to be looked up in a particular DNS
2232 39. Added log_selector=+queue_time_overall.
2234 40. When running the queue in the test harness, wait just a tad after forking a
2235 delivery process, to get repeatability of debugging output.
2237 41. Include certificate and key file names in error message when GnuTLS fails
2238 to set them up, because the GnuTLS error message doesn't include the name
2239 of the failing file when there is a problem reading it.
2241 42. Allow both -bf and -bF in the same test run.
2243 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2245 44. The "Exiscan patch" is now merged into the mainline Exim source.
2247 45. Sometimes the final signoff response after QUIT could fail to get
2248 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2249 < 0 before doing a fflush(). This bug looks as though it goes back to the
2250 introduction of TLS in release 3.20, but "sometimes" must have been rare
2251 because the tests only now provoked it.
2253 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2254 (this can affect the format of dates).
2256 47. exim_tidydb, when checking for the continued existence of a message for
2257 which it has found a message-specific retry record, was not finding
2258 messages that were in split spool directories. Consequently, it was
2259 deleting retry records that should have stayed in existence.
2261 48. Steve fixed some bugs in eximstats.
2263 49. The SPA authentication driver was not abandoning authentication and moving
2264 on to the next authenticator when an expansion was forced to fail,
2265 contradicting the general specification for all authenticators. Instead it
2266 was generating a temporary error. It now behaves as specified.
2268 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2269 (the order specifies the preference for clients). The order is now AES256,
2270 AES128, 3DES, ARCFOUR128.
2272 51. Small patch to Sieve code - explicitly set From: when generating an
2275 52. Exim crashed if a remote delivery caused a very long error message to be
2276 recorded - for instance if somebody sent an entire SpamAssassin report back
2277 as a large number of 550 error lines. This bug was coincidentally fixed by
2278 increasing the size of one of Exim's internal buffers (big_buffer) that
2279 happened as part of the Exiscan merge. However, to be on the safe side, I
2280 have made the code more robust (and fixed the comments that describe what
2283 53. Now that there can be additional text after "Completed" in log lines (if
2284 the queue_time_overall log selector is set), a one-byte patch to exigrep
2285 was needed to allow it to recognize "Completed" as not the last thing in
2288 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2289 patch that reportedly fixes this has been added. I am not expert enough to
2290 create a test for it. This is what the patch creator wrote:
2292 "I found a little strange behaviour of ldap code when working with
2293 Windows 2003 AD Domain, where users was placed in more than one
2294 Organization Units. When I tried to give exim partial DN, the exit code
2295 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2296 But simultaneously result of request was absolutely normal ldap result,
2297 so I produce this patch..."
2299 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2300 REFERENCE, so I have modified the code to exclude the patch when that macro
2303 55. Some experimental protocols are using DNS PTR records for new purposes. The
2304 keys for these records are domain names, not reversed IP addresses. The
2305 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2306 leaves it alone. Component reversal etc. now happens only for IP addresses.
2309 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2311 57. Double the size of the debug message buffer (to 2048) so that more of very
2312 long debug lines gets shown.
2314 58. The exicyclog utility now does better if the number of log files to keep
2315 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2317 59. Two changes related to the smtp_active_hostname option:
2319 (1) $smtp_active_hostname is now available as a variable.
2320 (2) The default for smtp_banner uses $smtp_active_hostname instead
2321 of $primary_hostname.
2323 60. The host_aton() function is supposed to be passed a string that is known
2324 to be a valid IP address. However, in the case of IPv6 addresses, it was
2325 not checking this. This is a hostage to fortune. Exim now panics and dies
2326 if the condition is not met. A case was found where this could be provoked
2327 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2328 components; fortuitously, this particular loophole had already been fixed
2329 by change 4.50/55 above.
2331 If there are any other similar loopholes, the new check in host_aton()
2332 itself should stop them being exploited. The report I received stated that
2333 data on the command line could provoke the exploit when Exim was running as
2334 exim, but did not say which command line option was involved. All I could
2335 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2336 running as the user.
2339 61. There was a buffer overflow vulnerability in the SPA authentication code
2340 (which came originally from the Samba project). I have added a test to the
2341 spa_base64_to_bits() function which I hope fixes it.
2344 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2345 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2347 63. The daemon start-up calls getloadavg() while still root for those OS that
2348 need the first call to be done as root, but it missed one case: when
2349 deliver_queue_load_max is set with deliver_drop_privilege. This is
2350 necessary for the benefit of the queue runner, because there is no re-exec
2351 when deliver_drop_privilege is set.
2353 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2354 This has been fixed.
2356 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2357 was in use, was not putting the data itself into the right store pool;
2358 consequently, it could be overwritten for a subsequent message in the same
2359 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2362 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2364 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2365 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2366 the function were treating the return as a boolean value, which happened to
2367 work because 0=false and not-0=true, but is not correct code.
2369 68. The host_aton() function was not handling scoped IPv6 addresses (those
2370 with, for example, "%eth0" on the end) correctly.
2372 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2373 negated items (that is, ~something) in unsigned ints. Some compilers
2374 apparently mutter when there is no cast.
2376 70. If an address verification called from an ACL failed, and did not produce a
2377 user-specific message (i.e. there was only a "system" message), nothing was
2378 put in $acl_verify_message. In this situation, it now puts the system
2381 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2382 session; change 4.31/43 added the unwanted input to the log line - except
2383 that it did not do this in the start of session case. It now does.
2385 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2386 This is wrong and can cause the other end to generate a synchronization
2387 error if it is another Exim or anything else that does the synchronization
2388 check. A QUIT command is no longer sent after a timeout.
2390 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2391 during host lookups.
2393 74. The defer_ok option of callout verification was not working if it was used
2394 when verifying addresses in header lines, that is, for this case:
2396 verify = header_sender/callout=defer_ok
2398 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2399 those file descriptors could be used for SMTP connections. If anything
2400 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2401 could be sent to the SMTP client, causing chaos. The daemon now opens
2402 stdin, stdout, and stderr to /dev/null when it puts itself into the
2405 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2406 log by default. The user can override this with suitable Perl magic.
2408 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2409 the log message when discard triggers, was not working for the DATA ACL or
2410 for the non-SMTP ACL.
2412 78. Error message wording change in sieve.c.
2414 79. If smtp_accept_max_per_host was set, the number of connections could be
2415 restricted to fewer than expected, because the daemon was trying to set up
2416 a new connection before checking whether the processes handling previous
2417 connections had finished. The check for completed processes is now done
2418 earlier. On busy systems, this bug wouldn't be noticed because something
2419 else would have woken the daemon, and it would have reaped the completed
2422 80. If a message was submitted locally by a user whose login name contained one
2423 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2424 It caused a spool format error. I have fixed the spool reading code. A
2425 related problem was that the "from" clause in the Received: line became
2426 illegal because of the space(s). It is now covered by ${quote_local_part.
2428 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2431 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2433 83. Patch from Sieve maintainer to fix unterminated string problem in
2434 "vacation" handling.
2436 84. Some minor changes to the Linux configuration files to help with other
2437 OS variants using glibc.
2439 85. One more patch for Sieve to update vacation handling to latest spec.
2442 ----------------------------------------------------
2443 See the note above about the 4.44 and 4.50 releases.
2444 ----------------------------------------------------
2450 1. Change 4.43/35 introduced a bug that caused file counts to be
2451 incorrectly computed when quota_filecount was set in an appendfile
2454 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2455 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2458 3. Give more explanation in the error message when the command for a transport
2459 filter fails to execute.
2461 4. There are several places where Exim runs a non-Exim command in a
2462 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2463 was being done only for the command run by the queryprogram router. It is
2464 now done for all such subprocesses. The other cases are: ${run, transport
2465 filters, and the commands run by the lmtp and pipe transports.
2467 5. Some older OS have a limit of 256 on the maximum number of file
2468 descriptors. Exim was using setrlimit() to set 1000 as a large value
2469 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2470 systems. I've change it so that if it can't get 1000, it tries for 256.
2472 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2473 was an oversight, and furthermore, ever since the addition of extra
2474 controls (e.g. 4.43/32), the checks on when to allow different forms of
2475 "control" were broken. There should now be diagnostics for all cases when a
2476 control that does not make sense is encountered.
2478 7. $recipients is now available in the predata ACL (oversight).
2480 8. Tidy the search cache before the fork to do a delivery from a message
2481 received from the command line. Otherwise the child will trigger a lookup
2482 failure and thereby defer the delivery if it tries to use (for example) a
2483 cached ldap connection that the parent has called unbind on.
2485 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2486 of $address_data from the recipient verification was clobbered by the
2487 sender verification.
2489 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2490 was its contents. (It was OK if the option was not defined at all.)
2492 11. A "Completed" log line is now written for messages that are removed from
2493 the spool by the -Mrm option.
2495 12. $host_address is now set to the target address during the checking of
2496 ignore_target_hosts.
2498 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2499 being passed; this would have caused $sender_host_name to have been used if
2500 matching the list had actually called for a host name (not very likely,
2501 since this list is usually IP addresses). A host name is now passed as
2504 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2505 code that specifies a non-restarting handler (typically sigaction() in
2506 modern systems) in an attempt to fix a rare and obscure crash bug.
2508 15. Narrowed the window for a race in the daemon that could cause it to ignore
2509 SIGCHLD signals. This is not a major problem, because they are used only to
2510 wake it up if nothing else does.
2512 16. A malformed maildirsize file could cause Exim to calculate negative values
2513 for the mailbox size or file count. Odd effects could occur as a result.
2514 The maildirsize information is now recalculated if the size or filecount
2517 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2518 support for a long time. Removed HAVE_SYS_VFS_H.
2520 18. Updated exipick to current release from John Jetmore.
2522 19. Allow an empty sender to be matched against a lookup in an address list.
2523 Previously the only cases considered were a regular expression, or an
2526 20. Exim went into a mad DNS lookup loop when doing a callout where the
2527 host was specified on the transport, if the DNS lookup yielded more than
2530 21. The RFC2047 encoding function was originally intended for short strings
2531 such as real names; it was not keeping to the 75-character limit for
2532 encoded words that the RFC imposes. It now respects the limit, and
2533 generates multiple encoded words if necessary. To be on the safe side, I
2534 have increased the buffer size for the ${rfc2047: expansion operator from
2537 22. Failure to deliver a bounce message always caused it to be frozen, even if
2538 there was an errors_to setting on the router. The errors_to setting is now
2541 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2542 canonical form (fully expanded) before being placed in
2543 $sender_host_address.
2545 24. Updated eximstats to version 1.33
2547 25. Include certificate and key file names in error message when GnuTLS fails
2548 to set them up, because the GnuTLS error message doesn't include the name
2549 of the failing file when there is a problem reading it.
