1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.494 2007/03/13 16:37:57 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.
159 PH/37 If a message is not accepted after it has had an id assigned (e.g.
160 because it turns out to be too big or there is a timeout) there is no
161 "Completed" line in the log. When some messages of this type were
162 selected by exigrep, they were listed as "not completed". Others were
163 picked up by some special patterns. I have improved the selection
164 criteria to be more general.
166 PH/38 The host_find_failed option in the manualroute router can now be set
167 to "ignore", to completely ignore a host whose IP address cannot be
168 found. If all hosts are ignored, the behaviour is controlled by the new
169 host_all_ignored option.
175 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
176 fixed by 4.65/MH/01 (is this a record?) are fixed:
178 (i) An empty string was always treated as zero by the numeric comparison
179 operators. This behaviour has been restored.
181 (ii) It is documented that the numeric comparison operators always treat
182 their arguments as decimal numbers. This was broken in that numbers
183 starting with 0 were being interpreted as octal.
185 While fixing these problems I realized that there was another issue that
186 hadn't been noticed. Values of message_size_limit (both the global option
187 and the transport option) were treated as octal if they started with 0.
188 The documentation was vague. These values are now always treated as
189 decimal, and I will make that clear in the documentation.
195 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
196 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
199 MH/01 Don't check that the operands of numeric comparison operators are
200 integers when their expansion is in "skipping" mode (fixes bug
201 introduced by 4.64-PH/07).
203 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
204 child addresses, Exim now panics and dies. Previously, because the count
205 is held in a short int, deliveries were likely to be lost. As such a
206 large number of recipients for a single message is ridiculous
207 (performance will be very, very poor), I have chosen to impose a limit
208 rather than extend the field.
214 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
215 leftover -K file (the existence of which was triggered by #402).
216 While we were at it, introduced process PID as part of the -K
217 filename. This should rule out race conditions when creating
220 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
221 processing considerably. Previous code took too long for large mails,
222 triggering a timeout which in turn triggers #401.
224 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
225 in the DK code in transports.c. sendfile() is not really portable,
226 hence the _LINUX specificness.
228 TF/01 In the add_headers option to the mail command in an Exim filter,
229 there was a bug that Exim would claim a syntax error in any
230 header after the first one which had an odd number of characters
233 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
234 callout verification, Exim cached a "reject" for the entire domain. This
235 is correct for most verifications, but it is not correct for a recipient
236 verification with use_sender or use_postmaster set, because in that case
237 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
238 case of MAIL FROM:<> rejection from other early rejections (e.g.
239 rejection of HELO). When verifying a recipient using a non-null MAIL
240 address, the cache is ignored if it shows MAIL FROM:<> rejection.
241 Whatever the result of the callout, the value of the domain cache is
242 left unchanged (for any other kind of callout, getting as far as trying
243 RCPT means that the domain itself is ok).
245 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
248 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
249 manifest itself as EPIPE rather than ECONNECT. When tidying away a
250 session, the daemon ignores ECONNECT errors and logs others; it now
251 ignores EPIPE as well.
253 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
254 (quoted-printable decoding).
256 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
257 later the small subsequent patch to fix an introduced bug.
259 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
261 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
263 PH/08 An error is now given if message_size_limit is specified negative.
265 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
266 to be given (somewhat) arbitrary names.
268 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
271 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
272 miscellaneous code fixes
274 PH/10 Added the log_reject_target ACL modifier to specify where to log
277 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
278 hostname. This is wrong, because it relates to the incoming message (and
279 probably the interface on which it is arriving) and not to the outgoing
280 callout (which could be using a different interface). This has been
281 changed to use the value of the helo_data option from the smtp transport
282 instead - this is what is used when a message is actually being sent. If
283 there is no remote transport (possible with a router that sets up host
284 addresses), $smtp_active_hostname is used.
286 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
287 tweaks were necessary in order to get it to work (see also 21 below):
288 (a) The code assumed that strncpy() returns a negative number on buffer
289 overflow, which isn't the case. Replaced with Exim's string_format()
291 (b) There were several signed/unsigned issues. I just did the minimum
292 hacking in of casts. There is scope for a larger refactoring.
293 (c) The code used strcasecmp() which is not a standard C function.
294 Replaced with Exim's strcmpic() function.
295 (d) The code set only $1; it now sets $auth1 as well.
296 (e) A simple test gave the error "authentication client didn't specify
297 service in request". It would seem that Dovecot has changed its
298 interface. Fortunately there's a specification; I followed it and
299 changed what the client sends and it appears to be working now.
301 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
304 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
305 address A is aliased to B and C, where B exists and C does not. Without
306 -v the output is "A verified" because verification stops after a
307 successful redirection if more than one address is generated. However,
308 with -v the child addresses are also verified. Exim was outputting "A
309 failed to verify" and then showing the successful verification for C,
310 with its parentage. It now outputs "B failed to verify", showing B's
311 parentage before showing the successful verification of C.
313 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
314 look up a TXT record in a specific list after matching in a combined
317 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
318 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
319 they consult the DNS. I had assumed they would set it the way they
320 wanted; and indeed my experiments on Linux seem to show that in some
321 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
322 To be on the safe side, however, I have now made the interface to
323 host_find_byname() similar to host_find_bydns(), with an argument
324 containing the DNS resolver options. The host_find_byname() function now
325 sets these options at its start, just as host_find_bydns() does. The smtp
326 transport options dns_qualify_single and dns_search_parents are passed to
327 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
328 of host_find_byname() use the default settings of RES_DEFNAMES
329 (qualify_single) but not RES_DNSRCH (search_parents).
331 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
332 spool_read_header() do less string testing, by means of a preliminary
333 switch on the second character of optional "-foo" lines. (This is
334 overdue, caused by the large number of possibilities that now exist.
335 Originally there were few.) While I was there, I also converted the
336 str(n)cmp tests so they don't re-test the leading "-" and the first
337 character, in the hope this might squeeze out yet more improvement.
339 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
340 flag allowing group syntax was set by the header_syntax check but not
341 turned off, possible causing trouble later; (2) The flag was not being
342 set at all for the header_verify test, causing "group"-style headers to
343 be rejected. I have now set it in this case, and also caused header_
344 verify to ignore an empty address taken from a group. While doing this, I
345 came across some other cases where the code for allowing group syntax
346 while scanning a header line wasn't quite right (mostly, not resetting
347 the flag correctly in the right place). These bugs could have caused
348 trouble for malformed header lines. I hope it is now all correct.
350 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
351 with the "reply" argument non-NULL. The code, however (which originally
352 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
353 but it didn't always do it. This confused somebody who was copying the
354 code for some other use. I have removed all the tests.
356 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
357 feature that was used to support insecure browsers during the U.S. crypto
358 embargo. It requires special client support, and Exim is probably the
359 only MTA that supported it -- and would never use it because real RSA is
360 always available. This code has been removed, because it had the bad
361 effect of slowing Exim down by computing (never used) parameters for the
362 RSA_EXPORT functionality.
364 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
365 authenticator to fail if there's a tab character in the incoming data
366 (there should never be unless someone is messing about, as it's supposed
367 to be base64-encoded). Also added, on Timo's advice, the "secured" option
368 if the connection is using TLS or if the remote IP is the same as the
369 local IP, and the "valid-client-cert option" if a client certificate has
372 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
373 authenticators. This can be used for authorization after authentication
374 succeeds. (In the case of plaintext, it servers for both authentication
377 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
378 if any retry times were supplied.
380 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
381 connection, where there is no client IP address to check. In this
382 situation, the verify now always succeeds.
384 PH/25 Applied John Jetmore's -Mset patch.
386 PH/26 Added -bem to be like -Mset, but loading a message from a file.
388 PH/27 In a string expansion for a processed (not raw) header when multiple
389 headers of the same name were present, leading whitespace was being
390 removed from all of them, but trailing whitespace was being removed only
391 from the last one. Now trailing whitespace is removed from each header
392 before concatenation. Completely empty headers in a concatenation (as
395 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
396 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
398 PH/29 [Removed. This was a change that I later backed out, and forgot to
399 correct the ChangeLog entry (that I had efficiently created) before
400 committing the later change.]
402 PH/30 Exim was sometimes attempting to deliver messages that had suffered
403 address errors (4xx response to RCPT) over the same connection as other
404 messages routed to the same hosts. Such deliveries are always "forced",
405 so retry times are not inspected. This resulted in far too many retries
406 for the affected addresses. The effect occurred only when there were more
407 hosts than the hosts_max_try setting in the smtp transport when it had
408 the 4xx errors. Those hosts that it had tried were not added to the list
409 of hosts for which the message was waiting, so if all were tried, there
410 was no problem. Two fixes have been applied:
412 (i) If there are any address or message errors in an SMTP delivery, none
413 of the hosts (tried or untried) are now added to the list of hosts
414 for which the message is waiting, so the message should not be a
415 candidate for sending over the same connection that was used for a
416 successful delivery of some other message. This seems entirely
417 reasonable: after all the message is NOT "waiting for some host".
418 This is so "obvious" that I'm not sure why it wasn't done
419 previously. Hope I haven't missed anything, but it can't do any
420 harm, as the worst effect is to miss an optimization.
422 (ii) If, despite (i), such a delivery is accidentally attempted, the
423 routing retry time is respected, so at least it doesn't keep
424 hammering the server.
426 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
427 in ${readsocket because some servers need this prod.
429 PH/32 Added some extra debug output when updating a wait-xxx database.
431 PH/33 The hint "could be header name not terminated by colon", which has been
432 given for certain expansion errors for a long time, was not being given
433 for the ${if def:h_colon_omitted{... case.
435 PH/34 The spec says: "With one important exception, whenever a domain list is
436 being scanned, $domain contains the subject domain." There was at least
437 one case where this was not true.
439 PH/35 The error "getsockname() failed: connection reset by peer" was being
440 written to the panic log as well as the main log, but it isn't really
441 panic-worthy as it just means the connection died rather early on. I have
442 removed the panic log writing for the ECONNRESET error when getsockname()
445 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
446 runs only) independently of the message's sender address. This meant
447 that, if the 4xx error was in fact related to the sender, a different
448 message to the same recipient with a different sender could confuse
449 things. In particualar, this can happen when sending to a greylisting
450 server, but other circumstances could also provoke similar problems.
451 I have changed the default so that the retry time for these errors is now
452 based a combination of the sender and recipient addresses. This change
453 can be overridden by setting address_retry_include_sender=false in the
456 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
457 remote server are returned as part of bounce messages. This was not
458 happening for LMTP over a pipe (the lmtp transport), but now it is the
459 same for both kinds of LMTP.
461 PH/38 Despite being documented as not happening, Exim was rewriting addresses
462 in header lines that were in fact CNAMEs. This is no longer the case.
464 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
465 and queue runs started by the daemon processed all messages. This has
466 been fixed so that -R and -S can now usefully be given with -q<time>.
468 PH/40 Import PCRE release 6.7 (fixes some bugs).
470 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
472 PH/42 Give an error if -q is specified more than once.
474 PH/43 Renamed the variables $interface_address and $interface_port as
475 $received_ip_address and $received_port, to make it clear that these
476 values apply to message reception, and not to the outgoing interface when
477 a message is delivered. (The old names remain recognized, of course.)
479 PH/44 There was no timeout on the connect() call when using a Unix domain
480 socket in the ${readsocket expansion. There now is.
482 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
483 be meaningful with "accept".
485 SC/01 Eximstats V1.43
486 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
488 SC/02 Eximstats V1.44
489 Use a glob alias rather than an array ref in the generated
490 parser. This improves both readability and performance.
492 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
493 Collect SpamAssassin and rejection statistics.
494 Don't display local sender or destination tables unless
495 there is data to show.
496 Added average volumes into the top table text output.
498 SC/04 Eximstats V1.46
499 Collect data on the number of addresses (recipients)
500 as well as the number of messages.
502 SC/05 Eximstats V1.47
503 Added 'Message too big' to the list of mail rejection
504 reasons (thanks to Marco Gaiarin).
506 SC/06 Eximstats V1.48
507 Mainlog lines which have GMT offsets and are too short to
508 have a flag are now skipped.
510 SC/07 Eximstats V1.49 (Alain Williams)
511 Added the -emptyok flag.
513 SC/08 Eximstats V1.50
514 Fixes for obtaining the IP address from reject messages.
516 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
517 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
518 whitesspace changes from 4.64-PH/27
520 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
523 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
524 are found, allow negative numbers in numeric criteria)
526 JJ/06 exipick.20061117.2, added new $message_body_missing variable
528 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
529 to match changes made in 4.64-PH/43
531 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
533 PH/47 Put in an explicit test for a DNS lookup of an address record where the
534 "domain" is actually an IP address, and force a failure. This locks out
535 those revolvers/nameservers that support "A-for-A" lookups, in
536 contravention of the specifications.
538 PH/48 When a host name was looked up from an IP address, and the subsequent
539 forward lookup of the name timed out, the host name was left in
540 $sender_host_name, contrary to the specification.
542 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
543 restricted to single-key lookups, Exim was not diagnosing an error if
544 * or *@ was used with a query-style lookup.
546 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
548 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
549 long ago, but noone interested enough thought of it. Let's just say that
550 the "1.1" means that there are some new functions that weren't there at
551 some point in the past.
553 PH/51 Error processing for expansion failure of helo_data from an smtp
554 transport during callout processing was broken.
556 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
557 tested/used via the -bh/-bhc/-bs options.
559 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
560 bug, fixed in subsequent PCRE releases).
562 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
563 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
565 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
571 SC/01 Use a glob alias rather than an array ref in eximstats generated
572 parser. This improves both readability and performance.
574 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
575 Don't display local sender or destination tables in eximstats unless
576 there is data to show.
577 Added average volumes into the eximstats top table text output.
579 SC/03 Collect data on the number of addresses (recipients) as well
580 as the number of messages in eximstats.
582 TF/01 Correct an error in the documentation for the redirect router. Exim
583 does (usually) call initgroups() when daemonizing.
585 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
586 with consistent privilege compared to when running as a daemon.
588 TF/03 Note in the spec that $authenticated_id is not set for local
589 submissions from trusted users.
591 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
592 Thanks to Dean Brooks <dean@iglou.com> for the patch.
594 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
595 by adding some example configuration directives to the default
596 configuration file. A little bit of work is required to uncomment the
597 directives and define how usernames and passwords are checked, but
598 there is now a framework to start from.
600 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
601 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
602 without this. I don't know how relevant this is to other LDAP libraries.