2551 26. Expand error message when OpenSSL has problems setting up cert/key files.
2554 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2555 (this can affect the format of dates).
2557 28. exim_tidydb, when checking for the continued existence of a message for
2558 which it has found a message-specific retry record, was not finding
2559 messages that were in split spool directories. Consequently, it was
2560 deleting retry records that should have stayed in existence.
2562 29. eximstats updated to version 1.35
2563 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2564 1.35 - bugfix such that pie charts by volume are generated correctly
2566 30. The SPA authentication driver was not abandoning authentication and moving
2567 on to the next authenticator when an expansion was forced to fail,
2568 contradicting the general specification for all authenticators. Instead it
2569 was generating a temporary error. It now behaves as specified.
2571 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2572 (the order specifies the preference for clients). The order is now AES256,
2573 AES128, 3DES, ARCFOUR128.
2575 31. Small patch to Sieve code - explicitly set From: when generating an
2578 32. Exim crashed if a remote delivery caused a very long error message to be
2579 recorded - for instance if somebody sent an entire SpamAssassin report back
2580 as a large number of 550 error lines. This bug was coincidentally fixed by
2581 increasing the size of one of Exim's internal buffers (big_buffer) that
2582 happened as part of the Exiscan merge. However, to be on the safe side, I
2583 have made the code more robust (and fixed the comments that describe what
2586 33. Some experimental protocols are using DNS PTR records for new purposes. The
2587 keys for these records are domain names, not reversed IP addresses. The
2588 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2589 leaves it alone. Component reversal etc. now happens only for IP addresses.
2592 34. The host_aton() function is supposed to be passed a string that is known
2593 to be a valid IP address. However, in the case of IPv6 addresses, it was
2594 not checking this. This is a hostage to fortune. Exim now panics and dies
2595 if the condition is not met. A case was found where this could be provoked
2596 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2597 components; fortuitously, this particular loophole had already been fixed
2598 by change 4.50/55 or 4.44/33 above.
2600 If there are any other similar loopholes, the new check in host_aton()
2601 itself should stop them being exploited. The report I received stated that
2602 data on the command line could provoke the exploit when Exim was running as
2603 exim, but did not say which command line option was involved. All I could
2604 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2605 running as the user.
2608 35. There was a buffer overflow vulnerability in the SPA authentication code
2609 (which came originally from the Samba project). I have added a test to the
2610 spa_base64_to_bits() function which I hope fixes it.
2613 36. The daemon start-up calls getloadavg() while still root for those OS that
2614 need the first call to be done as root, but it missed one case: when
2615 deliver_queue_load_max is set with deliver_drop_privilege. This is
2616 necessary for the benefit of the queue runner, because there is no re-exec
2617 when deliver_drop_privilege is set.
2619 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2620 was in use, was not putting the data itself into the right store pool;
2621 consequently, it could be overwritten for a subsequent message in the same
2622 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2625 38. Sometimes the final signoff response after QUIT could fail to get
2626 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2627 < 0 before doing a fflush(). This bug looks as though it goes back to the
2628 introduction of TLS in release 3.20, but "sometimes" must have been rare
2629 because the tests only now provoked it.
2635 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2636 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2637 Now it returns an integer. A number of calls were still expecting a T/F
2638 return. Fortuitously, in all cases, the tests worked in OK situations,
2639 which is the norm. However, things would have gone wrong on any write
2640 failures on the smtp file descriptor. This function is used when sending
2641 messages over SMTP and also when doing verify callouts.
2643 2. When Exim is called to do synchronous delivery of a locally submitted
2644 message (the -odf or -odi options), it no longer closes stderr before doing
2647 3. Implemented the mua_wrapper option.
2649 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2651 5. Implemented the functions header_remove(), header_testname(),
2652 header_add_at_position(), and receive_remove_recipient(), and exported them
2655 6. If an ACL "warn" statement specified the addition of headers, Exim already
2656 inserted X-ACL-Warn: at the start if there was no header name. However, it
2657 was not making this test for the second and subsequent header lines if
2658 there were newlines in the string. This meant that an invalid header could
2659 be inserted if Exim was badly configured.
2661 7. Allow an ACL "warn" statement to add header lines at the start or after all
2662 the Received: headers, as well as at the end.
2664 8. Added the rcpt_4xx retry error code.
2666 9. Added postmaster_mailfrom=xxx to callout verification option.
2668 10. Added mailfrom=xxxx to the callout verification option, for verify=
2671 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2672 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2674 12. Inserted some casts to stop certain compilers warning when using pointer
2675 differences as field lengths or precisions in printf-type calls (mostly
2676 affecting debugging statements).
2678 13. Added optional readline() support for -be (dynamically loaded).
2680 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2681 same clock tick as a message's arrival, so that its received time was the
2682 same as the "first fail" time on the retry record, and that message
2683 remained on the queue past the ultimate address timeout, every queue runner
2684 would try a delivery (because it was past the ultimate address timeout) but
2685 after another failure, the ultimate address timeout, which should have then
2686 bounced the address, did not kick in. This was a "< instead of <=" error;
2687 in most cases the first failure would have been in the next clock tick
2688 after the received time, and all would be well.
2690 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2691 being recognized when the domain list was tested by the match_domain
2692 condition in an expansion string.
2694 16. Added the ${str2b64: operator.
2696 17. Exim was always calling setrlimit() to set a large limit for the number of
2697 processes, without checking whether the existing limit was already
2698 adequate. (It did check for the limit on file descriptors.) Furthermore,
2699 errors from getrlimit() and setrlimit() were being ignored. Now they are
2700 logged to the main and panic logs, but Exim does carry on, to try to do its
2701 job under whatever limits there are.
2703 18. Imported PCRE 5.0.
2705 19. Trivial typo in log message " temporarily refused connection" (the leading
2708 20. If the log selector return_path_on_delivery was set and an address was
2709 redirected to /dev/null, the delivery process crashed because it assumed
2710 that a return path would always be set for a "successful" delivery. In this
2711 case, the whole delivery is bypassed as an optimization, and therefore no
2714 21. Internal re-arrangement: the function for sending a challenge and reading
2715 a response while authentication was assuming a zero-terminated challenge
2716 string. It's now changed to take a pointer and a length, to allow for
2717 binary data in such strings.
2719 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2721 23. Exim was not respecting finduser_retries when seeking the login of the
2722 uid under which it was called; it was always trying 10 times. (The default
2723 setting of finduser_retries is zero.) Also, it was sleeping after the final
2724 failure, which is pointless.
2726 24. Implemented tls_on_connect_ports.
2728 25. Implemented acl_smtp_predata.
2730 26. If the domain in control=submission is set empty, Exim assumes that the
2731 authenticated id is a complete email address when it generates From: or
2732 Sender: header lines.
2734 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2735 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2736 chown and chgrp in /bin and hostname in /usr/bin.
2738 28. Exim was keeping the "process log" file open after each use, just as it
2739 does for the main log. This opens the possibility of it remaining open for
2740 long periods when the USR1 signal hits a daemon. Occasional processlog
2741 errors were reported, that could have been caused by this. Anyway, it seems
2742 much more sensible not to leave this file open at all, so that is what now
2745 29. The long-running daemon process does not normally write to the log once it
2746 has entered its main loop, and it closes the log before doing so. This is
2747 so that log files can straightforwardly be renamed and moved. However,
2748 there are a couple of unusual error situations where the daemon does write
2749 log entries, and I had neglected to close the log afterwards.
2751 30. The text of an SMTP error response that was received during a remote
2752 delivery was being truncated at 512 bytes. This is too short for some of
2753 the long messages that one sometimes sees. I've increased the limit to
2756 31. It is now possible to make retry rules that apply only when a message has a
2757 specific sender, in particular, an empty sender.
2759 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2760 it possible to be selective about when SMTP synchronization is enforced.
2762 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2764 32. Implemented hosts_connection_nolog.
2766 33. Added an ACL for QUIT.
2768 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2771 35. Added mailbox_size and mailbox_filecount to appendfile.
2773 36. Added control = no_multiline_responses to ACLs.
2775 37. There was a bug in the logic of the code that waits for the clock to tick
2776 in the case where the clock went backwards by a substantial amount such
2777 that the microsecond fraction of "now" was more than the microsecond
2778 fraction of "then" (but the whole seconds number was less).
2780 38. Added support for the libradius Radius client library this is found on
2781 FreeBSD (previously only the radiusclient library was supported).
2787 1. When certain lookups returned multiple values in the form name=value, the
2788 quoting of the values was not always being done properly. Specifically:
2789 (a) If the value started with a double quote, but contained no whitespace,
2791 (b) If the value contained whitespace other than a space character (i.e.
2792 tabs or newlines or carriage returns) it was not quoted.
2793 This fix has been applied to the mysql and pgsql lookups by writing a
2794 separate quoting function and calling it from the lookup code. The fix
2795 should probably also be applied to nisplus, ibase and oracle lookups, but
2796 since I cannot test any of those, I have not disturbed their existing code.
2798 2. A hit in the callout cache for a specific address caused a log line with no
2799 reason for rejecting RCPT. Now it says "Previous (cached) callout
2800 verification failure".
2802 3. There was an off-by-one bug in the queryprogram router. An over-long
2803 return line was truncated at 256 instead of 255 characters, thereby
2804 overflowing its buffer with the terminating zero. As well as fixing this, I
2805 have increased the buffer size to 1024 (and made a note to document this).
2807 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2808 when Exim is waiting for an SMTP response from a remote server, Exim
2809 restarts its select() call on the socket, thereby resetting its timeout.
2810 This is not a problem when such interrupts are rare. Somebody set up a cron
2811 job to run exiwhat every 2 minutes, which is less than the normal select()
2812 timeout (5 or 10 minutes). This meant that the select() timeout never
2813 kicked in because it was always reset. I have fixed this by comparing the
2814 time when an interrupt arrives with the time at the start of the first call
2815 to select(). If more time than the timeout has elapsed, the interrupt is
2816 treated as a timeout.
2818 5. Some internal re-factoring in preparation for the addition of Sieve
2819 extensions (by MH). In particular, the "personal" test is moved to a
2820 separate function, and given an option for scanning Cc: and Bcc: (which is
2821 not set for Exim filters).
2823 6. When Exim created an email address using the login of the caller as the
2824 local part (e.g. when creating a From: or Sender: header line), it was not
2825 quoting the local part when it contained special characters such as @.
2827 7. Installed new OpenBSD configuration files.
2829 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2830 try to make them clearer.