604 PH/02 Add the verb name to the "unknown ACL verb" error.
606 PH/03 Magnus Holmgren's patch for filter_prepend_home.
608 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
610 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
611 directory not expanded when it should be if an expanded home directory
612 was set for the address (which is overridden by the transport).
614 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
617 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
618 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
619 because it is too late at that time, and has no effect.
621 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
622 security issue with \' (bugzilla #107). I could not use the
623 PQescapeStringConn() function, because it needs a PGconn value as one of
626 PH/08 When testing addresses using -bt, indicate those final addresses that
627 are duplicates that would not cause an additional delivery. At least one
628 person was confused, thinking that -bt output corresponded to deliveries.
629 (Suppressing duplicates isn't a good idea as you lose the information
630 about possibly different redirections that led to the duplicates.)
632 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
633 systems where poll() doesn't work, in particular OS X.
635 PH/10 Added more information to debugging output for retry time not reached.
637 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
638 operations in malware.c.
640 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
643 PH/13 If write_rejectlog was set false when logging was sent to syslog with
644 syslog_duplication set false, log lines that would normally be written
645 both the the main log and to the reject log were not written to syslog at
648 PH/14 In the default configuration, change the use of "message" in ACL warn
649 statements to "add_header".
651 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
652 not followed by a command (e.g. "seen endif").
654 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
655 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
658 PH/17 Added extra conditions to the default value of delay_warning_condition
662 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
663 { match{$h_precedence:}{(?i)bulk|list|junk} } \
664 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
667 The Auto-Submitted: and various List- headers are standardised, whereas I
668 don't think Precedence: ever was.
670 PH/18 Refactored debugging code in route_finduser() to show more information,
671 in particular, the error code if getpwnam() issues one.
673 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
674 This is apparently needed in addition to the PH/07 change above to avoid
675 any possible encoding problems.
677 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
678 but not after initializing Perl.
680 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
681 output them only if debugging. By default they are written stderr,
682 apparently, which is not desirable.
684 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
687 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
690 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
692 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
693 authenticated or an ident call has been made. Suppress the default
694 values for $authenticated_id and $authenticated_sender (but permit -oMai
695 and -oMas) when testing with -bh.
697 PH/24 Re-jigged the order of the tests in the default configuration so that the
698 tests for valid domains and recipients precede the DNS black list and CSA
699 tests, on the grounds that those ones are more expensive.
701 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
702 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
703 This bug exists in every version of Exim that I still have, right back to
706 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
707 However, an attempt to turn on case-sensitivity in a regex key by
708 including (?-i) didn't work because the subject string was already
709 lowercased, and the effects were non-intuitive. It turns out that a
710 one-line patch can be used to allow (?-i) to work as expected.
716 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
717 other effects) broke the use of negated acl sub-conditions.
719 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
722 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
723 "Deny" causes Exim to reject the incoming connection with a 554 error.
724 Unfortunately, if there is a major crisis, such as a disk failure,
725 tcp-wrappers gives "deny", whereas what one would like would be some
726 kind of temporary error. A kludge has been added to help with this.
727 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
728 554 error is used if errno is still zero or contains ENOENT (which occurs
729 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
732 PH/03 Add -lutil to the default FreeBSD LIBS setting.
734 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
735 errors. Otherwise a message that provokes a temporary error (when other
736 messages do not) can cause a whole host to time out.
738 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
739 addresses were routed directly to files or pipes from a redirect router.
740 File deliveries just didn't batch; pipe deliveries might have suffered
743 PH/06 A failure to get a lock for a hints database would erroneously always say
744 "Failed to get write lock", even when it was really a read lock.
746 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
747 of 0600. This has been changed to use the value of the lockfile_mode
748 option (which defaults to 0600).
750 PH/08 Applied small patch from the Sieve maintainer.
752 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
753 folder from quota calculations, a direct delivery into this folder messed
754 up the contents of the maildirsize file. This was because the regex was
755 used only to exclude .Trash (or whatever) when the size of the mailbox
756 was calculated. There was no check that a delivery was happening into an
757 excluded directory. This bug has been fixed by ignoring all quota
758 processing for deliveries into excluded directories.
760 PH/10 Added the maildirfolder_create_regex option to appendfile.
766 PH/01 The code for finding all the local interface addresses on a FreeBSD
767 system running IPv6 was broken. This may well have applied to all BSD
768 systems, as well as to others that have similar system calls. The broken
769 code found IPv4 interfaces correctly, but gave incorrect values for the
770 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
771 that it would not match correctly against @[] and not recognize the IPv6
774 PH/02 The ipliteral router was not recognizing addresses of the form user@
775 [ipv6:....] because it didn't know about the "ipv6:" prefix.
777 PH/03 Added disable_ipv6.
779 PH/04 Changed $reply_address to use the raw form of the headers instead of the
780 decoded form, because it is most often used to construct To: headers
781 lines in autoreplies, and the decoded form may well be syntactically
782 invalid. However, $reply_address has leading white space removed, and all
783 newlines turned into spaces so that the autoreply transport does not
786 PH/05 If group was specified without a user on a router, and no group or user
787 was specified on a transport, the group from the router was ignored.
789 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
790 for visible compile-time settings that can be used to change these
791 numbers, for those that want even more. Backwards compatibility with old
792 spool files has been maintained. However, going back to a previous Exim
793 release will lost any variables that are in spool files.
795 PH/07 Two small changes when running in the test harness: increase delay when
796 passing a TCP/IP connection to a new process, in case the original
797 process has to generate a bounce, and remove special handling of
798 127.0.0.2 (sic), which is no longer necessary.
800 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
801 be the same on different OS.
803 PH/09 Moved a debug statement in filter processing to avoid a race problem when
806 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
807 whether --show-vars was specified or not
809 JJ/02 exipick: Added support for new ACL variable spool format introduced
812 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
813 syntactically invalid From: or Reply-to: line, and a filter used this to
814 generate an autoreply, and therefore failed to obtain an address for the
815 autoreply, Exim could try to deliver to a non-existent relative file
816 name, causing unrelated and misleading errors. What now happens is that
817 it logs this as a hard delivery error, but does not attempt to create a
820 PH/11 The exinext utility has a -C option for testing purposes, but although
821 the given file was scanned by exinext itself; it wasn't being passed on
822 when Exim was called.
824 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
825 an end-of-file indication when reading a command response.
827 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
828 compiled. In many other places in Exim, IPv6 addresses are always
829 recognized, so I have changed this. It also means that IPv4 domain
830 literals of the form [IPV4:n.n.n.n] are now always recognized.
832 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
833 used if the router is not running as root, for example, when verifying at
834 ACL time, or when using -bh. The debugging output from this situation was
835 non-existent - all you got was a failure to exec. I have made two
838 (a) Failures to set uid/gid, the current directory, or a process leader
839 in a subprocess such as that created by queryprogram now generate
840 suitable debugging ouput when -d is set.
842 (b) The queryprogram router detects when it is not running as root,
843 outputs suitable debugging information if -d is set, and then runs
844 the subprocess without attempting to change uid/gid.
846 PH/15 Minor change to Makefile for building test_host (undocumented testing
849 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
850 additional section of a DNS packet that returns MX or SRV records.
851 Instead, it always explicitly searches for A/AAAA records. This avoids
852 major problems that occur when a DNS server includes only records of one
853 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
854 fixed another bug: if SRV records were looked up and the corresponding
855 address records were *not* found in the additional section, the port
856 values from the SRV records were lost.
858 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
859 using the correct key (the original address) when searching the retry
860 rules in order to find which one to use for generating the retry hint.
862 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
863 adding the default one. Similarly, if it contains a Reply-To: header, the
864 errors_reply_to option, if set, is not used.
866 PH/19 When calculating a retry time, Exim used to measure the "time since
867 failure" by looking at the "first failed" field in the retry record. Now
868 it does not use this if it is later than than the arrival time of the
869 message. Instead it uses the arrival time. This makes for better
870 behaviour in cases where some deliveries succeed, thus re-setting the
871 "first failed" field. An example is a quota failure for a huge message
872 when small messages continue to be delivered. Without this change, the
873 "time since failure" will always be short, possible causing more frequent
874 delivery attempts for the huge message than are intended.
875 [Note: This change was subsequently modified - see PH/04 for 4.62.]
877 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
878 $1, $2, $3) because the numerical variables can be reset during some
879 expansion items (e.g. "match"), thereby losing the authentication data.
881 PH/21 Make -bV show the size of off_t variables so that the test suite can
882 decide whether to run tests for quotas > 2G.
884 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
885 mailbox_size, and mailbox_filecount in the appendfile transport. If a
886 filecount value is greater than 2G or if a quota value is greater than 2G
887 on a system where the size of off_t is not greater than 4, a panic error
890 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
891 never match. The debug and -bh output now contains an explicit error
892 message indicating a malformed IPv4 address or mask.
894 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
895 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
898 PH/25 Do not write to syslog when running in the test harness. The only
899 occasion when this arises is a failure to open the main or panic logs
900 (for which there is an explicit test).
902 PH/26 Added the /no_tell option to "control=freeze".
904 PH/27 If a host name lookup failed very early in a connection, for example, if
905 the IP address matched host_lookup and the reverse lookup yielded a name
906 that did not have a forward lookup, an error message of the form "no IP
907 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
908 could be logged. Now it outputs the IP address instead of "NULL".
910 PH/28 An enabling patch from MH: add new function child_open_exim2() which
911 allows the sender and the authenticated sender to be set when
912 submitting a message from within Exim. Since child_open_exim() is
913 documented for local_scan(), the new function should be too.
915 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
916 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
917 results in an empty string is now treated as unset.
919 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
921 PH/31 Added sender_verify_fail logging option.
923 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
924 needed by RFC 822 but not by RFC 2822 was commented out. I have now
925 tidied the source and removed it altogether.
927 PH/33 When a queue run was abandoned because the load average was too high, a
928 log line was always written; now it is written only if the queue_run log
929 selector is set. In addition, the log line for abandonment now contains
930 information about the queue run such as the pid. This is always present
931 in "start" and "stop" lines but was omitted from the "abandon" line.
933 PH/34 Omit spaces between a header name and the colon in the error message that
934 is given when verify = headers_syntax fails (if there are lots of them,
935 the message gets confusing).
937 PH/35 Change the default for dns_check_names_pattern to allow slashes within
938 names, as there are now some PTR records that contain slashes. This check
939 is only to protect against broken name servers that fall over on strange
940 characters, so the fact that it applies to all lookups doesn't matter.
942 PH/36 Now that the new test suite is complete, we can remove some of the
943 special code in Exim that was needed for the old test suite. For example,
944 sorting DNS records because real resolvers return them in an arbitrary
945 order. The new test suite's fake resolver always returns records in the
948 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
949 bounces) except when queue_only is set, to avoid logging races between
950 the different processes.
952 PH/38 Panic-die if .include specifies a non-absolute path.
954 PH/39 A tweak to the "H" retry rule from its user.
956 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
957 a label. They prevented compilation on older perls.
959 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
960 a warning to be raised on newish perls.
962 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
963 on queue. Changes to match documented behaviour of showing count of
964 messages matching specified criteria.
966 PH/40 Changed the default ident timeout from 30s to 5s.
968 PH/41 Added support for the use of login_cap features, on those BSD systems
969 that have them, for controlling the resources used by pipe deliveries.
971 PH/42 The content-scanning code uses fopen() to create files in which to put
972 message data. Previously it was not paying any attention to the mode of
973 the files. Exim runs with umask(0) because the rest of the code creates
974 files with open(), and sets the required mode explicitly. Thus, these
975 files were ending up world-writeable. This was not a big issue, because,
976 being within the spool directory, they were not world-accessible. I have
977 created a function called modefopen, which takes an additional mode
978 argument. It sets umask(777), creates the file, chmods it to the required
979 mode, then resets the umask. All the relevant calls to fopen() in the
980 content scanning code have been changed to use this function.
982 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
983 to 24 hours. This avoids potential overflow problems when processing G
984 and H retry rules. I suspect nobody ever tinkers with this value.
986 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
988 PH/45 When the plaintext authenticator is running as a client, the server's
989 challenges are checked to ensure they are valid base64 strings. By
990 default, the authentication attempt is cancelled if an invalid string is
991 received. Setting client_ignore_invalid_base64 true ignores these errors.
992 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
993 they are received. Thus, the responses can be made to depend on the
994 challenges. If an invalid string is ignored, an empty string is placed in
997 PH/46 Messages that are created by the autoreply transport now contains a
998 References: header, in accordance with RFCs 2822 and 3834.
1000 PH/47 Added authenticated_sender_force to the smtp transport.
1002 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1004 PH/49 Installed latest patch from the Sieve maintainer.
1006 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1007 was also set, the appendfile transport was unnecessarily scanning a
1008 directory of message files (e.g. for maildir delivery) to find the count
1009 of files (along with the size), even though it did not need this
1010 information. It now does the scan only if it needs to find either the
1011 size of the count of files.
1013 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1015 PH/52 Two bugs concerned with error handling when the smtp transport is
1018 (i) Exim was not creating retry information for temporary errors given
1019 for individual recipients after the DATA command when the smtp transport
1020 was used in LMTP mode. This meant that they could be retried too
1021 frequently, and not timed out correctly.
1023 (ii) Exim was setting the flag that allows error details to be returned
1024 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1025 recipients that were returned after the DATA command.
1027 PH/53 This is related to PH/52, but is more general: for any failing address,
1028 when detailed error information was permitted to be returned to the
1029 sender, but the error was temporary, then after the final timeout, only
1030 "retry timeout exceeded" was returned. Now it returns the full error as
1031 well as "retry timeout exceeded".
1033 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1034 do this, and (what is worse) MTAs that accept it.
1036 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1037 will now be deprecated.
1039 PH/56 New os.c-cygwin from the Cygwin maintainer.
1041 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1042 formats (previously only available in exim formats via -bpr, -bpru,
1043 and -bpra. Now also available in native and exiqgrep formats)
1045 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1046 with very large, slow to parse queues
1048 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1050 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1052 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1053 responses to authentication challenges, though it was showing the
1054 challenges; (ii) I've removed the CR characters from the debug output for
1057 PH/58 Allow for the insertion of a newline as well as a space when a string
1058 is turned into more than one encoded-word during RFC 2047 encoding. The
1059 Sieve code now uses this.
1061 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1062 data_4xx, lost_connection, tls_required.
1064 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1065 message was being sent as an SMTP response.