2832 9. Callout options, other than the timeout value, were being ignored when
2833 verifying sender addresses in header lines. For example, when using
2835 verify = header_sender/callout=no_cache
2837 the cache was (incorrectly) being used.
2839 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2840 only the Cygwin environment.
2842 11. When return_path_on_delivery was set as a log selector, if different remote
2843 addresses in the same message used different return paths and parallel
2844 remote delivery occurred, the wrong values would sometimes be logged.
2845 (Whenever a remote delivery process finished, the return path value from
2846 the most recently started remote delivery process was logged.)
2848 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2849 lines when AUTH and/or TLS are in use. This is the "received protocol"
2850 field. Exim used to use "asmtp" for authenticated SMTP, without any
2851 indication (in the protocol name) for TLS use. Now it follows the RFC and
2852 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2853 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2854 names appear in log lines as well as in Received: header lines.
2856 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2857 extensions, and comparison tests, and to fix some bugs.
2859 14. Changes to the "personal" filter test:
2861 (1) The test was buggy in that it was just doing the equivalent of
2862 "contains" tests on header lines. For example, if a user's address was
2863 anne@some.where, the "personal" test would incorrectly be true for
2865 To: susanne@some.where
2867 This test is now done by extracting each address from the header in turn,
2868 and checking the entire address. Other tests that are part of "personal"
2869 are now done using regular expressions (for example, to check local parts
2870 of addresses in From: header lines).
2872 (2) The list of non-personal local parts in From: addresses has been
2873 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2874 taken from the Sieve specification recommendations.
2876 (3) If the message contains any header line starting with "List-" it is
2877 treated as non-personal.
2879 (4) The test for "circular" in the Subject: header line has been removed
2880 because it now seems ill-conceived.
2882 15. Minor typos in src/EDITME comments corrected.
2884 16. Installed latest exipick from John Jetmore.
2886 17. If headers_add on a router specified a text string that was too long for
2887 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2888 of string_sprintf() is now avoided.
2890 18. $message_body_size was not set (it was always zero) when running the DATA
2891 ACL and the local_scan() function.
2893 19. For the "mail" command in an Exim filter, no default was being set for
2894 the once_repeat time, causing a random time value to be used if "once" was
2895 specified. (If the value happened to be <= 0, no repeat happened.) The
2896 default is now 0s, meaning "never repeat". The "vacation" command was OK
2897 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2898 (I found it when inspecting the code).
2900 20. There is now an overall timeout for performing a callout verification. It
2901 defaults to 4 times the callout timeout, which applies to individual SMTP
2902 commands during the callout. The overall timeout applies when there is more
2903 than one host that can be tried. The timeout is checked before trying the
2904 next host. This prevents very long delays if there are a large number of
2905 hosts and all are timing out (e.g. when the network connections are timing
2906 out). The value of the overall timeout can be changed by specifying an
2907 additional sub-option for "callout", called "maxwait". For example:
2909 verify = sender/callout=5s,maxwait=20s
2911 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2912 to the end before writing, but this should make it even safer).
2914 22. Exim was forgetting that it had advertised PIPELINING for the second and
2915 subsequent messages on an SMTP connection. It was also not resetting its
2916 memory on STARTTLS and an internal HELO.
2918 23. When Exim logs an SMTP synchronization error within a session, it now
2919 records whether PIPELINING has been advertised or not.
2921 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2922 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2923 rather than long int.
2925 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2927 26. Added the never_mail option to autoreply.
2933 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2934 crash if the getsockname() call failed; this can happen if a connection is
2935 closed very soon after it is established. The problem was simply in the
2936 order in which certain operations were done, causing Exim to try to write
2937 to the SMTP stream before it had set up the file descriptor. The bug has
2938 been fixed by making things happen in the correct order.
2944 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2945 before the connection was closed, thus losing the rejection response.
2947 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2948 some early Solaris releases, but causes trouble in current releases where
2949 socklen_t is defined.
2951 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2954 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2957 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2958 that releasing the top of it at the end releases what was used for sub-
2959 expansions (unless the block got too big). However, discard this block if
2960 the first thing is a variable or header, so that we can use its block when
2961 it is dynamic (useful for very large $message_headers, for example).
2963 6. Lookups now cache *every* query, not just the most recent. A new, separate
2964 store pool is used for this. It can be recovered when all lookup caches are
2965 flushed. Lookups now release memory at the end of their result strings.
2966 This has involved some general refactoring of the lookup sources.
2968 7. Some code has been added to the store_xxx() functions to reduce the amount
2969 of flapping under certain conditions.
2971 8. log_incoming_interface used to affect only the <= reception log lines. Now
2972 it causes the local interface and port to be added to several more SMTP log
2973 lines, for example "SMTP connection from", and rejection lines.
2975 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2977 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2979 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2980 could be overwritten at the end of the current message (or the start of a
2981 new message if it was set in a HELO ACL). The value is now preserved for
2982 the duration of the SMTP connection.
2984 12. If a transport had a headers_rewrite setting, and a matching header line
2985 contained an unqualified address, that address was qualified, even if it
2986 did not match any rewriting rules. The underlying bug was that the values
2987 of the flags that permit the existence of unqualified sender and recipient
2988 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2989 non-local messages, and by -bnq for local messages) were not being
2990 preserved with the message after it was received.
2992 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2993 "next input=" as part of the text comprising the host identity instead of
2994 the correct text. The code was using the same buffer for two different
2995 strings. However, depending on which order the printing function evaluated
2996 its arguments, the bug did not always show up. Under Linux, for example, my
2997 test suite worked just fine.
2999 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3000 doesn't work with some older versions of Perl. It has been changed to "my",
3001 which in any case is probably the better facility to use.
3003 15. A really picky compiler found some instances of statements for creating
3004 error messages that either had too many or two few arguments for the format
3007 16. The size of the buffer for calls to the DNS resolver has been increased
3008 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3009 for addresses that have a lot of PTR records. This alleviates a problem; it
3010 does not fully solve it.
3012 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3013 buffer now truncates the list and logs the incident, which is the same
3014 action as happens when Exim is looking up a host name and its aliases.
3015 Previously in this situation something unpredictable would happen;
3016 sometimes it was "internal error: store_reset failed".
3018 18. If a server dropped the connection unexpectedly when an Exim client was
3019 using GnuTLS and trying to read a response, the client delivery process
3020 crashed while trying to generate an error log message.
3022 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3023 string, for example:
3025 warn message = H1: something\nH2: something
3027 the text was added as a single header line from Exim's point of view
3028 though it ended up OK in the delivered message. However, searching for the
3029 second and subsequent header lines using $h_h2: did not work. This has been
3030 fixed. Similarly, if a system filter added multiple headers in this way,
3031 the routers could not see them.
3033 20. Expanded the error message when iplsearch is called with an invalid key to
3034 suggest using net-iplsearch in a host list.
3036 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3037 ACLs are no longer actually imposed (and a message to that effect is
3040 22. If a "gecos" field in a passwd entry contained escaped characters, in
3041 particular, if it contained a \" sequence, Exim got it wrong when building
3042 a From: or a Sender: header from that name. A second bug also caused
3043 incorrect handling when an unquoted " was present following a character
3044 that needed quoting.
3046 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3047 was not being matched caselessly.
3049 24. Arranged for all hyphens in the exim.8 source to be escaped with
3052 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3053 properly. Recipient callout cache records were still being keyed to include
3054 the sender, even when use_sender was set false. This led to far more
3055 callouts that were necessary. The sender is no longer included in the key
3056 when use_sender is false.
3058 26. Added "control = submission" modifier to ACLs.
3060 27. Added the ${base62d: operator to decode base 62 numbers.
3062 28. dnsdb lookups can now access SRV records.
3064 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3065 the configuration file.
3067 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3068 (-v) mode. This makes the output for a verbose queue run more intelligible.
3070 31. Added a use_postmaster feature to recipient callouts.
3072 32. Added the $body_zerocount variable, containing the number of binary zero
3073 bytes in the message body.
3075 33. The time of last modification of the "new" subdirectory is now used as the
3076 "mailbox time last read" when there is a quota error for a maildir
3079 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3081 35. Added +ignore_unknown as a special item in host lists.
3083 36. Code for decoding IPv6 addresses in host lists is now included, even if
3084 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3085 address was recognized as an IP address, but was then not correctly decoded
3086 into binary, causing unexpected and incorrect effects when compared with
3093 1. Very minor rewording of debugging text in manualroute to say "list of
3094 hosts" instead of "hostlist".
3096 2. If verify=header_syntax was set, and a header line with an unqualified
3097 address (no domain) and a large number of spaces between the end of the
3098 name and the colon was received, the reception process suffered a buffer
3099 overflow, and (when I tested it) crashed. This was caused by some obsolete
3100 code that should have been removed. The fix is to remove it!
3102 3. When running in the test harness, delay a bit after writing a bounce
3103 message to get a bit more predictability in the log output.
3105 4. Added a call to search_tidyup() just before forking a reception process. In
3106 theory, someone could use a lookup in the expansion of smtp_accept_max_
3107 per_host which, without the tidyup, could leave open a database connection.
3109 5. Added the variables $recipient_data and $sender_data which get set from a
3110 lookup success in an ACL "recipients" or "senders" condition, or a router
3111 "senders" option, similar to $domain_data and $local_part_data.
3113 6. Moved the writing of debug_print from before to after the "senders" test
3116 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3117 problems for message scanning, either using a data ACL, or using
3118 local_scan() because the Received: header was not generated till after they
3119 were called (in order to set the time as the time of reception completion).
3120 I have revised the way this works. The header is now generated after the
3121 body is received, but before the ACL or local_scan() are called. After they
3122 are run, the timestamp in the header is updated.
3128 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3129 before starting a queue runner without re-exec. This happened only when
3130 deliver_drop_privilege was set or when the Exim user was set to root. The
3131 effect of the bug was that timeouts during subsequent deliveries caused
3132 crashes instead of being properly handled. The handler is now left at its
3133 default (and expected) setting.
3135 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3136 message, again when deliver_drop_privilege is set or Exim is run as root.
3137 The bug described in (1) was not present in this case, but the tidying up
3138 of the other signals was missing. I have made the two cases consistent.
3140 3. The ignore_target_hosts setting on a manualroute router was being ignored
3141 for hosts that were looked up using the /MX notation.
3143 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3146 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3147 operated on the sender address. After changing the $sender_address to <>
3148 for the sender address verify, Exim was re-instated it as the original
3149 (before rewriting) address, but remembering that it had rewritten it, so it
3150 wasn't rewriting it again. This bug also had the effect of breaking the
3151 sender address verification caching when the sender address was rewritten.
3153 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3154 This has been changed so that if the ip literal address matches
3155 ignore_target_hosts, the router declines.
3157 7. Added expansion conditions match_domain, match_address, and match_local_
3158 part (NOT match_host).
3160 8. The placeholder for the Received: header didn't have a length field set.
3162 9. Added code to Exim itself and to exim_lock to test for a specific race
3163 condition that could lead to file corruption when using MBX delivery. The
3164 issue is with the lockfile that is created in /tmp. If this file is removed
3165 after a process has opened it but before that process has acquired a lock,
3166 there is the potential for a second process to recreate the file and also
3167 acquire a lock. This could lead to two Exim processes writing to the file
3168 at the same time. The added code performs the same test as UW imapd; it
3169 checks after acquiring the lock that its file descriptor still refers to
3170 the same named file.
3172 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3173 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3176 11. Added the smtp_active_hostname option. If used, this will typically be made
3177 to depend on the incoming interface address. Because $interface_address is
3178 not set up until the daemon has forked a reception process, error responses
3179 that can happen earlier (such as "too many connections") no longer contain
3182 12. If an expansion in a condition on a "warn" statement fails because a lookup
3183 defers, the "warn" statement is abandoned, and the next ACL statement is
3184 processed. Previously this caused the whole ACL to be aborted.
3186 13. Added the iplsearch lookup type.
3188 14. Added ident_timeout as a log selector.
3190 15. Added tls_certificate_verified as a log selector.
3192 16. Added a global option tls_require_ciphers (compare the smtp transport
3193 option of the same name). This controls incoming TLS connections.
3195 17. I finally figured out how to make tls_require_ciphers do a similar thing
3196 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3197 before starting the TLS session.
3199 18. Tabs are now shown as \t in -bP output.
3201 19. If the log selector return_path_on_delivery was set, Exim crashed when
3202 bouncing a message because it had too many Received: header lines.
3204 20. If two routers both had headers_remove settings, and the first one included
3205 a superfluous trailing colon, the final name in the first list and the
3206 first name in the second list were incorrectly joined into one item (with a
3207 colon in the middle).
3213 1. Added -C and -D options to the exinext utility, mainly to make it easier
3214 to include in the automated testing, but these could be helpful when
3215 multiple configurations are in use.
3217 2. The exinext utility was not formatting the output nicely when there was
3218 an alternate port involved in the retry record key, nor when there was a
3219 message id as well (for retries that were specific to a specific message
3220 and a specific host). It was also confused by IPv6 addresses, because of
3221 the additional colons they contain. I have fixed the IPv4 problem, and
3222 patched it up to do a reasonable job for IPv6.
3224 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3225 delivery, the log line now contains "pipelined" if PIPELINING was used.
3227 4. An SMTP transport process used to panic and die if the bind() call to set
3228 an explicit outgoing interface failed. This has been changed; it is now
3229 treated in the same way as a connect() failure.
3231 5. A reference to $sender_host_name in the part of a conditional expansion
3232 that was being skipped was still causing a DNS lookup. This no longer
3235 6. The def: expansion condition was not recognizing references to header lines
3236 that used bh_ and bheader_.
3238 7. Added the _cache feature to named lists.
3240 8. The code for checking quota_filecount in the appendfile transport was
3241 allowing one more file than it should have been.
3243 9. For compatibility with Sendmail, the command line option
3251 and sets the incoming protocol and host name (for trusted callers). The
3252 host name and its colon can be omitted when only the protocol is to be set.
3253 Note the Exim already has two private options, -pd and -ps, that refer to
3254 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3255 "s", but I don't think that's a major issue.
3257 10. A number of refactoring changes to the code, none of which should affect
3260 (a) The number of logging options was getting close to filling up the
3261 32-bit word that was used as a bit map. I have split them into two classes:
3262 those that are passed in the argument to log_write(), and those that are
3263 only ever tested independently outside of that function. These are now in
3264 separate 32-bit words, so there is plenty of room for expansion again.
3265 There is no change in the user interface or the logging behaviour.
3267 (b) When building, for example, log lines, the code previously used a
3268 macro that called string_cat() twice, in order to add two strings. This is
3269 not really sufficiently general. Furthermore, there was one instance where
3270 it was actually wrong because one of the argument was used twice, and in
3271 one call a function was used. (As it happened, calling the function twice
3272 did not affect the overall behaviour.) The macro has been replaced by a
3273 function that can join an arbitrary number of extra strings onto a growing
3276 (c) The code for expansion conditions now uses a table and a binary chop
3277 instead of a serial search (which was left over from when there were very
3278 few conditions). Also, it now recognizes conditions like "pam" even when
3279 the relevant support is not compiled in: a suitably worded error message is
3280 given if an attempt is made to use such a condition.
3282 11. Added ${time_interval:xxxxx}.
3284 12. A bug was causing one of the ddress fields not to be passed back correctly
3285 from remote delivery subprocesses. The field in question was not being
3286 subsequently used, so this caused to problems in practice.
3288 13. Added new log selectors queue_time and deliver_time.
3290 14. Might have fixed a bug in maildirsizefile handling that threw up
3291 "unexpected character" debug warnings, and recalculated the data
3292 unnecessarily. In any case, I expanded the warning message to give more
3295 15. Added the message "Restricted characters in address" to the statements in
3296 the default ACL that block characters like @ and % in local parts.
3298 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3299 Three changes have been made:
3301 (a) There was a serious bug; a negative response to MAIL caused the whole
3302 recipient domain to be cached as invalid, thereby blocking all messages
3303 to all local parts at the same domain, from all senders. This bug has
3304 been fixed. The domain is no longer cached after a negative response to
3305 MAIL if the sender used is not empty.
3307 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3310 (c) A new callout option, "use_sender" has been added for people who want
3311 the modified behaviour.
3317 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3320 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3321 indeed breaks things for older releases.
3323 3. Added additional logging to the case where there is a problem reading data
3324 from a filter that is running in a subprocess using a pipe, in order to
3325 try to track down a specific problem.
3327 4. Testing facility fudge: when running in the test harness and attempting
3328 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3329 getting "No route to host". Convert this to a timeout.
3331 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3334 6. Some OS don't have socklen_t but use size_t instead. This affects the
3335 fifth argument of getsockopt() amongst other things. This is now
3336 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3337 can be set for individual OS. I have set it for SunOS5, OSF1, and
3338 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3339 some earlier ones do not.
3341 7. Change 4.30/15 was not doing the test caselessly.
3343 8. The standard form for an IPv6 address literal was being rejected by address
3344 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3345 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3346 this, as well as the form without the "IPv6" on the front (but only when
3347 address literals are enabled, of course).
3349 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3351 10. Exim crashed if a message with an empty sender address specified by -f
3352 encountered a router with an errors_to setting. This could be provoked only
3353 by a command such as
3357 where an empty string was supplied; "<>" did not hit this bug.
3359 11. Installed PCRE release 4.5.
3361 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3362 remained set. It is now erased.
3364 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3365 times from message ids (which are base 36 rather than the normal 62).
3367 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3368 were being counted as actual protocol errors, and logged if the log
3369 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3370 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3371 and DATA following a set of rejected RCPTs do not count as protocol errors.
3372 In other words, Exim assumes they were pipelined, though this may not
3373 actually be the case. Of course, in all cases the client gets an
3374 appropriate error code.
3376 15. If a lookup fails in an ACL condition, a message about the failure may
3377 be available; it is used if testing the ACL cannot continue, because most
3378 such messages specify what the cause of the deferral is. However, some
3379 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3380 that caused an old message to be retained and used if a later statement
3381 caused a defer, replacing the real cause of the deferral.
3383 16. If an IP address had so many PTR records that the DNS lookup buffer
3384 was not large enough to hold them, Exim could crash while trying to process
3385 the truncated data. It now detects and logs this case.
3387 17. Further to 4.21/58, another change has been made: if (and only if) the
3388 first line of a message (the first header line) ends with CRLF, a bare LF
3389 in a subsequent header line has a space inserted after it, so as not to
3390 terminate the header.
3392 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3393 unnecessarily, used atoi() instead of strtol(), and didn't check the
3394 termination when getting file sizes from file names by regex.
3396 19. Completely re-implemented the support for maildirsize files, in the light
3397 of a number of problems with the previous contributed implementation
3398 (4.30/29). In particular:
3400 . If the quota is zero, the maildirsize file is maintained, but no quota is
3403 . If the maildir directory does not exist, it is created before any attempt
3404 to write a maildirsize file.
3406 . The quota value in the file is just a cache; if the quota is changed in
3407 the transport, the new value overrides.
3409 . A regular expression is available for excluding directories from the
3412 20. The autoreply transport checks the characters in options that define the
3413 message's headers; it allows continued headers, but it was checking with
3414 isspace() after an embedded newline instead of explicitly looking for a
3417 21. If all the "regular" hosts to which an address was routed had passed their
3418 expiry times, and had not reached their retry times, the address was
3419 bounced, even if fallback hosts were defined. Now Exim should go on to try
3422 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3423 equivalent code in the SMTP transport. Some hosts send humungous responses
3424 to HELO/EHLO, more than 1024 it seems.
3426 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3427 gives compiler warnings in some environments. I've now done it "properly",
3430 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3431 (because of problems with the built-in one) was declared to return uschar *
3432 instead of char *, causing compiler failure.
3434 25. Fixed a file descriptor leak when processing alias/forward files.
3436 26. Fixed a minor format string issue in dbfn.c.
3438 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3440 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3441 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3444 29. When a sender address is verified, it is cached, to save repeating the test
3445 when there is more than one recipient in a message. However, when the
3446 verification involves a callout, it is possible for different callout
3447 options to be set for different recipients. It is too complicated to keep
3448 track of this in the cache, so now Exim always runs a verification when a
3449 callout is required, relying on the callout cache for the optimization.
3450 The overhead is duplication of the address routing, but this should not be
3453 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3454 to be verified with callout=postmaster, and the main callout worked but the
3455 postmaster check failed, the verification correctly failed. However, if a
3456 subsequent RCPT command asked for sender verification *without* the
3457 postmaster check, incorrect caching caused this verification also to fail,
3460 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3461 it was not caching the DNS options (qualify_single, search_parents) that
3462 were used when the lookup failed. A subsequent lookup with different
3463 options therefore always gave the same answer, though there were cases
3464 where it should not have. (Example: a "domains = !$mx_any" option on a
3465 dnslookup router: the "domains" option is always processed without any
3466 widening, but the router might have qualify_single set.) Now Exim uses the
3467 cached value only when the same options are set.
3469 32. Added John Jetmore's "exipick" utility to the distribution.
3471 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3472 than a timeout (e.g. a certificate is required, and is not provided), an
3473 Exim server now closes the connection immediately. Previously it waited for
3474 the client to close - but if the client is SSL, it seems that they each
3475 wait for each other, leading to a delay before one of them times out.