1067 PH/61 Add -l and -k options to exicyclog.
1069 PH/62 When verifying, if an address was redirected to one new address, so that
1070 verification continued, and the new address failed or deferred after
1071 having set something in $address_data, the value of $address_data was not
1072 passed back to the ACL. This was different to the case when no
1073 redirection occurred. The value is now passed back in both cases.
1075 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1076 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1077 use login_cap.h, so on its own it isn't the distinguishing feature. The
1078 new name refers directly to the setclassresources() function.
1080 PH/65 Added configuration files for NetBSD3.
1082 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1084 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1085 is preferred over IPv4.
1087 PH/68 The bounce_return_message and bounce_return_body options were not being
1088 honoured for bounces generated during the reception of non-SMTP messages.
1089 In particular, this applied to messages rejected by the ACL. This bug has
1090 been fixed. However, if bounce_return_message is true and bounce_return_
1091 body is false, the headers that are returned for a non-SMTP message
1092 include only those that have been read before the error was detected.
1093 (In the case of an ACL rejection, they have all been read.)
1095 PH/69 The HTML version of the specification is now built in a directory called
1096 spec_html instead of spec.html, because the latter looks like a path with
1097 a MIME-type, and this confuses some software.
1099 PH/70 Catch two compiler warnings in sieve.c.
1101 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1102 function verify_get_ident() calls ip_connect() to connect a socket, but
1103 if the "connect()" function timed out, ip_connect() used to close the
1104 socket. However, verify_get_ident() also closes the socket later, and in
1105 between Exim writes to the log, which may get opened at this point. When
1106 the socket was closed in ip_connect(), the log could get the same file
1107 descriptor number as the socket. This naturally causes chaos. The fix is
1108 not to close the socket in ip_connect(); the socket should be closed by
1109 the function that creates it. There was only one place in the code where
1110 this was missing, in the iplookup router, which I don't think anybody now
1111 uses, but I've fixed it anyway.
1113 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1114 well as to direct DNS lookups. Otherwise the handling of names in host
1115 lists is inconsistent and therefore confusing.
1121 PH/01 Two changes to the default runtime configuration:
1123 (1) Move the checks for relay_from_hosts and authenticated clients from
1124 after to before the (commented out) DNS black list checks.
1126 (2) Add control=submission to the relay_from_hosts and authenticated
1127 clients checks, on the grounds that messages accepted by these
1128 statements are most likely to be submissions.
1130 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1132 (1) Generate an error if the third argument for the ${prvs expansion is
1135 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1138 (3) Reset the variables that are obtained from the first argument of
1139 ${prvscheck and used in the second argument before leaving the code,
1140 because their memory is reclaimed, so using them afterwards may do
1143 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1144 one (it's much easier than Tom thought :-).
1146 (5) Because of (4), we can now allow for the use of $prvscheck_result
1147 inside the third argument.
1149 PH/03 For some reason, the default setting of PATH when running a command from
1150 a pipe transport was just "/usr/bin". I have changed it to
1153 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1154 anything to be listed in the output from -bV.
1156 PH/05 When a filter generated an autoreply, the entire To: header line was
1157 quoted in the delivery log line, like this:
1159 => >A.N.Other <ano@some.domain> <original@ddress> ...
1161 This has been changed so that it extracts the operative address. There
1162 may be more than one such address. If so, they are comma-separated, like
1165 => >ano@some.domain,ona@other.domain <original@ddress> ...
1167 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1168 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1169 not being looked up in the rDNS to get a host name, Exim was showing the
1170 IP address twice in Received: lines, even though the IP addresses were
1171 identical. For example:
1173 Received: from [1.2.3.4] (helo=[1.2.3.4])
1175 However, if the real host name was known, it was omitting the HELO data
1176 if it matched the actual IP address. This has been tidied up so that it
1177 doesn't show the same IP address twice.
1179 PH/07 When both +timestamp and +memory debugging was on, the value given by
1180 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1181 called by the debug printing, thereby overwriting the timestamp buffer.
1182 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1184 PH/08 When the original message was included in an autoreply transport, it
1185 always said "this is a copy of the message, including all the headers",
1186 even if body_only or headers_only was set. It now gives an appropriate
1189 PH/09 Applied a patch from the Sieve maintainer which:
1191 o fixes some comments
1192 o adds the (disabled) notify extension core
1193 o adds some debug output for the result of if/elsif tests
1194 o points to the current vacation draft in the documentation
1195 and documents the missing references header update
1199 o fixes a bug in processing the envelope test (when testing
1200 multiple envelope elements, the last element determinted the
1203 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1204 Electronic Mail") by including:
1206 Auto-submitted: auto-generated
1208 in the messages that it generates (bounce messages and others, such as
1209 warnings). In the case of bounce messages for non-SMTP mesages, there was
1210 also a typo: it was using "Auto_submitted" (underscore instead of
1211 hyphen). Since every message generated by Exim is necessarily in response
1212 to another message, thes have all been changed to:
1214 Auto-Submitted: auto-replied
1216 in accordance with these statements in the RFC:
1218 The auto-replied keyword:
1220 - SHOULD be used on messages sent in direct response to another
1221 message by an automatic process,
1223 - MUST NOT be used on manually-generated messages,
1225 - MAY be used on Delivery Status Notifications (DSNs) and Message
1226 Disposition Notifications (MDNs),
1228 - MUST NOT be used on messages generated by automatic or periodic
1229 processes, except for messages which are automatic responses to
1232 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1233 to the default Received: header definition.
1235 PH/12 Added log selector acl_warn_skipped (default on).
1237 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1238 variables because (a) they are in the wrong storage pool and (b) even if
1239 they were copied, it wouldn't work properly because of the caching.
1241 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1242 checking when decoding. Apparently there are clients that generate
1243 overlong encoded strings. Why am I not surprised?
1245 PH/15 If the first argument of "${if match_address" was not empty, but did not
1246 contain an "@" character, Exim crashed. Now it writes a panic log message
1247 and treats the condition as false.
1249 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1251 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1252 extension "envelope-auth". The code is finished and in agreement with
1253 other implementations, but there is no documentation so far and in fact,
1254 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1255 not changing the active code.
1257 Print executed "if" and "elsif" statements when debugging is used. This
1258 helps a great deal to understand what a filter does.
1260 Document more things not specified clearly in RFC3028. I had all this
1261 sorted out, when out of a sudden new issues came to my mind. Oops."
1263 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1266 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1267 canonical form (as documented). However, after a host name lookup from
1268 the IP address, check_host() was doing a simple string comparison with
1269 addresses acquired from the DNS when checking that the found name did
1270 have the original IP as one of its addresses. Since any found IPv6
1271 addresses are likely to be in abbreviated form, the comparison could
1272 fail. Luckily, there already exists a function for doing the comparison
1273 by converting both addresses to binary, so now that is used instead of
1274 the text comparison.
1276 PH/20 There was another similar case to PH/19, when a complete host name was
1277 given in a host list; looking up its IP address could give an abbreviated
1278 form, whereas the current host's name might or might not be abbreviated.
1279 The same fix has been applied.
1285 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1286 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1289 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1290 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1292 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1294 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1295 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1296 Also, for "Auto-Submitted", treat anything other than "no" as
1297 non-personal, in accordance with RFC 3834. (Previously it treated
1298 anything starting "auto-" as non-personal.)
1300 TF/01 The control=submission/name=... option had a problem with syntax
1301 errors if the name included a slash character. The /name= option
1302 now slurps the rest of the string, so it can include any characters
1303 but it must come last in the list of options (after /sender_retain
1306 PH/05 Some modifications to the interface to the fake nameserver for the new
1314 TK/01 Added the "success_on_redirect" address verification option. See
1315 NewStuff for rationale and an example.
1317 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1319 PH/02 Patch to exigrep to allow it to work on syslog lines.
1321 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1322 fread() to skip over the body file's header line, because in Cygwin the
1323 header line is locked and is inaccessible.
1325 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1326 co-exist for some time) to make it clear that it is the Exim ID that is
1327 referenced, not the Message-ID: header line.
1329 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1330 string_format() function, because snprintf() does not exist on all
1333 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1334 ${stat: expansion item.
1336 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1337 protocol synchronization error", to keep the pedants happy.
1339 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1340 well as for IRIX systems, when gcc is being used. See the host.c source
1343 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1345 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1348 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1349 transport and to the smtp transport in LMTP mode.
1351 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1353 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1354 run a filter in a subprocess. This could lead to confusion in subsequent
1355 lookups in the parent process. There should also be a search_tidyup() at
1356 the end of the subprocess.
1358 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1359 only if the host matched helo_try_verify_hosts, which caused the
1360 verification to occur when the EHLO/HELO command was issued. The ACL just
1361 tested the remembered result. Now, if a previous verification attempt has
1362 not happened, "verify = helo" does it there and then.
1364 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1366 TK/03 Fix log output including CR from clamd.
1368 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1369 exist provoked a memory error which could cause a segfault.
1371 PH/15 Installed PCRE 6.2
1373 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1375 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1376 of the problem. Specifically, suggested +O2 rather than +O1 for the
1379 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1381 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1382 record, the port was not correctly logged when the outgoing_port log
1383 selector was set (it logged the transort's default port).
1385 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1386 fallback_hosts, and "hosts" in the smtp transport.
1388 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1389 host errors such as "Connection refused".
1391 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1392 authentication with radiusclient 0.4.9:
1394 - Error returned from rc_read_config was caught wrongly
1395 - Username/password not passed on to radius server due to wrong length.
1397 The presumption is that some radiusclient API changes for 4.51/PH/17
1398 were not taken care of correctly. The code is still untested by me (my
1399 Linux distribution still has 0.3.2 of radiusclient), but it was
1400 contributed by a Radius user.
1402 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1403 expanding the "port" option of the smtp transport.
1405 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1406 while reading a MIME header. Thanks to Tom Hughes for a patch.
1408 PH/24 Include config.h inside local_scan.h so that configuration settings are
1411 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1412 that in an ACL for RCPT (for example), you can examine exactly what was
1415 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1416 commands, but it was not correctly comparing the address with the actual
1417 client host address. Thus, it would show the EHLO address in Received:
1418 header lines when this was not necessary.
1420 PH/27 Added the % operator to ${eval:}.
1422 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1423 it should be ignoring failures (because with -C, for example, it has lost
1424 privilege). It wasn't ignoring creation failures other than "already
1427 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1430 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1431 because an input line was too long, either on its own, or by virtue of
1432 too many continuations, the temporary file was not being removed, and the
1433 return code was incorrect.
1435 PH/31 Missing "BOOL" in function definition in filtertest.c.
1437 PH/32 Applied Sieve patches from the maintainer.
1439 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1441 PH/33 Added "verify = not_blind".
1443 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1444 Local/Makefile (with some defaults set). These are used in built scripts
1445 such as exicyclog, but they have never been used in the exim_install
1446 script (though there are many overriding facilities there). I have
1447 arranged that the exim_install script now takes note of these two
1450 PH/35 Installed configuration files for Dragonfly.
1452 PH/36 When a locally submitted message by a trusted user did not contain a
1453 From: header, and the sender address was obtained from -f or from an SMTP
1454 MAIL command, and the trusted user did not use -F to supply a sender
1455 name, $originator_name was incorrectly used when constructing a From:
1456 header. Furthermore, $originator_name was used for submission mode
1457 messages from external hosts without From: headers in a similar way,
1458 which is clearly wrong.
1460 PH/37 Added control=suppress_local_fixups.
1462 PH/38 When log_selector = +received_sender was set, and the addition of the
1463 sender made the log line's construction buffer exactly full, or one byte
1464 less than full, an overflow happened when the terminating "\n" was
1467 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1468 when the result of a list match is failure because a DNS lookup failed.
1470 PH/40 RM_COMMAND is now used in the building process.
1472 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1473 the "build-* directories that it finds.
1475 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1476 address in a domain literal was a prefix of an interface address.
1478 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1479 when verifying a sender address, unless rewrite_headers is false.
1481 PH/44 Wrote a long comment about why errors_to addresses are verified as
1482 recipients, not senders.
1484 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1485 the ratelimit ACL was added.
1487 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1489 PH/46 Added extra information about PostgreSQL errors to the error string.
1491 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1492 avoiding the need to install special zones in a real server. This is
1493 backwards compatible; if it can't find the fake resolver, it drops back.
1494 Thus, both old and new test suites can be run.
1496 TF/02 Added util/ratelimit.pl
1498 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1499 clock is set back in time.
1501 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1502 Candler <B.Candler@pobox.com>.
1504 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1505 OK for addresses that are the result of redirections.
1507 PH/48 A number of further additions for the benefit of the new test suite,
1508 including a fake gethostbyname() that interfaces to the fake DNS resolver
1511 TF/06 The fix for widen_domains has also been applied to qualify_single and
1512 search_parents which are the other dnslookup options that can cause
1515 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1518 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1520 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1521 service" if it encountered an MX record with an empty target hostname.
1522 The message is now "an MX or SRV record indicated no SMTP service".
1524 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1525 if the DNS of the sending site is misconfigured. This is quite a
1526 common situation. This change restores the behaviour of treating a
1527 helo verification defer as a failure.
1529 PH/51 If self=fail was set on a router, the bounce message did not include the
1530 actual error message.
1536 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1538 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1539 command itself ended in error, the underlying message about the transport
1540 filter timeout was being overwritten with the pipe command error. Now the
1541 underlying error message should be appended to the second error message.
1543 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1545 PH/02 Reduce the amount of output that "make" produces by default. Full output
1546 can still be requested.
1548 PH/03 The warning log line about a condition test deferring for a "warn" verb
1549 was being output only once per connection, rather than after each
1550 occurrence (because it was using the same function as for successful
1551 "warn" verbs). This seems wrong, so I have changed it.
1553 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1554 it should not have, which might have caused a crash in the right
1555 circumstances, but probably never did.
1557 PH/04 Installed a modified version of Tony Finch's patch to make submission
1558 mode fix the return path as well as the Sender: header line, and to
1559 add a /name= option so that you can make the user's friendly name appear
1562 TF/03 Added the control = fakedefer ACL modifier.
1564 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1565 Mark Lowes for thorough testing.
1567 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1569 TK/03 Merged latest SRS patch from Miles Wilton.
1571 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1572 with the definition in sysexits.h (which is #included earlier).
1573 Fortunately, Exim does not actually use EX_OK. The code used to try to
1574 preserve the sysexits.h value, by assumimg that macro definitions were
1575 scanned for macro replacements. I have been disabused of this notion,
1576 so now the code just undefines EX_OK before #including unistd.h.