3477 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3478 maintained 0.8.x compatibility because I don't think many are using it, and
3479 it is clearly obsolete.
3481 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3484 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3485 client certificate was expired. A simple patch fixes this, though I don't
3486 understand the full logic of why the verify callback is called multiple
3489 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3490 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3491 which causes problems with some clients (such as the Certicom SSL Plus
3492 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3493 disables the coutermeasure allowing Eudora to connect."
3495 38. Exim was not checking that a write() to a log file succeeded. This could
3496 lead to Bad Things if a log got too big, in particular if it hit a file
3497 size limit. Exim now panics and dies if it cannot write to a log file, just
3498 as it does if it cannot open a log file.
3500 39. Modified OS/Makefile-Linux so that it now contains
3502 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3504 The two -D definitions ensure that Exim is compiled with large file
3505 support, which makes it possible to handle log files that are bigger than
3508 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3509 instance) a domain was checked against a named list that involved a lookup,
3510 causing $domain_data to be set, then another domain was checked against the
3511 same list, then the first domain was re-checked, the value of $domain_data
3512 after the final check could be wrong. In particular, if the second check
3513 failed, it could be set empty. This bug probably also applied to
3516 41. The strip_trailing_dot option was not being applied to the address given
3517 with the -f command-line option.
3519 42. The code for reading a message's header from the spool was incrementing
3520 $received_count, but never initializing it. This meant that the value was
3521 incorrect (doubled) while delivering a message in the same process in which
3522 it was received. In the most common configuration of Exim, this never
3523 happens - a fresh exec is done - but it can happen when
3524 deliver_drop_privilege is set.
3526 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3527 it now includes up to 150 characters of the unexpected data in the log
3530 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3531 and building data strings. The size of both of these buffers was 10 000
3532 bytes - far larger than anybody would *ever* want, thought I. Needless to
3533 say, somebody hit the limit. I have increased the maximum line length to
3534 20 000 and the maximum data length of concatenated lines to 100 000. I have
3535 also fixed two bugs, because there was no checking on these buffers. Tsk,
3536 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3537 buffer is too small.
3539 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3540 lsearch lookups. Now it does.
3542 46. When parsing a route_list item in a manualroute router, a fixed-length
3543 buffer was used for the list of hosts. I made this 1024 bytes long,
3544 thinking that nobody would ever have a list of hosts that long. Wrong.
3545 Somebody had a whole pile of complicated expansion conditions, and the
3546 string was silently truncated, leading to an expansion error. It turns out
3547 that it is easier to change to an unlimited length (owing to other changes
3548 that have happened since this code was originally written) than to build
3549 structure for giving a limitation error. The length of the item that
3550 expands into the list of hosts is now unlimited.
3552 47. The lsearch lookup could not handle data where the length of text line was
3553 more than 4095 characters. Such lines were truncated, leading to shortened
3554 data being returned. It should now handle lines of any length.
3556 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3557 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3560 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3561 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3562 no longer happens in comment lines. A list of replacements is now placed
3563 at the head of all of the source files, except those whose only change is
3564 to replace PERL_COMMAND in the very first #! line.
3566 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3567 messages on the queue, with a bottom-up merge sort, using code contributed
3568 by Michael Haardt. This should make operations like -bp somewhat faster on
3569 large queues. It won't affect queue runners, except when queue_run_in_order
3572 51. Installed eximstats 1.31 in the distribution.
3574 52. Added support for SRV lookups to the dnslookup router.
3576 53. If an ACL referred to $message_body or $message_body_end, the value was not
3577 reset for any messages that followed in the same SMTP session.
3579 54. The store-handling optimization for building very long strings was not
3580 differentiating between the different store pools. I don't think this
3581 actually made any difference in practice, but I've tidied it.
3583 55. While running the routers to verify a sender address, $sender_address
3584 was still set to the sender address. This is wrong, because when routing to
3585 send a bounce to the sender, it would be empty. Therefore, I have changed
3586 it so that, while verifying a sender address, $sender_address is set to <>.
3587 (There is no change to what happens when verifying a recipient address.)
3589 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3590 target A or AAAA records (if not already returned) without resetting the
3591 qualify_single or search_parents options of the DNS resolver. These are
3592 inappropriate in this case because the targets of MX and SRV records must
3593 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3594 target that, when qualified, matched something in the local domain. These
3595 two options are now turned off when doing these lookups.
3597 57. It seems that at least some releases of Reiserfs (which does not have the
3598 concept of a fixed number of inodes) returns zero and not -1 for the
3599 number of available inodes. This interacted badly with check_spool_inodes,
3600 which assumed that -1 was the "no such thing" setting. What I have done is
3601 to check that the total number of inodes is greater than zero before doing
3602 the test of how many are available.
3604 58. When a "warn" ACL statement has a log_message modifier, the message is
3605 remembered, and not repeated. This is to avoid a lot of repetition when a
3606 message has many recipients that cause the same warning to be written.
3607 Howewer, Exim was preserving the list of already written lines for an
3608 entire SMTP session, which doesn't seem right. The memory is now reset if a
3609 new message is started.
3611 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3612 debugging output unless log_rewrite was also set.
3614 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3615 of (int)(handle) when we know that handle contains (void *)(-1).
3617 61. The Exim daemon panic-logs an error return when it closes the incoming
3618 connection. However "connection reset by peer" seems to be common, and
3619 isn't really an error worthy of noting specially, so that particular error
3622 62. When Exim is trying to find all the local interfaces, it used to panic and
3623 die if the ioctl to get the interface flags failed. However, it seems that
3624 on at least one OS (Solaris 9) it is possible to have an interface that is
3625 included in the list of interfaces, but for which you get a failure error
3626 for this call. This happens when the interface is not "plumbed" into a
3627 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3628 failure of the "get flags" call assumes that the interface is down.
3630 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3631 makes life easier for people who are doing arithmetic on fields extracted
3632 from dates, where you often get leading zeros that should not be
3633 interpreted as octal.
3635 64. Added qualify_domain to the redirect router, to override the global
3638 65. If a pathologically long header line contained very many addresses (the
3639 report of this problem mentioned 10 000) and each of them was rewritten,
3640 Exim could use up a very large amount of memory. (It kept on making new
3641 copies of the header line as it rewrote, and never released the old ones.)
3642 At the expense of a bit more processing, the header rewriting function has
3643 been changed so that it no longer eats memory in this way.
3645 66. The generation of the Received: header has been moved from the time that a
3646 message starts to be received, to the time that it finishes. The timestamp
3647 in the Received: header should now be very close to that of the <= log
3648 line. There are two side-effects of this change:
3650 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3651 logged header lines no longer include the local Received: line, because
3652 it has not yet been created. The same applies to a copy of the message
3653 that is returned to a non-SMTP sender when a message is rejected.
3655 (b) When a filter file is tested using -bf, no additional Received: header
3656 is added to the test message. After some thought, I decided that this
3659 This change does not affect the value of $received_for. It is still set
3660 after address rewriting, but before local_scan() is called.
3662 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3664 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3665 gave an unhelpful panic error message, and a defer error. I have managed to
3666 change this behaviour so that it now rejects any supplied certificate,
3667 which seems right, as the list of acceptable certificates is empty.
3669 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3670 gave an unhelpful defer error. I have not managed to make this reject any
3671 supplied certificates, but the error message it gives is "no certificate
3672 supplied", which is not helpful.
3674 70. exigrep's output now also includes lines that are not associated with any
3675 message, but which match the given pattern. Implemented by a patch from
3676 Martin Sluka, which also tidied up the Perl a bit.
3678 71. Recipient callout verification, like sender verification, was using <> in
3679 the MAIL FROM command. This isn't really the right thing, since the actual
3680 sender may affect whether the remote host accepts the recipient or not. I
3681 have changed it to use the actual sender in the callout; this means that
3682 the cache record is now keyed on a recipient/sender pair, not just the
3683 recipient address. There doesn't seem to be a real danger of callout loops,
3684 since a callout by the remote host to check the sender would use <>.
3685 [SEE ABOVE: changed after hitting problems.]
3687 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3688 temporary errors. However, in the case of such a code being given after
3689 the end of a data transmission (i.e. after ".") Exim was failing to write
3690 a retry record for the message. (Yes, there was some broken host that was
3691 actually sending 8xx at this point.)
3693 73. An unknown lookup type in a host list could cause Exim to panic-die when
3694 the list was checked. (An example that provoked this was putting <; in the
3695 middle of a list instead of at the start.) If this happened during a DATA
3696 ACL check, a -D file could be left lying around. This kind of configuration
3697 error no longer causes Exim to die; instead it causes a defer errror. The
3698 incident is still logged to the main and panic logs.
3700 74. Buglet left over from Exim 3 conversion. The message "too many messages
3701 in one connection" was written to the rejectlog but not the mainlog, except
3702 when address rewriting (yes!) was being logged.
3704 75. Added write_rejectlog option.
3706 76. When a system filter was run not as root (that is, when system_filter_user
3707 was set), the values of the $n variables were not being returned to the
3708 main process; thus, they were not subsequently available in the $sn
3711 77. Added +return_path_on_delivery log selector.
3713 78. A connection timeout was being treated differently from recipients deferred
3714 when testing hosts_max_try with a message that was older than the host's
3715 retry timeout. (The host should not be counted, thus allowing all hosts to
3716 be tried at least once before bouncing.) This may have been the cause of an
3717 occasionally reported bug whereby a message would remain on the queue
3718 longer than the retry timeout, but would be bounced if a delivery was
3719 forced. I say "may" because I never totally pinned down the problem;
3720 setting up timeout/retry tests is difficult. See also the next item.
3722 79. The ultimate address timeout was not being applied to errors that involved
3723 a combination of host plus message (for example, a timeout on a MAIL
3724 command). When an address resolved to a number of possible hosts, and they
3725 were not all tried for each delivery (e.g. because of hosts_max_try), a
3726 message could remain on the queue longer than the retry timeout.
3728 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3731 81. Fixed an obscure SMTP outgoing bug which required at least the following
3732 conditions: (a) there was another message waiting for the same server;
3733 (b) the server returned 5xx to all RCPT commands in the first message so
3734 that the message was not completed; (c) the server dropped the connection
3735 or gave a negative response to the RSET that Exim sends to abort the
3736 transaction. The observed case was a dropped connection after DATA that had
3737 been sent in pipelining mode. That is, the server had advertised PIPELINING
3738 but was not implementing it correctly. The effect of the bug was incorrect
3739 behaviour, such as trying another host, and this could lead to a crash.
3745 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3746 and daemon.c were passed as pointers to ints; they should have been
3747 pointers to socklen_t variables (which are typically unsigned ints).
3749 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3752 3. Fixed a really odd bug that affected only the testing scheme; patching a
3753 certain fixed string in the binary changed the value of another string that
3754 happened to be identical to the end of the original first string.
3756 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3757 name", it returns that address as the IP address. On some operating
3758 systems (e.g. Solaris), it also passes back the IP address string as the
3759 "host name". However, on others (e.g. Linux), it passes back an empty
3760 string. Exim wasn't checking for this, and was changing the host name to an
3761 empty string, assuming it had been canonicized.
3763 5. Although rare, it is permitted to have more than one PTR record for a given
3764 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3765 all the names associated with an address, because they do in Solaris.
3766 However, it seems that they do not in Linux for data that comes from the
3767 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3768 I found this out when I moved to a new Linux workstation and tried to run
3769 the Exim test suite.
3771 To get round this problem I have changed the code so that it now does its
3772 own call to the DNS to look up PTR records when searching for a host name.
3773 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3774 addresses that are only in /etc/hosts are still found.
3776 This behaviour is, however, controlled by an option called host_lookup_
3777 order, which defaults to "bydns:byaddr". If people want to use the other
3778 order, or indeed, just use one or the other means of lookup, they can
3779 specify it in this variable.
3781 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3782 some operating systems, this comes back from gethostbyaddr() as an empty
3783 string, and this is what Exim used to test for. However, it seems that in
3784 other systems, "." is yielded. Exim now tests for this case too.
3786 7. The values of check_spool_space and check_log_space are now held internally
3787 as a number of kilobytes instead of an absolute number of bytes. If a
3788 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3789 kilobyte. This means that much larger values can be stored.
3791 8. Exim monitor: an attempt to get the action menu when not actually pointing
3792 at a message produces an empty menu entitled "No message selected". This
3793 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3794 no entries in it ("Shell widget menu has zero width and/or height"). So I
3795 have added a single, blank menu entry in this case.
3797 9. Added ${quote_local_part.
3799 10. MIME decoding is now applied to the contents of Subject: header lines when
3802 11. Now that a reference to $sender_host_address automatically causes a reverse
3803 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3804 host lookup before query-style lookups in lists that might use this
3805 variable. This has therefore been abolished, and the "net-" prefix is no
3806 longer necessary for query-style lookups.
3808 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3809 have been a typo for LFLAGS, so it has been changed.
3811 13. The install script calls Exim with "-C /dev/null" in order to find the
3812 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3813 to be output. Howeve, since Exim outputs its version number before the
3814 error, it didn't break the script. It just looked ugly. I fixed this by
3815 always allowing "-C /dev/null" if the caller is root.
3817 14. Ignore overlarge ACL variable number when reading spool file - insurance
3818 against a later release with more variables having written the file.
3820 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3821 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3822 this, as well as the form without the "IPv6" on the front.
3824 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3825 OS/Makefile-Darwin file.
3827 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3828 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3830 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3831 variable. However, some OS use ints for inodes. I've added cast to long int
3832 to get rid of the compiler warning.
3834 19. I had forgotten to lock out "/../" in configuration file names when
3835 ALT_CONFIG_PREFIX was set.
3837 20. Routers used for verification do not need to specify transports. However,
3838 if such a router generated a host list, and callout was configured, Exim
3839 crashed, because it could not find a port number from the (non-existent)
3840 transport. It now assumes port 25 in this circumstance.
3842 21. Added the -t option to exigrep.
3844 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3845 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3846 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3847 them is set, LOOKUP_LSEARCH is forced.
3849 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3851 24. Added sender and host information to the "rejected by local_scan()" log
3852 line; previously there was no indication of these.
3854 25. Added .include_if_exists.
3856 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3857 Linux. It turns out that not all file systems support this. Apparently some
3858 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3859 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3860 not supported on the file descriptor, is now ignored when Exim is trying to
3861 sync a directory. This applies only to Linux.
3863 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3865 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3866 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3869 29. Added support for maildirsize files from supplied patch (modified a bit).
3871 30. The use of :fail: followed by an empty string could lead Exim to respond to
3872 sender verification failures with (e.g.):
3874 550 Verification failed for <xxx>
3875 550 Sender verify failed
3877 where the first response line was missing the '-' that indicates it is not
3878 the final line of the response.
3880 31. The loop for finding the name of the user that called Exim had a hardwired
3881 limit of 10; it now uses the value of finduser_retries, which is used for
3882 all other user lookups.
3884 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3887 33. Exim was neglecting to zero errno before one call of strtol() when
3888 expanding a string and expecting an integer value. On some systems this
3889 resulted in spurious "integer overflow" errors. Also, it was casting the
3890 result into an int without checking.
3892 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3893 did not work. The code looks as if it has *never* worked, though it appears
3894 to have been documented since at least releast 1.62. I have made it work.
3896 35. The "timeout_DNS" error in retry rules, also documented since at least
3897 1.62, also never worked. As it isn't clear exactly what this means, and
3898 clearly it isn't a major issue, I have abolished the feature by treating it
3899 as "timeout", and writing a warning to the main and panic logs.
3901 36. The display of retry rules for -brt wasn't always showing the error code
3904 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3905 timeout_connect_A, timeout_connect_MX.
3907 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3908 to the empty sender.
3910 39. The daemon was not analysing the content of -oX till after it had closed
3911 stderr and disconnected from the controlling terminal. This meant that any
3912 syntax errors were only noted on the panic log, and the return code from
3913 the command was 0. By re-arranging the code a little, I've made the
3914 decoding happen first, so such errors now appear on stderr, and the return
3915 code is 1. However, the actual setting up of the sockets still happens in
3916 the disconnected process, so errors there are still only recorded on the
3919 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3920 connections (as happens on some IP stacks) was logged at start up time as
3921 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3922 it from "IPv6 and IPv4", which means that two separate sockets are being
3925 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3926 says whether AF_INET or AF_INET6 was passed as an argument.
3928 42. Exiwhat output was messed up when time zones were included in log
3931 43. Exiwhat now gives more information about the daemon's listening ports,
3932 and whether -tls-on-connect was used.
3934 44. The "port" option of the smtp transport is now expanded.
3936 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3937 silently ignored. Now an error message is written to the main and panic
3940 46. There's a new ACL modifier called "logwrite" which writes to a log file
3941 as soon as it is encountered.
3943 47. Added $local_user_uid and $local_user_gid at routing time.
3945 48. Exim crashed when trying to verify a sender address that was being
3948 49. Exim was recognizing only a space character after ".include". It now also
3949 recognizes a tab character.
3951 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3952 extracting the relevant information from the specification. The man page no
3953 longer contains scrambled data for the -d option, and I've added a section
3954 at the front about calling Exim under different names.
3956 51. Added "extra_headers" argument to the "mail" command in filter files.
3958 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3961 53. Installed eximstats 1.29.
3963 54. Added transport_filter_timeout as a generic transport option.
3965 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3966 Cc: header lines. This was required by RFC 822, but it not required by RFC
3969 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3970 incoming messages that did not have them. Now it does so only if the
3971 message originates locally, that is, if there is no associated remote host
3972 address. When Resent- header lines are present, this applies to the Resent-
3973 lines rather than the non-Resent- lines.
3975 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3976 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3978 58. Messages for configuration errors now include the name of the main
3979 configuration file - useful now that there may be more than one file in a
3980 list (.included file names were always shown).
3982 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3983 for those rare installations that do not start the daemon as root or run it
3984 setuid root. I've cut out the call to initgroups() if the daemon is not
3987 60. The Exim user and group can now be bound into the binary as text strings
3988 that are looked up at the start of Exim's processing.
3990 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3992 62. Added $mailstore_basename variable.
3994 63. Installed patch to sieve.c from Michael Haardt.
3996 64. When Exim failed to open the panic log after failing to open the main log,
3997 the original message it was trying to log was written to stderr and debug
3998 output, but if they were not available (the usual case in production), it
3999 was lost. Now it is written to syslog before the two lines that record the
4000 failures to open the logs.
4002 65. Users' Exim filters run in subprocesses under the user's uid. It is
4003 possible for a "deliver" command or an alias in a "personal" command to
4004 provoke an address rewrite. If logging of address rewriting is configured,
4005 this fails because the process is not running as root or exim. There may be
4006 a better way of dealing with this, but for the moment (because 4.30 needs
4007 to be released), I have disabled address rewrite logging when running a
4008 filter in a non-root, non-exim process.
4014 1. The buildconfig auxiliary program wasn't quoting the value set for
4015 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4016 not defined. This bug was masked in 4.22 by the effect that was fixed in
4019 2. Some messages that were rejected after a message id was allocated were
4020 shown as "incomplete" by exigrep. It no longer does this for messages that
4021 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4023 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4024 have allow_domain_literals set, the ID did not get logged in the <= line.
4025 Domain literals are now always recognized in Message-ID: header lines.
4027 4. The first argument for a ${extract expansion item is the key name or field
4028 number. Leading and trailing spaces in this item were not being ignored,
4029 causing some misleading effects.
4031 5. When deliver_drop_privilege was set, single queue runner processes started
4032 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4033 same command in the process it spins off) were not dropping privilege.
4035 6. When the daemon running as "exim" started a queue runner, it always
4036 re-executed Exim in the spun-off process. This is a waste of effort when
4037 deliver_drop_privilege is set. The new process now just calls the
4038 queue-runner function directly.
4044 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4047 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4048 ignored. Though the use of -oP was forcing the writing of a pid file, it
4049 was always written to the default place.
4051 3. If the message "no IP address found for host xxxx" is generated during
4052 incoming verification, it is now followed by identification of the incoming
4053 connection (so you can more easily find what provoked it).
4055 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4057 5. Added some features to "harden" Exim a bit more against certain attacks:
4059 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4060 be put in Local/Makefile. This is like the never_users runtime option,
4061 but it cannot be overridden. The default setting is "root".
4063 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4064 prefix string with which any file named in a -C command line option
4067 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4068 is retained for -C and -D only if the caller of Exim is root. Without
4069 it, the exim user may also use -C and -D and retain privilege.
4071 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4072 command line option is disabled.
4074 6. Macro names set by the -D option must start with an upper case letter, just
4075 like macro names defined in the configuration file.
4077 7. Added "dereference=" facility to LDAP.
4079 8. Two instances of the typo "uknown" in the source files are fixed.
4081 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4082 the Configure-Makefile script screwed up while processing it.
4084 10. Incorporated PCRE 4.4.