1578 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1579 in the smtp transport. When a block could not be written in a single
1580 write() function, the timeout was being re-applied to each part-write.
1581 This seems wrong - if the receiver was accepting one byte at a time it
1582 would take for ever. The timeout is now adjusted when this happens. It
1583 doesn't have to be particularly precise.
1585 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1586 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1588 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1589 without a domain if the check to <postmaster@domain> fails.
1591 SC/01 Eximstats: added -xls and the ability to specify output files
1592 (patch written by Frank Heydlauf).
1594 SC/02 Eximstats: use FileHandles for outputing results.
1596 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1598 SC/04 Eximstats: fixed display of large numbers with -nvr option
1600 SC/05 Eximstats: fixed merging of reports with empty tables.
1602 SC/06 Eximstats: added the -include_original_destination flag
1604 SC/07 Eximstats: removed tabs and trailing whitespace.
1606 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1608 TK/06 MBOX spool code: Add real "From " MBOX separator line
1609 so the .eml file is really in mbox format (even though
1610 most programs do not really care). Patch from Alex Miller.
1612 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1613 The latter is generated from $received_to and is only set if the
1614 message has one envelope recipient. SA can use these headers,
1615 obviously out-of-the-box. Patch from Alex Miller.
1617 PH/08 The ${def test on a variable was returning false if the variable's
1618 value was "0", contrary to what the specification has always said!
1619 The result should be true unless the variable is empty.
1621 PH/09 The syntax error of a character other than { following "${if
1622 def:variable_name" (after optional whitespace) was not being diagnosed.
1623 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1624 accidental colon was present, for example, could give incorrect results.
1626 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1627 result is used (not including appendfile, where other changes are about
1630 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1631 This involved changing a lot of size variables from int to off_t. It
1632 should work with maildirs and everything.
1634 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1635 spamd dying while we are connected to it.
1637 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1640 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1641 function for BDB 4.3.
1643 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1645 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1646 the sources for PCRE have been re-arranged and more files are now
1649 PH/15 The code I had for printing potentially long long variables in PH/11
1650 above was not the best (it lost precision). The length of off_t variables
1651 is now inspected at build time, and an appropriate printing format (%ld
1652 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1653 to be "long long int" or "long int". This is needed for the internal
1654 formatting function string_vformat().
1656 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1657 the configuration file to be ":syslog", then the script "guesses" where
1658 the logs files are, rather than using the compiled in default. In our
1659 case the guess is not the same as the compiled default, so the script
1660 suddenly stopped working when I started to use syslog. The patch checks
1661 to see if log_file_path is "". If so, it attempts to read it from exim
1662 with no configuration file to get the compiled in version, before it
1663 falls back to the previous guessing code."
1665 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1666 implementing BATV in an Exim configuration. See NewStuff for the gory
1669 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1670 Makefile that are specific to HP-UX.
1672 PH/18 If the "use_postmaster" option was set for a recipient callout together
1673 with the "random" option, the postmaster address was used as the MAIL
1674 FROM address for the random test, but not for the subsequent recipient
1675 test. It is now used for both.
1677 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1678 patch removes a few documentation additions to RFC 3028, because the
1679 latest draft now contains them. It adds the new en;ascii-case comparator
1680 and a new error check for 8bit text in MIME parts. Comparator and
1681 require names are now matched exactly. I enabled the subaddress
1682 extension, but it is not well tested yet (read: it works for me)."
1684 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1685 rework some of the code of TK/09 above to avoid the hardwired use of
1686 "%lld" and "long long". Replaced the call to snprintf() with a call to
1689 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1690 records point to non-existent hosts", "retry timeout exceeded", and
1691 "retry time not reached for any host after a long failure period".
1693 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1694 experimental DomainKeys support:
1696 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1697 (2) On an error such as an illegally used "control", the wrong name for
1698 the control was given.
1700 These problems did NOT occur unless DomainKeys support was compiled.
1702 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1704 PH/24 Added ${if match_ip condition.
1706 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1707 databases so that it will be absolutely obvious if a crash occurs in the
1708 DB library. This is a regular occurrence (often caused by mis-matched
1711 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1712 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1713 picked up on a user's system that detects such things. There doesn't seem
1714 to be a gcc warning option for this - only an attribute that has to be
1715 put on the function's prototype. It seems that in Fedora Core 4 they have
1716 set this on a number of new functions. No doubt there will be more in due
1719 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1720 specify a transport. However, if an address that was verified by such a
1721 router was the subject of a callout, Exim crashed because it tried to
1722 read the rcpt_include_affixes from the non-existent transport. Now it
1723 just assumes that the setting of that option is false. This bug was
1724 introduced by 4.51/PH/31.
1726 PH/28 Changed -d+all to exclude +memory, because that information is very
1727 rarely of interest, but it makes the output a lot bigger. People tend to
1728 do -d+all out of habit.
1730 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1731 code in os-type was giving problems when libc.so lives in lib64, like on
1734 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1735 aren't the modern standard, and it seems that some systems' include files
1736 don't always have them. Exim was already checking for some of the newer
1737 ones like T_AAAA, and defining it itself. I've added checks for all the
1738 record types that Exim uses.
1740 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1741 not automatically generating a new one, as it is supposed to. This
1742 prevented TLS from working. If the file did exist, but contained invalid
1743 data, a new version was generated, as expected. It was only the case of a
1744 non-existent file that was broken.
1746 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1747 with a change in libdomainkeys > 0.64.
1749 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1750 from DNS. If the selector record carries the flag, it now has
1751 precedence over the domain-wide flag.
1753 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1755 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1756 the use of a port name that isn't defined in /etc/services) occurred, the
1757 message was deferred as in a normal delivery, and thus remained on the
1758 spool, instead of being failed because of the mua_wrapper setting. This
1759 is now fixed, and I tidied up some of the mua_wrapper messages at the
1762 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1763 the messages in a hash of arrays rather than using individual hashes.
1764 This is a bit cleaner and results in dramatic memory savings, albeit
1765 at a slight CPU cost.
1767 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1768 as requested by Marc Sherman.
1770 SC/10 Eximstats: added histograms for user specified patterns as requested
1773 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1775 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1776 fopen() in the content-scanning modules that did not already have it.
1782 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1783 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1785 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1787 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1789 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1790 to test Sieve filters that use "vacation".
1792 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1793 that changes the way the GnuTLS parameters are stored in the cache file.
1794 The new format can be generated externally. For backward compatibility,
1795 if the data in the cache doesn't make sense, Exim assumes it has read an
1796 old-format file, and it generates new data and writes a new file. This
1797 means that you can't go back to an older release without removing the
1800 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1801 work if there are any delivery delays because "one_time" forces the
1802 parent to be marked "delivered", so its unseen clone is never tried
1803 again. For this reason, Exim now forbids the simultaneous setting of
1806 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1807 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1808 entry if you want to know the details. The fix, however, neglected to
1809 consider the case when local delivery batching is involved. The test for
1810 "previously delivered" was not happening when checking to see if an
1811 address could be batched with a previous (undelivered) one; under
1812 certain circumstances this could lead to multiple deliveries to the same
1815 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1816 in its include files, and this causes problems building Exim.
1818 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1819 header_syntax) but Exim was just ignoring anything given after a slash.
1820 In particular, this caused confusion with an attempt to use "verify =
1821 reverse_host_lookup/defer_ok". An error is now given when options are
1822 supplied for verify items that do not have them. (Maybe reverse_host_
1823 lookup should have a defer_ok option, but that's a different point.)
1825 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1826 defined by RFC 821) to 2048, because there were problems with some AUTH
1827 commands, and RFC 1869 says the size should be increased for extended
1828 SMTP commands that take arguments.
1830 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1833 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1834 "unknown" error; now it says that the functionality isn't in the binary.
1836 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1837 an address' error message when a string expansion fails (syntax or
1838 whatever). Otherwise the password may appear in the log. Following change
1839 PH/42 below, there is no longer a chance of it appearing in a bounce
1842 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1844 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1845 of its addresses was ever tried. (Bugzilla bug #2.)
1847 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1848 the result incorrectly in the debug output. (It correctly added a newline
1849 to what was transported.)
1851 TF/01 Added $received_time.
1853 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1854 commented out examples of how to interface to a virus scanner and to
1855 SpamAssassin. Also added commented examples of av_scanner and
1856 spamd_address settings.
1858 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1859 and controls are allowed in which ACLs. There were a couple of minor
1860 errors. Some of the entries in the conditions table (which is a table of
1861 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1862 negation of where the condition IS allowed.
1864 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1866 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1867 header file does not have a version number, so I've had to invent a new
1868 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1869 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1870 radiusclient), but it was contributed by a Radius user.
1872 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1873 files or directories, for OpenSSL.
1875 PH/19 When an Exim process that is running as root has to create an Exim log
1876 file, it does so in a subprocess that runs as exim:exim so as to get the
1877 ownership right at creation (otherwise, other Exim processes might see
1878 the file with the wrong ownership). There was no test for failure of this
1879 fork() call, which would lead to the process getting stuck as it waited
1880 for a non-existent subprocess. Forks do occasionally fail when resources
1881 run out. I reviewed all the other calls to fork(); they all seem to check
1884 PH/20 When checking for unexpected SMTP input at connect time (before writing
1885 the banner), Exim was not dealing correctly with a non-positive return
1886 from the read() function. If the client had disconnected by this time,
1887 the result was a log entry for a synchronization error with an empty
1888 string after "input=" when read() returned zero. If read() returned -1
1889 (an event I could not check), uninitialized data bytes were printed.
1890 There were reports of junk text (parts of files, etc) appearing after
1893 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1895 PH/22 Added support for macro redefinition, and (re)definition in between
1896 driver and ACL definitions.
1898 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1899 forgetting to use the resulting value; it was using the unexpanded value.
1901 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1902 hadn't been configured. The fix is from Juergen Kreileder, who
1903 understands it better than I do:
1905 "Here's what I see happening with three configured cyrus_sasl
1906 authenticators configured (plain, login, cram-md5):
1908 On startup auth_cyrus_sasl_init() gets called for each of these.
1909 This means three calls to sasl_listmech() without a specified mech_list.
1910 => SASL tests which mechs of all available mechs actually work
1911 => three warnings about OTP not working
1912 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1914 With the patch, sasl_listmech() also gets called three times. But now
1915 SASL's mech_list option is set to the server_mech specified in the the
1916 authenticator. Or in other words, the answer from sasl_listmech()
1917 gets limited to just the mech you're testing for (which is different
1919 => the return list contains just 'plain' or 'login', 'cram-md5' or
1920 nothing depending on the value of ob->server_mech.
1922 I've just tested the patch: Authentication still works fine,
1923 unavailable mechs specified in the exim configuration are still
1924 caught, and the auth.log warnings about OTP are gone."
1926 PH/25 When debugging is enabled, the contents of the command line are added
1927 to the debugging output, even when log_selector=+arguments is not
1930 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1931 answer is "GNU", and only if the return is "GNU/something" is the answer
1934 PH/27 $acl_verify_message is now set immediately after the failure of a
1935 verification in an ACL, and so is available in subsequent modifiers. In
1936 particular, the message can be preserved by coding like this:
1938 warn !verify = sender
1939 set acl_m0 = $acl_verify_message
1941 Previously, $acl_verify_message was set only while expanding "message"
1942 and "log_message" when a very denied access.
1944 PH/28 Modified OS/os.c-Linux with
1946 -#ifndef OS_LOAD_AVERAGE
1947 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1949 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1950 nomenclature these days.)
1952 PH/29 Installed patch from the Sieve maintainer that adds the options
1953 sieve_useraddress and sieve_subaddress to the redirect router.
1955 PH/30 In these circumstances:
1956 . Two addresses routed to the same list of hosts;
1957 . First host does not offer TLS;
1958 . First host accepts first address;
1959 . First host gives temporary error to second address;
1960 . Second host offers TLS and a TLS session is established;
1961 . Second host accepts second address.
1962 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1963 and peerdn, if requested) that were in fact used only for the second
1966 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1967 attention to any local part prefix or suffix that was matched by the
1968 router that accepted the address. It now behaves in the same way as it
1969 does for delivery: the affixes are removed from the local part unless
1970 rcpt_include_affixes is set on the transport.
1972 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1973 timeout during the DATA phase of an incoming message.
1975 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1976 applied a patch sanctioned by the Sieve maintainer.
1978 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1979 the uid or gid is negative. A case of a negative gid caused this to be
1980 noticed. The fix allows for either to be negative.
1982 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1983 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1986 PH/36 The change PH/12 above was broken. Fixed it.
1988 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1989 the grounds that routing the same address twice would always produce the
1990 same answer. This might have been true once, but it is certainly no
1991 longer true now. Routing a child address may depend on the previous
1992 routing that produced that child. Some complicated redirection strategies
1993 went wrong when messages had multiple recipients, and made Exim's
1994 behaviour dependent on the order in which the addresses were given.
1996 I have moved the duplicate checking until after the routing is complete.
1997 Exim scans the addresses that are assigned to local and remote
1998 transports, and removes any duplicates. This means that more work will be
1999 done, as duplicates will always all be routed, but duplicates are
2000 presumably rare, so I don't expect this is of any significance.
2002 For deliveries to pipes, files, and autoreplies, the duplicate checking
2003 still happens during the routing process, since they are not going to be
2006 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2007 It corrects a timeout issue with spamd. This is Ian's comment: "The
2008 background is that sometimes spamd either never reads data from a
2009 connection it has accepted, or it never writes response data. The exiscan
2010 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2011 blindly assumes that writes won't block so it may never time out."
2013 PH/39 Allow G after quota size as well as K and M.
2015 PH/40 The value set for $authenticated_id in an authenticator may not contain
2016 binary zeroes or newlines because the value is written to log lines and
2017 to spool files. There was no check on this. Now the value is run through
2018 the string_printing() function so that such characters are converted to
2019 printable escape sequences.
2021 PH/41 $message_linecount is a new variable that contains the total number of
2022 lines in the message. Compare $body_linecount, which is the count for the
2025 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2026 bounce and delay warning messages, except in certain special cases, which
2029 (a) An SMTP error message from a remote host;
2030 (b) A message specified in a :fail: redirection;
2031 (c) A message specified in a "fail" command in a system filter;
2032 (d) A message specified in a FAIL return from the queryprogram router;
2033 (e) A message specified by the cannot_route_message router option.