4086 11. The SMTP synchronization check was not operating right at the start of an
4087 SMTP session. For example, it could not catch a HELO sent before the client
4088 waited for the greeting. There is now a check for outstanding input at the
4089 point when the greeting is written. Because of the duplex, asynchronous
4090 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4091 way, but not yet received, when the check is performed.
4093 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4094 on TCP/IP sockets, because this apparently causes some broken clients to
4097 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4098 unchanged) from the Cygwin maintainer.
4100 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4101 was supported instead of testing for mbx. Effectively a typo.
4103 15. The spa authenticator server code was not checking that the input it
4104 received was valid base64.
4106 16. The debug output line for the "set" modifier in ACLs was not showing the
4107 name of the variable that was being set.
4109 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4111 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4112 lookup on its own. Something else was needed to trigger the lookup. For
4113 example, a match in host_lookup or the need for a host name in a host list.
4114 Now, if $sender_host_name is referenced and the host name has not yet been
4115 looked up, a lookup is performed. If the lookup fails, the variable remains
4116 empty, and $host_lookup_failed is set to "1".
4118 19. Added "eqi" as a case-independent comparison operator.
4120 20. The saslauthd authentication condition could segfault if neither service
4121 nor realm was specified.
4123 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4124 error message that was logged was misleading, and incoming SMTP
4125 connections were dropped. The message is now more accurate, and temporary
4126 errors are given to SMTP connections.
4128 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4129 (except RSET) with a 421 error, until QUIT is received. However, it was
4130 failing to send a response to QUIT.
4132 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4133 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4134 ACL is now run after the helo_try_verify_hosts code.
4136 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4137 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4138 case-independent, so other case variants are also recognized). Apparently
4139 some systems use these upper case variants.
4141 25. If more than two messages were waiting for the same host, and a transport
4142 filter was specified for the transport, Exim sent two messages over the
4143 same TCP/IP connection, and then failed with "socket operation on non-
4144 socket" when it tried to send the third.
4146 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4148 27. The extern definition of crypt16() in expand.c was not being excluded when
4149 the OS had its own crypt16() function.
4151 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4152 as a preferred synonym for return_size_limit, both as an option and as an
4155 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4157 30. Changed the default configuration ACL to relax the local part checking rule
4158 for addresses that are not in any local domains. For these addresses,
4159 slashes and pipe symbols are allowed within local parts, but the sequence
4160 /../ is explicitly forbidden.
4162 31. SPA server authentication was not clearing the challenge buffer before
4165 32. log_message in a "warn" ACL statement was writing to the reject log as
4166 well as to the main log, which contradicts the documentation and doesn't
4167 seem right (because no rejection is happening). So I have stopped it.
4169 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4170 However, I am unable to do any testing of this.
4172 34. Fixed an infelicity in the appendfile transport. When checking directories
4173 for a mailbox, to see if any needed to be created, it was accidentally
4174 using path names with one or more superfluous leading slashes; tracing
4175 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4177 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4178 contents are added to the log line that is written for every discarded
4179 recipient. (Previously a log_message setting was ignored.)
4181 36. The ${quote: operator now quotes the string if it is empty.
4183 37. The install script runs exim in order to find its version number. If for
4184 some reason other than non-existence or emptiness, which it checks, it
4185 could not run './exim', it was installing it with an empty version number,
4186 i.e. as "exim-". This error state is now caught, and the installation is
4189 38. An argument was missing from the function that creates an error message
4190 when Exim fails to connect to the socket for saslauthd authentication.
4191 This could cause Exim to crash, or give a corrupted message.
4193 39. Added isip, isip4, and isip6 to ${if conditions.
4195 40. The ACL variables $acl_xx are now saved with the message, and can be
4196 accessed later in routers, transports, and filters.
4198 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4199 strings in the file are not string-expanded.
4201 42. If a MAIL command specified a SIZE value that was too large to fit into an
4202 int variable, the check against message_size_limit failed. Such values are
4203 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4204 day this will have to be increased, but I don't think I want to be around
4205 when emails are that large.
4212 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4213 iconv() is not standard in FreeBSD.
4215 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4216 IPv6 enabled. The observed symptom was a segmentation fault on return from
4217 the function os_common_find_running_interfaces() in src/os.c.
4219 3. In the check_special_case() function in daemon.c I had used "errno" as an
4220 argument name, which causes warnings on some systems. This was basically a
4221 typo, since it was named "eno" in the comments!
4223 4. The code that waits for the clock to tick (at a resolution of some fraction
4224 of a second) so as to ensure message-id uniqueness was always waiting for
4225 at least one whole tick, when it could have waited for less. [This is
4226 almost certainly not relevant at current processor speeds, where it is
4227 unlikely to ever wait at all. But we try to future-proof.]
4229 5. The function that sleeps for a time interval that includes fractions of a
4230 second contained a race. It did not block SIGALRM between setting the
4231 timer, and suspending (a couple of lines later). If the interval was short
4232 and the sigsuspend() was delayed until after it had expired, the suspension
4233 never ended. On busy systems this could lead to processes getting stuck for
4236 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4237 process, before it forks any delivery processes. The open lookup caching
4238 mechanism meant that the open file or database connection was passed into
4239 the delivery process. The problem was that delivery processes always tidy
4240 up cached lookup data. This could cause a problem for the next delivery
4241 process started by the queue runner, because the external queue runner
4242 process does not know about the closure. So the next delivery process
4243 still has data in the lookup cache. In the case of a file lookup, there was
4244 no problem because closing a file descriptor in a subprocess doesn't affect
4245 the parent. However, if the lookup was caching a connection to a database,
4246 the connection was closed, and the second delivery process was likely to
4247 see errors such as "PGSQL: query failed: server closed the connection
4248 unexpectedly". The problem has been fixed by closing all cached lookups
4249 in a queue runner before running a delivery process.
4251 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4252 seem to have the "const" qualifier which it has on other OS. I've
4255 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4256 *of the same type* (client or server) with the same public name that an
4257 error should be diagnosed.
4259 9. When Exim looked up a host name for an IP address, but failed to find the
4260 original IP address when looking up the host name (a safety check), it
4261 output the message "<ip address> does not match any IP for NULL", which was
4262 confusing, to say the least. The bug was that the host name should have
4263 appeared instead of "NULL".
4265 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4266 user that is built into the binary, and the -C or -D options is used, root
4267 privilege is dropped before the configuration file is read. In addition,
4268 logging is switched to stderr instead of the normal log files. If the
4269 configuration then re-defines the Exim user, the unprivileged environment
4270 is probably not what is expected, so Exim logs a panic warning message (but
4273 However, if deliver_drop_privilege is set, the unprivileged state may well
4274 be exactly what is intended, so the warning has been cut out in that case,
4275 and Exim is allowed to try to write to its normal log files.
4281 1. smtp_return_error_details was not giving details for temporary sender
4282 or receiver verification errors.
4284 2. Diagnose a configuration error if two authenticators have the same public
4287 3. Exim used not to create the message log file for a message until the first
4288 delivery attempt. This could be confusing when incoming messages were held
4289 for policy or load reasons. The message log file is now created at the time
4290 the message is received, and an initial "Received" line is written to it.
4292 4. The automatically generated man page for command line options had a minor
4293 bug that caused no ill effects; however, a more serious problem was that
4294 the procedure for building the man page automatically didn't always
4295 operate. Consequently, release 4.20 contains an out-of-date version. This
4296 shouldn't happen again.
4298 5. When building Exim with embedded Perl support, the script that builds the
4299 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4300 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4302 6. The freeze_tell option was not being used for messages that were frozen on
4303 arrival, either by an ACL or by local_scan().
4305 7. Added the smtp_incomplete_transaction log selector.
4307 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4308 was accepting AUTH without a new EHLO.
4310 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4311 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4314 10. Logging of TCP/IP connections (when configured) now happens in the main
4315 daemon process instead of the child process, so that the TCP/IP connection
4316 count is more accurate (but it can never be perfect).
4318 11. The use of "drop" in a nested ACL was not being handled correctly in the
4319 outer ACL. Now, if condition failure induced by the nested "drop" causes
4320 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4321 or "require"), the connection is dropped.
4323 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4324 that yield "discard" can now be used with an "accept" or a "discard" verb,
4325 but an error is generated for any others (because I can't see a useful way
4326 to define what should happen).
4328 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4329 are now processed in the same way as lines in the Exim configuration file.
4330 In particular, continuation lines are supported.
4332 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4334 15. Added -ti meaning -t -i.
4336 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4337 domains, and warn by logging if others are found.
4339 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4340 returned from GIFCONF when Exim is trying to find the list of interfaces on
4341 a host. The code in os.c has been modified to copy each ifreq to an aligned
4342 structure in all cases.
4344 Also, in some cases, the returned ifreq's were being copied to a 'struct
4345 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4346 means the last couple of bytes of an IPv6 address could be chopped if the
4347 ifreq contained only a normal sockaddr (14 bytes storage).
4349 18. Named domain lists were not supported in the hosts_treat_as_local option.
4350 An entry such as +xxxx was not recognized, and was treated as a literal
4353 19. Ensure that header lines added by a DATA ACL are included in the reject log
4354 if the ACL subsequently rejects the message.
4356 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4357 MD5 (which is deprecated).
4359 21. When testing a filter file using -bf, Exim was writing a message when it
4360 took the sender from a "From " line in the message, but it was not doing so
4361 when it took $return_path from a Return-Path: header line. It now does.
4363 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4364 with a valid header line field name (a series of printing characters
4365 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4367 23. Changed "disc" in the source to "disk" to conform to the documentation and
4368 the book and for uniformity.
4370 24. Ignore Sendmail's -Ooption=value command line item.
4372 25. When execve() failed while trying to run a command in a pipe transport,
4373 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4374 could be confused with a return value of 69 from the command itself. This
4375 has been changed to 127, the value the shell returns if it is asked to run
4376 a non-existent command. The wording for the related log line suggests a
4377 non-existent command as the problem.
4379 26. If received_header_text expands to an empty string, do not add a Received:
4380 header line to the message. (Well, it adds a token one on the spool, but
4381 marks it "old" so that it doesn't get used or transmitted.)
4383 27. Installed eximstats 1.28 (addition of -nt option).
4385 28. There was no check for failure on the call to getsockname() in the daemon
4386 code. This can fail if there is a shortage of resources on the system, with
4387 ENOMEM, for example. A temporary error is now given on failure.
4389 29. Contrary to the C standard, it seems that in some environments, the
4390 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4391 program. Exim now does this explicitly; it affects the formatting of
4392 timestamps using strftime().
4394 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4395 complaints. I've now initialized all the variables, to avoid this.
4397 32. Header lines added by a system filter were not being "seen" during
4398 transport-time rewrites.
4400 33. The info_callback() function passed to OpenSSL is set up with type void
4401 (*)(SSL *, int, int), as described somewhere. However, when calling the
4402 function (actually a macro) that sets it up, the type void(*)() is
4403 expected. I've put in a cast to prevent warnings from picky compilers.
4405 34. If a DNS black list lookup found a CNAME record, but there were no A
4406 records associated with the domain it pointed at, Exim crashed.
4408 35. If a DNS black list lookup returned more than one A record, Exim ignored
4409 all but the first. It now scans all returned addresses if a particular IP
4410 value is being sought. In this situation, the contents of the
4411 $dnslist_value variable are a list of all the addresses, separated by a
4414 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4415 to accept a host name and all its aliases if the forward lookup for any of
4416 them yielded the IP address of the incoming connection. Now it accepts only
4417 those names whose forward lookup yields the correct IP address. Any other
4418 names are discarded. This closes a loophole whereby a rogue DNS
4419 administrator could create reverse DNS records to break through a
4420 wildcarded host restriction in an ACL.
4422 37. If a user filter or a system filter that ran in a subprocess used any of
4423 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4424 the wrong values were passed to the pipe command ($thisaddress had the
4425 value of $0, $0 had the value of $1, etc). This bug was introduced by
4426 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4428 38. Improved the line breaking for long SMTP error messages from ACLs.
4429 Previously, if there was no break point between 40 and 75 characters, Exim
4430 left the rest of the message alone. Two changes have been made: (a) I've
4431 reduced the minimum length to 35 characters; (b) if it can't find a break
4432 point between 35 and 75 characters, it looks ahead and uses the first one
4433 that it finds. This may give the occasional overlong line, but at least the
4434 remaining text gets split now.
4436 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4437 file descriptors might be low, and that setting 1000 would always raise it.
4438 It turns out that in some environments, the limit is already over 1000 and
4439 that lowering it causes trouble. So now Exim takes care not to decrease it.
4441 40. When delivering a message, the value of $return_path is set to $sender_
4442 address at the start of routing (routers may change the value). By an
4443 oversight, this default was not being set up when an address was tested by
4444 -bt or -bv, which affected the outcome if any router or filter referred to
4447 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4448 log only when "message" or "log_message" are set. However, if one of the
4449 conditions was an address verification, or a call to a nested ACL, the
4450 messages generated by the underlying test were being passed through. This
4451 no longer happens. The underlying message is available in $acl_verify_
4452 message for both "message" and "log_message" expansions, so it can be
4453 passed through if needed.
4455 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4456 new expansion $bh_ to give the encoded byte string without charset
4457 translation. Translation happens only if iconv() is available; HAVE_ICONV
4458 indicates this at build time. HEADERS_CHARSET gives the charset to
4459 translate to; headers_charset can change it in the configuration, and
4460 "headers charset" can change it in an individual filter file.
4462 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4463 that creates RFC 2047 encoded "words" labels them as that charset instead
4464 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4465 expansion operator; (ii) when Exim creates a From: line for a local
4466 message; (iii) when a header line is rewritten to include a "phrase" part.
4468 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4469 buggy, causing it to skip the first lines of messages whose message ID
4470 ended in 'D'. This would not have bitten before Exim release 4.14, because
4471 message IDs were unlikely to end in 'D' before then. The effect was to have
4472 incorrect size information for certain domains.
4474 45. #include "config.h" was missing at the start of the crypt16.c module. This
4475 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4478 46. If there was a timeout during a "random" callout check, Exim treated it as
4479 a failure of the random address, and carried on sending RSET and the real
4480 address. If the delay was just some slowness somewhere, the response to the
4481 original RCPT would be taken as a response to RSET and so on, causing
4482 mayhem of various kinds.
4484 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4485 when I implemented it. It didn't allow for the fact that some option values
4486 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4487 the right test for positive values.
4489 48. Domain names in DNS records are case-independent. Exim always looks them up
4490 in lower case. Some resolvers return domain names in exactly the case they
4491 appear in the zone file, that is, they may contain uppercase letters. Not
4492 all resolvers do this - some return always lower case. Exim was treating a
4493 change of case by a resolver as a change of domain, similar to a widening
4494 of a domain abbreviation. This triggered its re-routing code and so it was
4495 trying to route what was effectively the same domain again. This normally
4496 caused routing to fail (because the router wouldn't handle the domain
4497 twice). Now Exim checks for this case specially, and just changes the
4498 casing of the domain that it ultimately uses when it transmits the message
4501 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4504 50. If a filter generated a file delivery with a non-absolute name (possible if
4505 no home directory exists for the router), the forbid_file option was not
4508 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4509 the existing equality matching.
4511 52. Exim was using ints instead of ino_t variables in some places where it was
4512 dealing with inode numbers.
4514 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4515 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4516 called TMPDIR, and if it finds it is different, it changes its value.
4518 54. The smtp_printf() function is now made available to local_scan() so
4519 additional output lines can be written before returning. There is also an
4520 smtp_fflush() function to enable the detection of a dropped connection.
4521 The variables smtp_input and smtp_batched_input are exported to
4524 55. Changed the default runtime configuration: the message "Unknown user"
4525 has been removed from the ACL, and instead placed on the localuser router,
4526 using the cannot_route_message feature. This means that any verification
4527 failures that generate their own messages won't get overridden. Similarly,
4528 the "Unrouteable address" message that was in the ACL for unverifiable
4529 relay addresses has also been removed.
4531 56. Added hosts_avoid_esmtp to the smtp transport.
4533 57. The exicyclog script was not checking for the esoteric option
4534 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4535 will work only if exicyclog is run under the appropriate euid.
4537 58. Following a discussion on the list, the rules by which Exim recognises line
4538 endings on incoming messages have been changed. The -dropcr and drop_cr
4539 options are now no-ops, retained only for backwards compatibility. The
4540 following line terminators are recognized: LF CRLF CR. However, special
4541 processing applies to CR:
4543 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4544 nor a local message in the state where . is a terminator.
4546 (ii) If a bare CR is encountered in a header line, an extra space is added
4547 after the line terminator so as not to end the header. The reasoning
4548 behind this is that bare CRs in header lines are most likely either
4549 to be mistakes, or people trying to play silly games.
4551 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4552 was being incorrectly given as 18 bytes larger than it should have been.
4553 This is a VOB (very old bug).
4555 60. This may never have affected anything current, but just in case it has:
4556 When the local host is found other than at the start of a list of hosts,
4557 the local host, those with the same MX, and any that follow, are discarded.
4558 When the list in question was part of a longer list of hosts, the following
4559 hosts (not currently being processed) were also being discarded. This no
4560 longer happens. I'm not sure if this situation could ever has previously
4563 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4566 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4567 header. This is something that is recommended in a new Internet Draft, and
4568 is something that is documented as being done by Sendmail. There are two
4569 possible values. For messages generated by the autoreply transport, Exim
4572 Auto-Submitted: auto-replied
4574 whereas for all other generated messages (e.g. bounces) it adds
4576 Auto-Submitted: auto-generated
4578 63. The "personal" condition in filters now includes a test for the
4579 Auto-Submitted: header. If it contains the string "auto-" the message it
4580 not considered personal.
4582 64. Added rcpt_include_affixes as a generic transport option.
4584 65. Added queue_only_override (default true).
4586 66. Added the syslog_duplication option.
4588 67. If what should have been the first header line of a message consisted of
4589 a space followed by a colon, Exim was mis-interpreting it as a header line.
4590 It isn't of course - it is syntactically invalid and should therefore be
4591 treated as the start of the message body. The misbehaviour could have
4592 caused a number of strange effects, including loss of data in subsequent
4593 header lines, and spool format errors.
4595 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4596 client host had authenticated. This control can now be exercised by an ACL
4597 for more flexibility.
4599 69. By default, callouts do not happen when testing with -bh. There is now a
4600 variant, -bhc, which does actually run the callout code, including
4601 consulting and updating the callout cache.
4603 70. Added support for saslauthd authentication, courtesy of Alexander
4606 71. If statvfs() failed on the spool or log directories while checking their
4607 size for availability, Exim confusingly gave the error "space shortage".
4608 Furthermore, in debugging mode it crashed with a floating point exception.
4609 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4610 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4611 serious problem, Exim now writes to the main and panic logs when this
4612 happens, with details of the failure. It then refuses to accept the
4613 incoming message, giving the message "spool directory problem" or "log
4614 directory problem" with a 421 code for SMTP messages.
4616 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4617 0, 1, and 2 exist, because some OS complain for execs without them (see
4618 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4619 descriptors. However, the code omitted to check that the open succeeded,
4620 causing mysterious errors if for some reason the permissions on /dev/null
4621 got screwed. Now Exim writes a message to the main and panic logs, and
4622 bombs out if it can't open /dev/null.
4624 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4625 interact so that it is all more flexible. It is supposed to remain
4626 backwards compatible. Also added extra_local_interfaces.
4628 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4629 to bomb out with an assertion failure - to the client this appears as a
4630 connection drop. This problem occurs in the part of the code that was taken
4631 from the Samba project. Fortunately, the assertion is in a very simple
4632 function, so I have fixed this by reproducing the function inline in the
4633 one place where it is called, and arranging for authentication to fail
4634 instead of killing the process with assert().
4636 75. The SPA client code was not working when the server requested OEM rather
4637 than Unicode encoding.
4639 76. Added code to make require_files with a specific uid setting more usable in
4640 the case where statting the file as root fails - usually a non-root-mounted
4641 NFS file system. When this happens and the failure is EACCES, Exim now
4642 forks a subprocess and does the per-uid checking as the relevant uid.
4644 77. Added process_log_path.
4646 78. If log_file_path was not explicitly set, a setting of check_log_space or
4647 check_log_inodes was ignored.
4649 79. If a space check for the spool or log partitions fails, the incident is now
4650 logged. Of course, in the latter case the data may get lost...
4652 80. Added the %p formatting code to string_format() so that it can be used to
4653 print addresses in debug_print(). Adjusted all the address printing in the
4654 debugging in store.c to use %p rather than %d.
4656 81. There was a concern that a line of code in smtp_in.c could overflow a
4657 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4658 initially expressed, the concern was not well-founded, because trailing
4659 spaces are removed early. However, if the trailing spaces were followed by
4660 a NULL, they did not get removed, so the overflow was possible. Two fixes
4663 (a) I re-wrote the offending code in a cleaner fashion.
4664 (b) If an incoming SMTP command contains a NULL character, it is rejected
4667 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4668 runs initgroups(), so that if the Exim user is in any additional groups,
4669 they will be used during message reception.
4675 The change log for 4.20 and earlier releases has been archived.