2035 In these cases only, Exim does include the error details in bounce and
2036 warning messages. There are also a few cases where bland messages such
2037 as "unrouteable address" or "local delivery error" are given.
2039 PH/43 $value is now also set for the "else" part of a ${run expansion.
2041 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2042 being worked on, but at least Exim now implements the latest version to
2045 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2046 process to complete was treated as a delivery failure, a timeout while
2047 writing the message to the pipe was logged, but erroneously treated as a
2048 successful delivery. Such timeouts include transport filter timeouts. For
2049 consistency with the overall process timeout, these timeouts are now
2050 treated as errors, giving rise to delivery failures by default. However,
2051 there is now a new Boolean option for the pipe transport called
2052 timeout_defer, which, if set TRUE, converts the failures into defers for
2053 both kinds of timeout. A transport filter timeout is now identified in
2056 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2057 systems where "make" and "gmake" are different, calling "gmake" at top
2058 level broke things. I've arranged for the value of $(MAKE) to be passed
2059 from the Makefile to this script so that it can call the same version of
2063 A note about Exim versions 4.44 and 4.50
2064 ----------------------------------------
2066 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2067 changes of various kinds. As a consequence, a big documentation update was
2068 needed. This delayed the release for rather longer than seemed good, especially
2069 in the light of a couple of (minor) security issues. Therefore, the changes
2070 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2071 release. So 4.44 and 4.50 are in effect two different branches that both start
2074 I have left the 4.50 change log unchanged; it contains all the changes since
2075 4.43. The change log for 4.44 is below; many of its items are identical to
2076 those for 4.50. This seems to be the most sensible way to preserve the
2077 historical information.
2083 1. Minor wording change to the doc/README.SIEVE file.
2085 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2086 computation of the current number of files was incorrect.
2088 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2089 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2092 4. Give more explanation in the error message when the command for a transport
2093 filter fails to execute.
2095 5. There are several places where Exim runs a non-Exim command in a
2096 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2097 was being done only for the command run by the queryprogram router. It is
2098 now done for all such subprocesses. The other cases are: ${run, transport
2099 filters, and the commands run by the lmtp and pipe transports.
2101 6. Added CONFIGURE_GROUP build-time option.
2103 7. Some older OS have a limit of 256 on the maximum number of file
2104 descriptors. Exim was using setrlimit() to set 1000 as a large value
2105 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2106 systems. I've change it so that if it can't get 1000, it tries for 256.
2108 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2109 was an oversight, and furthermore, ever since the addition of extra
2110 controls (e.g. 4.43/32), the checks on when to allow different forms of
2111 "control" were broken. There should now be diagnostics for all cases when a
2112 control that does not make sense is encountered.
2114 9. Added the /retain_sender option to "control=submission".
2116 10. $recipients is now available in the predata ACL (oversight).
2118 11. Tidy the search cache before the fork to do a delivery from a message
2119 received from the command line. Otherwise the child will trigger a lookup
2120 failure and thereby defer the delivery if it tries to use (for example) a
2121 cached ldap connection that the parent has called unbind on.
2123 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2124 of $address_data from the recipient verification was clobbered by the
2125 sender verification.
2127 13. The value of address_data from a sender verification is now available in
2128 $sender_address_data in subsequent conditions in the ACL statement.
2130 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2132 15. Added a new option "connect=<time>" to callout options, to set a different
2135 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2136 was its contents. (It was OK if the option was not defined at all.)
2138 17. A "Completed" log line is now written for messages that are removed from
2139 the spool by the -Mrm option.
2141 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2142 information about exactly what failed.
2144 19. Added -dd to debug only the daemon process.
2146 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2147 handles timeouts, both on the server side and network timeouts. Renamed the
2148 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2150 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2151 It is now set to "smtps".
2153 22. $host_address is now set to the target address during the checking of
2154 ignore_target_hosts.
2156 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2157 being passed; this would have caused $sender_host_name to have been used if
2158 matching the list had actually called for a host name (not very likely,
2159 since this list is usually IP addresses). A host name is now passed as
2162 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2163 code that specifies a non-restarting handler (typically sigaction() in
2164 modern systems) in an attempt to fix a rare and obscure crash bug.
2166 25. Narrowed the window for a race in the daemon that could cause it to ignore
2167 SIGCHLD signals. This is not a major problem, because they are used only to
2168 wake it up if nothing else does.
2170 26. A malformed maildirsize file could cause Exim to calculate negative values
2171 for the mailbox size or file count. Odd effects could occur as a result.
2172 The maildirsize information is now recalculated if the size or filecount
2175 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2176 support for a long time. Removed HAVE_SYS_VFS_H.
2178 28. Installed the latest version of exipick from John Jetmore.
2180 29. In an address list, if the pattern was not a regular expression, an empty
2181 subject address (from a bounce message) matched only if the pattern was an
2182 empty string. Non-empty patterns were not even tested. This was the wrong
2183 because it is perfectly reasonable to use an empty address as part of a
2184 database query. An empty address is now tested by patterns that are
2185 lookups. However, all the other forms of pattern expect the subject to
2186 contain a local part and a domain, and therefore, for them, an empty
2187 address still always fails if the pattern is not itself empty.
2189 30. Exim went into a mad DNS loop when attempting to do a callout where the
2190 host was specified on an smtp transport, and looking it up yielded more
2191 than one IP address.
2193 31. Re-factored the code for checking spool and log partition space into a
2194 function that finds that data and another that does the check. The former
2195 is then used to implement four new variables: $spool_space, $log_space,
2196 $spool_inodes, and $log_inodes.
2198 32. The RFC2047 encoding function was originally intended for short strings
2199 such as real names; it was not keeping to the 75-character limit for
2200 encoded words that the RFC imposes. It now respects the limit, and
2201 generates multiple encoded words if necessary. To be on the safe side, I
2202 have increased the buffer size for the ${rfc2047: expansion operator from
2205 33. It is now permitted to omit both strings after an "if" condition; if the
2206 condition is true, the result is "true". As before, when the second string
2207 is omitted, a false condition yields an empty string. This makes it less
2208 cumbersome to write custom ACL and router conditions.
2210 34. Failure to deliver a bounce message always caused it to be frozen, even if
2211 there was an errors_to setting on the router. The errors_to setting is now
2214 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2215 canonical form (fully expanded) before being placed in
2216 $sender_host_address.
2218 36. The table in the code that translates DNS record types into text (T_A to
2219 "A" for instance) was missing entries for NS and CNAME. It is just possible
2220 that this could have caused confusion if both these types were looked up
2221 for the same domain, because the text type is used as part of Exim's
2222 per-process caching. But the chance of anyone hitting this buglet seems
2225 37. The dnsdb lookup has been extended in a number of ways.
2227 (1) There is a new type, "zns", which walks up the domain tree until it
2228 finds some nameserver records. It should be used with care.
2230 (2) There is a new type, "mxh", which is like "mx" except that it returns
2231 just the host names, not the priorities.
2233 (3) It is now possible to give a list of domains (or IP addresses) to be
2234 looked up. The behaviour when one of the lookups defers can be
2235 controlled by a keyword.
2237 (4) It is now possible to specify the separator character for use when
2238 multiple records are returned.
2240 38. The dnslists ACL condition has been extended: it is now possible to supply
2241 a list of IP addresses and/or domains to be looked up in a particular DNS
2244 39. Added log_selector=+queue_time_overall.
2246 40. When running the queue in the test harness, wait just a tad after forking a
2247 delivery process, to get repeatability of debugging output.
2249 41. Include certificate and key file names in error message when GnuTLS fails
2250 to set them up, because the GnuTLS error message doesn't include the name
2251 of the failing file when there is a problem reading it.
2253 42. Allow both -bf and -bF in the same test run.
2255 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2257 44. The "Exiscan patch" is now merged into the mainline Exim source.
2259 45. Sometimes the final signoff response after QUIT could fail to get
2260 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2261 < 0 before doing a fflush(). This bug looks as though it goes back to the
2262 introduction of TLS in release 3.20, but "sometimes" must have been rare
2263 because the tests only now provoked it.
2265 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2266 (this can affect the format of dates).
2268 47. exim_tidydb, when checking for the continued existence of a message for
2269 which it has found a message-specific retry record, was not finding
2270 messages that were in split spool directories. Consequently, it was
2271 deleting retry records that should have stayed in existence.
2273 48. Steve fixed some bugs in eximstats.
2275 49. The SPA authentication driver was not abandoning authentication and moving
2276 on to the next authenticator when an expansion was forced to fail,
2277 contradicting the general specification for all authenticators. Instead it
2278 was generating a temporary error. It now behaves as specified.
2280 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2281 (the order specifies the preference for clients). The order is now AES256,
2282 AES128, 3DES, ARCFOUR128.
2284 51. Small patch to Sieve code - explicitly set From: when generating an
2287 52. Exim crashed if a remote delivery caused a very long error message to be
2288 recorded - for instance if somebody sent an entire SpamAssassin report back
2289 as a large number of 550 error lines. This bug was coincidentally fixed by
2290 increasing the size of one of Exim's internal buffers (big_buffer) that
2291 happened as part of the Exiscan merge. However, to be on the safe side, I
2292 have made the code more robust (and fixed the comments that describe what
2295 53. Now that there can be additional text after "Completed" in log lines (if
2296 the queue_time_overall log selector is set), a one-byte patch to exigrep
2297 was needed to allow it to recognize "Completed" as not the last thing in
2300 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2301 patch that reportedly fixes this has been added. I am not expert enough to
2302 create a test for it. This is what the patch creator wrote:
2304 "I found a little strange behaviour of ldap code when working with
2305 Windows 2003 AD Domain, where users was placed in more than one
2306 Organization Units. When I tried to give exim partial DN, the exit code
2307 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2308 But simultaneously result of request was absolutely normal ldap result,
2309 so I produce this patch..."
2311 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2312 REFERENCE, so I have modified the code to exclude the patch when that macro
2315 55. Some experimental protocols are using DNS PTR records for new purposes. The
2316 keys for these records are domain names, not reversed IP addresses. The
2317 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2318 leaves it alone. Component reversal etc. now happens only for IP addresses.
2321 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2323 57. Double the size of the debug message buffer (to 2048) so that more of very
2324 long debug lines gets shown.
2326 58. The exicyclog utility now does better if the number of log files to keep
2327 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2329 59. Two changes related to the smtp_active_hostname option:
2331 (1) $smtp_active_hostname is now available as a variable.
2332 (2) The default for smtp_banner uses $smtp_active_hostname instead
2333 of $primary_hostname.
2335 60. The host_aton() function is supposed to be passed a string that is known
2336 to be a valid IP address. However, in the case of IPv6 addresses, it was
2337 not checking this. This is a hostage to fortune. Exim now panics and dies
2338 if the condition is not met. A case was found where this could be provoked
2339 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2340 components; fortuitously, this particular loophole had already been fixed
2341 by change 4.50/55 above.
2343 If there are any other similar loopholes, the new check in host_aton()
2344 itself should stop them being exploited. The report I received stated that
2345 data on the command line could provoke the exploit when Exim was running as
2346 exim, but did not say which command line option was involved. All I could
2347 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2348 running as the user.
2351 61. There was a buffer overflow vulnerability in the SPA authentication code
2352 (which came originally from the Samba project). I have added a test to the
2353 spa_base64_to_bits() function which I hope fixes it.
2356 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2357 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2359 63. The daemon start-up calls getloadavg() while still root for those OS that
2360 need the first call to be done as root, but it missed one case: when
2361 deliver_queue_load_max is set with deliver_drop_privilege. This is
2362 necessary for the benefit of the queue runner, because there is no re-exec
2363 when deliver_drop_privilege is set.
2365 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2366 This has been fixed.
2368 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2369 was in use, was not putting the data itself into the right store pool;
2370 consequently, it could be overwritten for a subsequent message in the same
2371 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2374 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2376 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2377 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2378 the function were treating the return as a boolean value, which happened to
2379 work because 0=false and not-0=true, but is not correct code.
2381 68. The host_aton() function was not handling scoped IPv6 addresses (those
2382 with, for example, "%eth0" on the end) correctly.
2384 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2385 negated items (that is, ~something) in unsigned ints. Some compilers
2386 apparently mutter when there is no cast.
2388 70. If an address verification called from an ACL failed, and did not produce a
2389 user-specific message (i.e. there was only a "system" message), nothing was
2390 put in $acl_verify_message. In this situation, it now puts the system
2393 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2394 session; change 4.31/43 added the unwanted input to the log line - except
2395 that it did not do this in the start of session case. It now does.
2397 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2398 This is wrong and can cause the other end to generate a synchronization
2399 error if it is another Exim or anything else that does the synchronization
2400 check. A QUIT command is no longer sent after a timeout.
2402 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2403 during host lookups.
2405 74. The defer_ok option of callout verification was not working if it was used
2406 when verifying addresses in header lines, that is, for this case:
2408 verify = header_sender/callout=defer_ok
2410 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2411 those file descriptors could be used for SMTP connections. If anything
2412 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2413 could be sent to the SMTP client, causing chaos. The daemon now opens
2414 stdin, stdout, and stderr to /dev/null when it puts itself into the
2417 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2418 log by default. The user can override this with suitable Perl magic.
2420 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2421 the log message when discard triggers, was not working for the DATA ACL or
2422 for the non-SMTP ACL.
2424 78. Error message wording change in sieve.c.
2426 79. If smtp_accept_max_per_host was set, the number of connections could be
2427 restricted to fewer than expected, because the daemon was trying to set up
2428 a new connection before checking whether the processes handling previous
2429 connections had finished. The check for completed processes is now done
2430 earlier. On busy systems, this bug wouldn't be noticed because something
2431 else would have woken the daemon, and it would have reaped the completed
2434 80. If a message was submitted locally by a user whose login name contained one
2435 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2436 It caused a spool format error. I have fixed the spool reading code. A
2437 related problem was that the "from" clause in the Received: line became
2438 illegal because of the space(s). It is now covered by ${quote_local_part.
2440 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2443 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2445 83. Patch from Sieve maintainer to fix unterminated string problem in
2446 "vacation" handling.
2448 84. Some minor changes to the Linux configuration files to help with other
2449 OS variants using glibc.
2451 85. One more patch for Sieve to update vacation handling to latest spec.
2454 ----------------------------------------------------
2455 See the note above about the 4.44 and 4.50 releases.
2456 ----------------------------------------------------
2462 1. Change 4.43/35 introduced a bug that caused file counts to be
2463 incorrectly computed when quota_filecount was set in an appendfile
2466 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2467 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2470 3. Give more explanation in the error message when the command for a transport
2471 filter fails to execute.
2473 4. There are several places where Exim runs a non-Exim command in a
2474 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2475 was being done only for the command run by the queryprogram router. It is
2476 now done for all such subprocesses. The other cases are: ${run, transport
2477 filters, and the commands run by the lmtp and pipe transports.
2479 5. Some older OS have a limit of 256 on the maximum number of file
2480 descriptors. Exim was using setrlimit() to set 1000 as a large value
2481 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2482 systems. I've change it so that if it can't get 1000, it tries for 256.
2484 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2485 was an oversight, and furthermore, ever since the addition of extra
2486 controls (e.g. 4.43/32), the checks on when to allow different forms of
2487 "control" were broken. There should now be diagnostics for all cases when a
2488 control that does not make sense is encountered.
2490 7. $recipients is now available in the predata ACL (oversight).
2492 8. Tidy the search cache before the fork to do a delivery from a message
2493 received from the command line. Otherwise the child will trigger a lookup
2494 failure and thereby defer the delivery if it tries to use (for example) a
2495 cached ldap connection that the parent has called unbind on.
2497 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2498 of $address_data from the recipient verification was clobbered by the
2499 sender verification.
2501 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2502 was its contents. (It was OK if the option was not defined at all.)
2504 11. A "Completed" log line is now written for messages that are removed from
2505 the spool by the -Mrm option.
2507 12. $host_address is now set to the target address during the checking of
2508 ignore_target_hosts.
2510 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2511 being passed; this would have caused $sender_host_name to have been used if
2512 matching the list had actually called for a host name (not very likely,
2513 since this list is usually IP addresses). A host name is now passed as
2516 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2517 code that specifies a non-restarting handler (typically sigaction() in
2518 modern systems) in an attempt to fix a rare and obscure crash bug.
2520 15. Narrowed the window for a race in the daemon that could cause it to ignore
2521 SIGCHLD signals. This is not a major problem, because they are used only to
2522 wake it up if nothing else does.
2524 16. A malformed maildirsize file could cause Exim to calculate negative values
2525 for the mailbox size or file count. Odd effects could occur as a result.
2526 The maildirsize information is now recalculated if the size or filecount
2529 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2530 support for a long time. Removed HAVE_SYS_VFS_H.
2532 18. Updated exipick to current release from John Jetmore.
2534 19. Allow an empty sender to be matched against a lookup in an address list.
2535 Previously the only cases considered were a regular expression, or an
2538 20. Exim went into a mad DNS lookup loop when doing a callout where the
2539 host was specified on the transport, if the DNS lookup yielded more than
2542 21. The RFC2047 encoding function was originally intended for short strings
2543 such as real names; it was not keeping to the 75-character limit for
2544 encoded words that the RFC imposes. It now respects the limit, and
2545 generates multiple encoded words if necessary. To be on the safe side, I
2546 have increased the buffer size for the ${rfc2047: expansion operator from
2549 22. Failure to deliver a bounce message always caused it to be frozen, even if
2550 there was an errors_to setting on the router. The errors_to setting is now
2553 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2554 canonical form (fully expanded) before being placed in
2555 $sender_host_address.
2557 24. Updated eximstats to version 1.33
2559 25. Include certificate and key file names in error message when GnuTLS fails
2560 to set them up, because the GnuTLS error message doesn't include the name
2561 of the failing file when there is a problem reading it.
2563 26. Expand error message when OpenSSL has problems setting up cert/key files.
2566 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2567 (this can affect the format of dates).
2569 28. exim_tidydb, when checking for the continued existence of a message for
2570 which it has found a message-specific retry record, was not finding
2571 messages that were in split spool directories. Consequently, it was
2572 deleting retry records that should have stayed in existence.
2574 29. eximstats updated to version 1.35
2575 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2576 1.35 - bugfix such that pie charts by volume are generated correctly
2578 30. The SPA authentication driver was not abandoning authentication and moving
2579 on to the next authenticator when an expansion was forced to fail,
2580 contradicting the general specification for all authenticators. Instead it
2581 was generating a temporary error. It now behaves as specified.
2583 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2584 (the order specifies the preference for clients). The order is now AES256,
2585 AES128, 3DES, ARCFOUR128.
2587 31. Small patch to Sieve code - explicitly set From: when generating an
2590 32. Exim crashed if a remote delivery caused a very long error message to be
2591 recorded - for instance if somebody sent an entire SpamAssassin report back
2592 as a large number of 550 error lines. This bug was coincidentally fixed by
2593 increasing the size of one of Exim's internal buffers (big_buffer) that
2594 happened as part of the Exiscan merge. However, to be on the safe side, I
2595 have made the code more robust (and fixed the comments that describe what
2598 33. Some experimental protocols are using DNS PTR records for new purposes. The
2599 keys for these records are domain names, not reversed IP addresses. The
2600 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2601 leaves it alone. Component reversal etc. now happens only for IP addresses.
2604 34. The host_aton() function is supposed to be passed a string that is known
2605 to be a valid IP address. However, in the case of IPv6 addresses, it was
2606 not checking this. This is a hostage to fortune. Exim now panics and dies
2607 if the condition is not met. A case was found where this could be provoked
2608 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2609 components; fortuitously, this particular loophole had already been fixed
2610 by change 4.50/55 or 4.44/33 above.
2612 If there are any other similar loopholes, the new check in host_aton()
2613 itself should stop them being exploited. The report I received stated that
2614 data on the command line could provoke the exploit when Exim was running as
2615 exim, but did not say which command line option was involved. All I could
2616 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2617 running as the user.
2620 35. There was a buffer overflow vulnerability in the SPA authentication code
2621 (which came originally from the Samba project). I have added a test to the
2622 spa_base64_to_bits() function which I hope fixes it.
2625 36. The daemon start-up calls getloadavg() while still root for those OS that
2626 need the first call to be done as root, but it missed one case: when
2627 deliver_queue_load_max is set with deliver_drop_privilege. This is
2628 necessary for the benefit of the queue runner, because there is no re-exec
2629 when deliver_drop_privilege is set.
2631 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2632 was in use, was not putting the data itself into the right store pool;
2633 consequently, it could be overwritten for a subsequent message in the same
2634 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2637 38. Sometimes the final signoff response after QUIT could fail to get
2638 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2639 < 0 before doing a fflush(). This bug looks as though it goes back to the
2640 introduction of TLS in release 3.20, but "sometimes" must have been rare
2641 because the tests only now provoked it.
2647 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2648 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2649 Now it returns an integer. A number of calls were still expecting a T/F
2650 return. Fortuitously, in all cases, the tests worked in OK situations,
2651 which is the norm. However, things would have gone wrong on any write
2652 failures on the smtp file descriptor. This function is used when sending
2653 messages over SMTP and also when doing verify callouts.
2655 2. When Exim is called to do synchronous delivery of a locally submitted
2656 message (the -odf or -odi options), it no longer closes stderr before doing
2659 3. Implemented the mua_wrapper option.
2661 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2663 5. Implemented the functions header_remove(), header_testname(),
2664 header_add_at_position(), and receive_remove_recipient(), and exported them
2667 6. If an ACL "warn" statement specified the addition of headers, Exim already
2668 inserted X-ACL-Warn: at the start if there was no header name. However, it
2669 was not making this test for the second and subsequent header lines if
2670 there were newlines in the string. This meant that an invalid header could
2671 be inserted if Exim was badly configured.
2673 7. Allow an ACL "warn" statement to add header lines at the start or after all
2674 the Received: headers, as well as at the end.
2676 8. Added the rcpt_4xx retry error code.
2678 9. Added postmaster_mailfrom=xxx to callout verification option.
2680 10. Added mailfrom=xxxx to the callout verification option, for verify=
2683 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2684 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2686 12. Inserted some casts to stop certain compilers warning when using pointer
2687 differences as field lengths or precisions in printf-type calls (mostly
2688 affecting debugging statements).
2690 13. Added optional readline() support for -be (dynamically loaded).
2692 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2693 same clock tick as a message's arrival, so that its received time was the
2694 same as the "first fail" time on the retry record, and that message
2695 remained on the queue past the ultimate address timeout, every queue runner
2696 would try a delivery (because it was past the ultimate address timeout) but
2697 after another failure, the ultimate address timeout, which should have then
2698 bounced the address, did not kick in. This was a "< instead of <=" error;
2699 in most cases the first failure would have been in the next clock tick
2700 after the received time, and all would be well.
2702 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2703 being recognized when the domain list was tested by the match_domain
2704 condition in an expansion string.
2706 16. Added the ${str2b64: operator.
2708 17. Exim was always calling setrlimit() to set a large limit for the number of
2709 processes, without checking whether the existing limit was already
2710 adequate. (It did check for the limit on file descriptors.) Furthermore,
2711 errors from getrlimit() and setrlimit() were being ignored. Now they are
2712 logged to the main and panic logs, but Exim does carry on, to try to do its
2713 job under whatever limits there are.
2715 18. Imported PCRE 5.0.
2717 19. Trivial typo in log message " temporarily refused connection" (the leading
2720 20. If the log selector return_path_on_delivery was set and an address was
2721 redirected to /dev/null, the delivery process crashed because it assumed
2722 that a return path would always be set for a "successful" delivery. In this
2723 case, the whole delivery is bypassed as an optimization, and therefore no
2726 21. Internal re-arrangement: the function for sending a challenge and reading
2727 a response while authentication was assuming a zero-terminated challenge
2728 string. It's now changed to take a pointer and a length, to allow for
2729 binary data in such strings.
2731 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2733 23. Exim was not respecting finduser_retries when seeking the login of the
2734 uid under which it was called; it was always trying 10 times. (The default
2735 setting of finduser_retries is zero.) Also, it was sleeping after the final
2736 failure, which is pointless.
2738 24. Implemented tls_on_connect_ports.
2740 25. Implemented acl_smtp_predata.
2742 26. If the domain in control=submission is set empty, Exim assumes that the
2743 authenticated id is a complete email address when it generates From: or
2744 Sender: header lines.
2746 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2747 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2748 chown and chgrp in /bin and hostname in /usr/bin.
2750 28. Exim was keeping the "process log" file open after each use, just as it
2751 does for the main log. This opens the possibility of it remaining open for
2752 long periods when the USR1 signal hits a daemon. Occasional processlog
2753 errors were reported, that could have been caused by this. Anyway, it seems
2754 much more sensible not to leave this file open at all, so that is what now
2757 29. The long-running daemon process does not normally write to the log once it
2758 has entered its main loop, and it closes the log before doing so. This is
2759 so that log files can straightforwardly be renamed and moved. However,
2760 there are a couple of unusual error situations where the daemon does write
2761 log entries, and I had neglected to close the log afterwards.
2763 30. The text of an SMTP error response that was received during a remote
2764 delivery was being truncated at 512 bytes. This is too short for some of
2765 the long messages that one sometimes sees. I've increased the limit to
2768 31. It is now possible to make retry rules that apply only when a message has a
2769 specific sender, in particular, an empty sender.
2771 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2772 it possible to be selective about when SMTP synchronization is enforced.
2774 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2776 32. Implemented hosts_connection_nolog.
2778 33. Added an ACL for QUIT.
2780 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2783 35. Added mailbox_size and mailbox_filecount to appendfile.
2785 36. Added control = no_multiline_responses to ACLs.
2787 37. There was a bug in the logic of the code that waits for the clock to tick
2788 in the case where the clock went backwards by a substantial amount such
2789 that the microsecond fraction of "now" was more than the microsecond
2790 fraction of "then" (but the whole seconds number was less).
2792 38. Added support for the libradius Radius client library this is found on
2793 FreeBSD (previously only the radiusclient library was supported).
2799 1. When certain lookups returned multiple values in the form name=value, the
2800 quoting of the values was not always being done properly. Specifically:
2801 (a) If the value started with a double quote, but contained no whitespace,
2803 (b) If the value contained whitespace other than a space character (i.e.
2804 tabs or newlines or carriage returns) it was not quoted.
2805 This fix has been applied to the mysql and pgsql lookups by writing a
2806 separate quoting function and calling it from the lookup code. The fix
2807 should probably also be applied to nisplus, ibase and oracle lookups, but
2808 since I cannot test any of those, I have not disturbed their existing code.
2810 2. A hit in the callout cache for a specific address caused a log line with no
2811 reason for rejecting RCPT. Now it says "Previous (cached) callout
2812 verification failure".
2814 3. There was an off-by-one bug in the queryprogram router. An over-long
2815 return line was truncated at 256 instead of 255 characters, thereby
2816 overflowing its buffer with the terminating zero. As well as fixing this, I
2817 have increased the buffer size to 1024 (and made a note to document this).
2819 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2820 when Exim is waiting for an SMTP response from a remote server, Exim
2821 restarts its select() call on the socket, thereby resetting its timeout.
2822 This is not a problem when such interrupts are rare. Somebody set up a cron
2823 job to run exiwhat every 2 minutes, which is less than the normal select()
2824 timeout (5 or 10 minutes). This meant that the select() timeout never
2825 kicked in because it was always reset. I have fixed this by comparing the
2826 time when an interrupt arrives with the time at the start of the first call
2827 to select(). If more time than the timeout has elapsed, the interrupt is
2828 treated as a timeout.
2830 5. Some internal re-factoring in preparation for the addition of Sieve
2831 extensions (by MH). In particular, the "personal" test is moved to a
2832 separate function, and given an option for scanning Cc: and Bcc: (which is
2833 not set for Exim filters).
2835 6. When Exim created an email address using the login of the caller as the
2836 local part (e.g. when creating a From: or Sender: header line), it was not
2837 quoting the local part when it contained special characters such as @.
2839 7. Installed new OpenBSD configuration files.
2841 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2842 try to make them clearer.
2844 9. Callout options, other than the timeout value, were being ignored when
2845 verifying sender addresses in header lines. For example, when using
2847 verify = header_sender/callout=no_cache
2849 the cache was (incorrectly) being used.
2851 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2852 only the Cygwin environment.
2854 11. When return_path_on_delivery was set as a log selector, if different remote
2855 addresses in the same message used different return paths and parallel
2856 remote delivery occurred, the wrong values would sometimes be logged.
2857 (Whenever a remote delivery process finished, the return path value from
2858 the most recently started remote delivery process was logged.)
2860 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2861 lines when AUTH and/or TLS are in use. This is the "received protocol"
2862 field. Exim used to use "asmtp" for authenticated SMTP, without any
2863 indication (in the protocol name) for TLS use. Now it follows the RFC and
2864 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2865 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2866 names appear in log lines as well as in Received: header lines.
2868 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2869 extensions, and comparison tests, and to fix some bugs.
2871 14. Changes to the "personal" filter test:
2873 (1) The test was buggy in that it was just doing the equivalent of
2874 "contains" tests on header lines. For example, if a user's address was
2875 anne@some.where, the "personal" test would incorrectly be true for
2877 To: susanne@some.where
2879 This test is now done by extracting each address from the header in turn,
2880 and checking the entire address. Other tests that are part of "personal"
2881 are now done using regular expressions (for example, to check local parts
2882 of addresses in From: header lines).
2884 (2) The list of non-personal local parts in From: addresses has been
2885 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2886 taken from the Sieve specification recommendations.
2888 (3) If the message contains any header line starting with "List-" it is
2889 treated as non-personal.
2891 (4) The test for "circular" in the Subject: header line has been removed
2892 because it now seems ill-conceived.
2894 15. Minor typos in src/EDITME comments corrected.
2896 16. Installed latest exipick from John Jetmore.
2898 17. If headers_add on a router specified a text string that was too long for
2899 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2900 of string_sprintf() is now avoided.
2902 18. $message_body_size was not set (it was always zero) when running the DATA
2903 ACL and the local_scan() function.
2905 19. For the "mail" command in an Exim filter, no default was being set for
2906 the once_repeat time, causing a random time value to be used if "once" was
2907 specified. (If the value happened to be <= 0, no repeat happened.) The
2908 default is now 0s, meaning "never repeat". The "vacation" command was OK
2909 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2910 (I found it when inspecting the code).
2912 20. There is now an overall timeout for performing a callout verification. It
2913 defaults to 4 times the callout timeout, which applies to individual SMTP
2914 commands during the callout. The overall timeout applies when there is more
2915 than one host that can be tried. The timeout is checked before trying the
2916 next host. This prevents very long delays if there are a large number of
2917 hosts and all are timing out (e.g. when the network connections are timing
2918 out). The value of the overall timeout can be changed by specifying an
2919 additional sub-option for "callout", called "maxwait". For example:
2921 verify = sender/callout=5s,maxwait=20s
2923 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2924 to the end before writing, but this should make it even safer).
2926 22. Exim was forgetting that it had advertised PIPELINING for the second and
2927 subsequent messages on an SMTP connection. It was also not resetting its
2928 memory on STARTTLS and an internal HELO.
2930 23. When Exim logs an SMTP synchronization error within a session, it now
2931 records whether PIPELINING has been advertised or not.
2933 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2934 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2935 rather than long int.
2937 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2939 26. Added the never_mail option to autoreply.
2945 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2946 crash if the getsockname() call failed; this can happen if a connection is
2947 closed very soon after it is established. The problem was simply in the
2948 order in which certain operations were done, causing Exim to try to write
2949 to the SMTP stream before it had set up the file descriptor. The bug has
2950 been fixed by making things happen in the correct order.
2956 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2957 before the connection was closed, thus losing the rejection response.
2959 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2960 some early Solaris releases, but causes trouble in current releases where
2961 socklen_t is defined.
2963 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2966 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2969 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2970 that releasing the top of it at the end releases what was used for sub-
2971 expansions (unless the block got too big). However, discard this block if
2972 the first thing is a variable or header, so that we can use its block when
2973 it is dynamic (useful for very large $message_headers, for example).
2975 6. Lookups now cache *every* query, not just the most recent. A new, separate
2976 store pool is used for this. It can be recovered when all lookup caches are
2977 flushed. Lookups now release memory at the end of their result strings.
2978 This has involved some general refactoring of the lookup sources.
2980 7. Some code has been added to the store_xxx() functions to reduce the amount
2981 of flapping under certain conditions.
2983 8. log_incoming_interface used to affect only the <= reception log lines. Now
2984 it causes the local interface and port to be added to several more SMTP log
2985 lines, for example "SMTP connection from", and rejection lines.
2987 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2989 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2991 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2992 could be overwritten at the end of the current message (or the start of a
2993 new message if it was set in a HELO ACL). The value is now preserved for
2994 the duration of the SMTP connection.
2996 12. If a transport had a headers_rewrite setting, and a matching header line
2997 contained an unqualified address, that address was qualified, even if it
2998 did not match any rewriting rules. The underlying bug was that the values
2999 of the flags that permit the existence of unqualified sender and recipient
3000 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3001 non-local messages, and by -bnq for local messages) were not being
3002 preserved with the message after it was received.
3004 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3005 "next input=" as part of the text comprising the host identity instead of
3006 the correct text. The code was using the same buffer for two different
3007 strings. However, depending on which order the printing function evaluated
3008 its arguments, the bug did not always show up. Under Linux, for example, my
3009 test suite worked just fine.
3011 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3012 doesn't work with some older versions of Perl. It has been changed to "my",
3013 which in any case is probably the better facility to use.
3015 15. A really picky compiler found some instances of statements for creating
3016 error messages that either had too many or two few arguments for the format
3019 16. The size of the buffer for calls to the DNS resolver has been increased
3020 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3021 for addresses that have a lot of PTR records. This alleviates a problem; it
3022 does not fully solve it.
3024 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3025 buffer now truncates the list and logs the incident, which is the same
3026 action as happens when Exim is looking up a host name and its aliases.
3027 Previously in this situation something unpredictable would happen;
3028 sometimes it was "internal error: store_reset failed".
3030 18. If a server dropped the connection unexpectedly when an Exim client was
3031 using GnuTLS and trying to read a response, the client delivery process
3032 crashed while trying to generate an error log message.
3034 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3035 string, for example:
3037 warn message = H1: something\nH2: something
3039 the text was added as a single header line from Exim's point of view
3040 though it ended up OK in the delivered message. However, searching for the
3041 second and subsequent header lines using $h_h2: did not work. This has been
3042 fixed. Similarly, if a system filter added multiple headers in this way,
3043 the routers could not see them.
3045 20. Expanded the error message when iplsearch is called with an invalid key to
3046 suggest using net-iplsearch in a host list.
3048 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3049 ACLs are no longer actually imposed (and a message to that effect is
3052 22. If a "gecos" field in a passwd entry contained escaped characters, in
3053 particular, if it contained a \" sequence, Exim got it wrong when building
3054 a From: or a Sender: header from that name. A second bug also caused
3055 incorrect handling when an unquoted " was present following a character
3056 that needed quoting.
3058 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3059 was not being matched caselessly.
3061 24. Arranged for all hyphens in the exim.8 source to be escaped with
3064 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3065 properly. Recipient callout cache records were still being keyed to include
3066 the sender, even when use_sender was set false. This led to far more
3067 callouts that were necessary. The sender is no longer included in the key
3068 when use_sender is false.
3070 26. Added "control = submission" modifier to ACLs.
3072 27. Added the ${base62d: operator to decode base 62 numbers.
3074 28. dnsdb lookups can now access SRV records.
3076 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3077 the configuration file.
3079 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3080 (-v) mode. This makes the output for a verbose queue run more intelligible.
3082 31. Added a use_postmaster feature to recipient callouts.
3084 32. Added the $body_zerocount variable, containing the number of binary zero
3085 bytes in the message body.
3087 33. The time of last modification of the "new" subdirectory is now used as the
3088 "mailbox time last read" when there is a quota error for a maildir
3091 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3093 35. Added +ignore_unknown as a special item in host lists.
3095 36. Code for decoding IPv6 addresses in host lists is now included, even if
3096 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3097 address was recognized as an IP address, but was then not correctly decoded
3098 into binary, causing unexpected and incorrect effects when compared with
3105 1. Very minor rewording of debugging text in manualroute to say "list of
3106 hosts" instead of "hostlist".
3108 2. If verify=header_syntax was set, and a header line with an unqualified
3109 address (no domain) and a large number of spaces between the end of the
3110 name and the colon was received, the reception process suffered a buffer
3111 overflow, and (when I tested it) crashed. This was caused by some obsolete
3112 code that should have been removed. The fix is to remove it!
3114 3. When running in the test harness, delay a bit after writing a bounce
3115 message to get a bit more predictability in the log output.
3117 4. Added a call to search_tidyup() just before forking a reception process. In
3118 theory, someone could use a lookup in the expansion of smtp_accept_max_
3119 per_host which, without the tidyup, could leave open a database connection.
3121 5. Added the variables $recipient_data and $sender_data which get set from a
3122 lookup success in an ACL "recipients" or "senders" condition, or a router
3123 "senders" option, similar to $domain_data and $local_part_data.
3125 6. Moved the writing of debug_print from before to after the "senders" test
3128 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3129 problems for message scanning, either using a data ACL, or using
3130 local_scan() because the Received: header was not generated till after they
3131 were called (in order to set the time as the time of reception completion).
3132 I have revised the way this works. The header is now generated after the
3133 body is received, but before the ACL or local_scan() are called. After they
3134 are run, the timestamp in the header is updated.
3140 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3141 before starting a queue runner without re-exec. This happened only when
3142 deliver_drop_privilege was set or when the Exim user was set to root. The
3143 effect of the bug was that timeouts during subsequent deliveries caused
3144 crashes instead of being properly handled. The handler is now left at its
3145 default (and expected) setting.
3147 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3148 message, again when deliver_drop_privilege is set or Exim is run as root.
3149 The bug described in (1) was not present in this case, but the tidying up
3150 of the other signals was missing. I have made the two cases consistent.
3152 3. The ignore_target_hosts setting on a manualroute router was being ignored
3153 for hosts that were looked up using the /MX notation.
3155 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3158 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3159 operated on the sender address. After changing the $sender_address to <>
3160 for the sender address verify, Exim was re-instated it as the original
3161 (before rewriting) address, but remembering that it had rewritten it, so it
3162 wasn't rewriting it again. This bug also had the effect of breaking the
3163 sender address verification caching when the sender address was rewritten.
3165 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3166 This has been changed so that if the ip literal address matches
3167 ignore_target_hosts, the router declines.
3169 7. Added expansion conditions match_domain, match_address, and match_local_
3170 part (NOT match_host).
3172 8. The placeholder for the Received: header didn't have a length field set.
3174 9. Added code to Exim itself and to exim_lock to test for a specific race
3175 condition that could lead to file corruption when using MBX delivery. The
3176 issue is with the lockfile that is created in /tmp. If this file is removed
3177 after a process has opened it but before that process has acquired a lock,
3178 there is the potential for a second process to recreate the file and also
3179 acquire a lock. This could lead to two Exim processes writing to the file
3180 at the same time. The added code performs the same test as UW imapd; it
3181 checks after acquiring the lock that its file descriptor still refers to
3182 the same named file.
3184 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3185 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3188 11. Added the smtp_active_hostname option. If used, this will typically be made
3189 to depend on the incoming interface address. Because $interface_address is
3190 not set up until the daemon has forked a reception process, error responses
3191 that can happen earlier (such as "too many connections") no longer contain
3194 12. If an expansion in a condition on a "warn" statement fails because a lookup
3195 defers, the "warn" statement is abandoned, and the next ACL statement is
3196 processed. Previously this caused the whole ACL to be aborted.
3198 13. Added the iplsearch lookup type.
3200 14. Added ident_timeout as a log selector.
3202 15. Added tls_certificate_verified as a log selector.
3204 16. Added a global option tls_require_ciphers (compare the smtp transport
3205 option of the same name). This controls incoming TLS connections.
3207 17. I finally figured out how to make tls_require_ciphers do a similar thing
3208 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3209 before starting the TLS session.
3211 18. Tabs are now shown as \t in -bP output.
3213 19. If the log selector return_path_on_delivery was set, Exim crashed when
3214 bouncing a message because it had too many Received: header lines.
3216 20. If two routers both had headers_remove settings, and the first one included
3217 a superfluous trailing colon, the final name in the first list and the
3218 first name in the second list were incorrectly joined into one item (with a
3219 colon in the middle).
3225 1. Added -C and -D options to the exinext utility, mainly to make it easier
3226 to include in the automated testing, but these could be helpful when
3227 multiple configurations are in use.
3229 2. The exinext utility was not formatting the output nicely when there was
3230 an alternate port involved in the retry record key, nor when there was a
3231 message id as well (for retries that were specific to a specific message
3232 and a specific host). It was also confused by IPv6 addresses, because of
3233 the additional colons they contain. I have fixed the IPv4 problem, and
3234 patched it up to do a reasonable job for IPv6.
3236 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3237 delivery, the log line now contains "pipelined" if PIPELINING was used.
3239 4. An SMTP transport process used to panic and die if the bind() call to set
3240 an explicit outgoing interface failed. This has been changed; it is now
3241 treated in the same way as a connect() failure.
3243 5. A reference to $sender_host_name in the part of a conditional expansion
3244 that was being skipped was still causing a DNS lookup. This no longer
3247 6. The def: expansion condition was not recognizing references to header lines
3248 that used bh_ and bheader_.
3250 7. Added the _cache feature to named lists.
3252 8. The code for checking quota_filecount in the appendfile transport was
3253 allowing one more file than it should have been.
3255 9. For compatibility with Sendmail, the command line option
3263 and sets the incoming protocol and host name (for trusted callers). The
3264 host name and its colon can be omitted when only the protocol is to be set.
3265 Note the Exim already has two private options, -pd and -ps, that refer to
3266 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3267 "s", but I don't think that's a major issue.
3269 10. A number of refactoring changes to the code, none of which should affect
3272 (a) The number of logging options was getting close to filling up the
3273 32-bit word that was used as a bit map. I have split them into two classes:
3274 those that are passed in the argument to log_write(), and those that are
3275 only ever tested independently outside of that function. These are now in
3276 separate 32-bit words, so there is plenty of room for expansion again.
3277 There is no change in the user interface or the logging behaviour.
3279 (b) When building, for example, log lines, the code previously used a
3280 macro that called string_cat() twice, in order to add two strings. This is
3281 not really sufficiently general. Furthermore, there was one instance where
3282 it was actually wrong because one of the argument was used twice, and in
3283 one call a function was used. (As it happened, calling the function twice
3284 did not affect the overall behaviour.) The macro has been replaced by a
3285 function that can join an arbitrary number of extra strings onto a growing
3288 (c) The code for expansion conditions now uses a table and a binary chop
3289 instead of a serial search (which was left over from when there were very
3290 few conditions). Also, it now recognizes conditions like "pam" even when
3291 the relevant support is not compiled in: a suitably worded error message is
3292 given if an attempt is made to use such a condition.
3294 11. Added ${time_interval:xxxxx}.
3296 12. A bug was causing one of the ddress fields not to be passed back correctly
3297 from remote delivery subprocesses. The field in question was not being
3298 subsequently used, so this caused to problems in practice.
3300 13. Added new log selectors queue_time and deliver_time.
3302 14. Might have fixed a bug in maildirsizefile handling that threw up
3303 "unexpected character" debug warnings, and recalculated the data
3304 unnecessarily. In any case, I expanded the warning message to give more
3307 15. Added the message "Restricted characters in address" to the statements in
3308 the default ACL that block characters like @ and % in local parts.
3310 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3311 Three changes have been made:
3313 (a) There was a serious bug; a negative response to MAIL caused the whole
3314 recipient domain to be cached as invalid, thereby blocking all messages
3315 to all local parts at the same domain, from all senders. This bug has
3316 been fixed. The domain is no longer cached after a negative response to
3317 MAIL if the sender used is not empty.
3319 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3322 (c) A new callout option, "use_sender" has been added for people who want
3323 the modified behaviour.
3329 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3332 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3333 indeed breaks things for older releases.
3335 3. Added additional logging to the case where there is a problem reading data
3336 from a filter that is running in a subprocess using a pipe, in order to
3337 try to track down a specific problem.
3339 4. Testing facility fudge: when running in the test harness and attempting
3340 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3341 getting "No route to host". Convert this to a timeout.
3343 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3346 6. Some OS don't have socklen_t but use size_t instead. This affects the
3347 fifth argument of getsockopt() amongst other things. This is now
3348 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3349 can be set for individual OS. I have set it for SunOS5, OSF1, and
3350 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3351 some earlier ones do not.
3353 7. Change 4.30/15 was not doing the test caselessly.
3355 8. The standard form for an IPv6 address literal was being rejected by address
3356 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3357 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3358 this, as well as the form without the "IPv6" on the front (but only when
3359 address literals are enabled, of course).
3361 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3363 10. Exim crashed if a message with an empty sender address specified by -f
3364 encountered a router with an errors_to setting. This could be provoked only
3365 by a command such as
3369 where an empty string was supplied; "<>" did not hit this bug.
3371 11. Installed PCRE release 4.5.
3373 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3374 remained set. It is now erased.
3376 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3377 times from message ids (which are base 36 rather than the normal 62).
3379 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3380 were being counted as actual protocol errors, and logged if the log
3381 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3382 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3383 and DATA following a set of rejected RCPTs do not count as protocol errors.
3384 In other words, Exim assumes they were pipelined, though this may not
3385 actually be the case. Of course, in all cases the client gets an
3386 appropriate error code.
3388 15. If a lookup fails in an ACL condition, a message about the failure may
3389 be available; it is used if testing the ACL cannot continue, because most
3390 such messages specify what the cause of the deferral is. However, some
3391 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3392 that caused an old message to be retained and used if a later statement
3393 caused a defer, replacing the real cause of the deferral.
3395 16. If an IP address had so many PTR records that the DNS lookup buffer
3396 was not large enough to hold them, Exim could crash while trying to process
3397 the truncated data. It now detects and logs this case.
3399 17. Further to 4.21/58, another change has been made: if (and only if) the
3400 first line of a message (the first header line) ends with CRLF, a bare LF
3401 in a subsequent header line has a space inserted after it, so as not to
3402 terminate the header.
3404 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3405 unnecessarily, used atoi() instead of strtol(), and didn't check the
3406 termination when getting file sizes from file names by regex.
3408 19. Completely re-implemented the support for maildirsize files, in the light
3409 of a number of problems with the previous contributed implementation
3410 (4.30/29). In particular:
3412 . If the quota is zero, the maildirsize file is maintained, but no quota is
3415 . If the maildir directory does not exist, it is created before any attempt
3416 to write a maildirsize file.
3418 . The quota value in the file is just a cache; if the quota is changed in
3419 the transport, the new value overrides.
3421 . A regular expression is available for excluding directories from the
3424 20. The autoreply transport checks the characters in options that define the
3425 message's headers; it allows continued headers, but it was checking with
3426 isspace() after an embedded newline instead of explicitly looking for a
3429 21. If all the "regular" hosts to which an address was routed had passed their
3430 expiry times, and had not reached their retry times, the address was
3431 bounced, even if fallback hosts were defined. Now Exim should go on to try
3434 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3435 equivalent code in the SMTP transport. Some hosts send humungous responses
3436 to HELO/EHLO, more than 1024 it seems.
3438 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3439 gives compiler warnings in some environments. I've now done it "properly",
3442 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3443 (because of problems with the built-in one) was declared to return uschar *
3444 instead of char *, causing compiler failure.
3446 25. Fixed a file descriptor leak when processing alias/forward files.
3448 26. Fixed a minor format string issue in dbfn.c.
3450 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3452 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3453 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3456 29. When a sender address is verified, it is cached, to save repeating the test
3457 when there is more than one recipient in a message. However, when the
3458 verification involves a callout, it is possible for different callout
3459 options to be set for different recipients. It is too complicated to keep
3460 track of this in the cache, so now Exim always runs a verification when a
3461 callout is required, relying on the callout cache for the optimization.
3462 The overhead is duplication of the address routing, but this should not be
3465 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3466 to be verified with callout=postmaster, and the main callout worked but the
3467 postmaster check failed, the verification correctly failed. However, if a
3468 subsequent RCPT command asked for sender verification *without* the
3469 postmaster check, incorrect caching caused this verification also to fail,
3472 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3473 it was not caching the DNS options (qualify_single, search_parents) that
3474 were used when the lookup failed. A subsequent lookup with different
3475 options therefore always gave the same answer, though there were cases
3476 where it should not have. (Example: a "domains = !$mx_any" option on a
3477 dnslookup router: the "domains" option is always processed without any
3478 widening, but the router might have qualify_single set.) Now Exim uses the
3479 cached value only when the same options are set.
3481 32. Added John Jetmore's "exipick" utility to the distribution.
3483 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3484 than a timeout (e.g. a certificate is required, and is not provided), an
3485 Exim server now closes the connection immediately. Previously it waited for
3486 the client to close - but if the client is SSL, it seems that they each
3487 wait for each other, leading to a delay before one of them times out.
3489 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3490 maintained 0.8.x compatibility because I don't think many are using it, and
3491 it is clearly obsolete.
3493 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3496 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3497 client certificate was expired. A simple patch fixes this, though I don't
3498 understand the full logic of why the verify callback is called multiple
3501 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3502 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3503 which causes problems with some clients (such as the Certicom SSL Plus
3504 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3505 disables the coutermeasure allowing Eudora to connect."
3507 38. Exim was not checking that a write() to a log file succeeded. This could
3508 lead to Bad Things if a log got too big, in particular if it hit a file
3509 size limit. Exim now panics and dies if it cannot write to a log file, just
3510 as it does if it cannot open a log file.
3512 39. Modified OS/Makefile-Linux so that it now contains
3514 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3516 The two -D definitions ensure that Exim is compiled with large file
3517 support, which makes it possible to handle log files that are bigger than
3520 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3521 instance) a domain was checked against a named list that involved a lookup,
3522 causing $domain_data to be set, then another domain was checked against the
3523 same list, then the first domain was re-checked, the value of $domain_data
3524 after the final check could be wrong. In particular, if the second check
3525 failed, it could be set empty. This bug probably also applied to
3528 41. The strip_trailing_dot option was not being applied to the address given
3529 with the -f command-line option.
3531 42. The code for reading a message's header from the spool was incrementing
3532 $received_count, but never initializing it. This meant that the value was
3533 incorrect (doubled) while delivering a message in the same process in which
3534 it was received. In the most common configuration of Exim, this never
3535 happens - a fresh exec is done - but it can happen when
3536 deliver_drop_privilege is set.
3538 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3539 it now includes up to 150 characters of the unexpected data in the log
3542 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3543 and building data strings. The size of both of these buffers was 10 000
3544 bytes - far larger than anybody would *ever* want, thought I. Needless to
3545 say, somebody hit the limit. I have increased the maximum line length to
3546 20 000 and the maximum data length of concatenated lines to 100 000. I have
3547 also fixed two bugs, because there was no checking on these buffers. Tsk,
3548 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3549 buffer is too small.
3551 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3552 lsearch lookups. Now it does.
3554 46. When parsing a route_list item in a manualroute router, a fixed-length
3555 buffer was used for the list of hosts. I made this 1024 bytes long,
3556 thinking that nobody would ever have a list of hosts that long. Wrong.
3557 Somebody had a whole pile of complicated expansion conditions, and the
3558 string was silently truncated, leading to an expansion error. It turns out
3559 that it is easier to change to an unlimited length (owing to other changes
3560 that have happened since this code was originally written) than to build
3561 structure for giving a limitation error. The length of the item that
3562 expands into the list of hosts is now unlimited.
3564 47. The lsearch lookup could not handle data where the length of text line was
3565 more than 4095 characters. Such lines were truncated, leading to shortened
3566 data being returned. It should now handle lines of any length.
3568 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3569 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3572 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3573 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3574 no longer happens in comment lines. A list of replacements is now placed
3575 at the head of all of the source files, except those whose only change is
3576 to replace PERL_COMMAND in the very first #! line.
3578 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3579 messages on the queue, with a bottom-up merge sort, using code contributed
3580 by Michael Haardt. This should make operations like -bp somewhat faster on
3581 large queues. It won't affect queue runners, except when queue_run_in_order
3584 51. Installed eximstats 1.31 in the distribution.
3586 52. Added support for SRV lookups to the dnslookup router.
3588 53. If an ACL referred to $message_body or $message_body_end, the value was not
3589 reset for any messages that followed in the same SMTP session.
3591 54. The store-handling optimization for building very long strings was not
3592 differentiating between the different store pools. I don't think this
3593 actually made any difference in practice, but I've tidied it.
3595 55. While running the routers to verify a sender address, $sender_address
3596 was still set to the sender address. This is wrong, because when routing to
3597 send a bounce to the sender, it would be empty. Therefore, I have changed
3598 it so that, while verifying a sender address, $sender_address is set to <>.
3599 (There is no change to what happens when verifying a recipient address.)
3601 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3602 target A or AAAA records (if not already returned) without resetting the
3603 qualify_single or search_parents options of the DNS resolver. These are
3604 inappropriate in this case because the targets of MX and SRV records must
3605 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3606 target that, when qualified, matched something in the local domain. These
3607 two options are now turned off when doing these lookups.
3609 57. It seems that at least some releases of Reiserfs (which does not have the
3610 concept of a fixed number of inodes) returns zero and not -1 for the
3611 number of available inodes. This interacted badly with check_spool_inodes,
3612 which assumed that -1 was the "no such thing" setting. What I have done is
3613 to check that the total number of inodes is greater than zero before doing
3614 the test of how many are available.
3616 58. When a "warn" ACL statement has a log_message modifier, the message is
3617 remembered, and not repeated. This is to avoid a lot of repetition when a
3618 message has many recipients that cause the same warning to be written.
3619 Howewer, Exim was preserving the list of already written lines for an
3620 entire SMTP session, which doesn't seem right. The memory is now reset if a
3621 new message is started.
3623 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3624 debugging output unless log_rewrite was also set.
3626 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3627 of (int)(handle) when we know that handle contains (void *)(-1).
3629 61. The Exim daemon panic-logs an error return when it closes the incoming
3630 connection. However "connection reset by peer" seems to be common, and
3631 isn't really an error worthy of noting specially, so that particular error
3634 62. When Exim is trying to find all the local interfaces, it used to panic and
3635 die if the ioctl to get the interface flags failed. However, it seems that
3636 on at least one OS (Solaris 9) it is possible to have an interface that is
3637 included in the list of interfaces, but for which you get a failure error
3638 for this call. This happens when the interface is not "plumbed" into a
3639 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3640 failure of the "get flags" call assumes that the interface is down.
3642 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3643 makes life easier for people who are doing arithmetic on fields extracted
3644 from dates, where you often get leading zeros that should not be
3645 interpreted as octal.
3647 64. Added qualify_domain to the redirect router, to override the global
3650 65. If a pathologically long header line contained very many addresses (the
3651 report of this problem mentioned 10 000) and each of them was rewritten,
3652 Exim could use up a very large amount of memory. (It kept on making new
3653 copies of the header line as it rewrote, and never released the old ones.)
3654 At the expense of a bit more processing, the header rewriting function has
3655 been changed so that it no longer eats memory in this way.
3657 66. The generation of the Received: header has been moved from the time that a
3658 message starts to be received, to the time that it finishes. The timestamp
3659 in the Received: header should now be very close to that of the <= log
3660 line. There are two side-effects of this change:
3662 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3663 logged header lines no longer include the local Received: line, because
3664 it has not yet been created. The same applies to a copy of the message
3665 that is returned to a non-SMTP sender when a message is rejected.
3667 (b) When a filter file is tested using -bf, no additional Received: header
3668 is added to the test message. After some thought, I decided that this
3671 This change does not affect the value of $received_for. It is still set
3672 after address rewriting, but before local_scan() is called.
3674 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3676 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3677 gave an unhelpful panic error message, and a defer error. I have managed to
3678 change this behaviour so that it now rejects any supplied certificate,
3679 which seems right, as the list of acceptable certificates is empty.
3681 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3682 gave an unhelpful defer error. I have not managed to make this reject any
3683 supplied certificates, but the error message it gives is "no certificate
3684 supplied", which is not helpful.
3686 70. exigrep's output now also includes lines that are not associated with any
3687 message, but which match the given pattern. Implemented by a patch from
3688 Martin Sluka, which also tidied up the Perl a bit.
3690 71. Recipient callout verification, like sender verification, was using <> in
3691 the MAIL FROM command. This isn't really the right thing, since the actual
3692 sender may affect whether the remote host accepts the recipient or not. I
3693 have changed it to use the actual sender in the callout; this means that
3694 the cache record is now keyed on a recipient/sender pair, not just the
3695 recipient address. There doesn't seem to be a real danger of callout loops,
3696 since a callout by the remote host to check the sender would use <>.
3697 [SEE ABOVE: changed after hitting problems.]
3699 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3700 temporary errors. However, in the case of such a code being given after
3701 the end of a data transmission (i.e. after ".") Exim was failing to write
3702 a retry record for the message. (Yes, there was some broken host that was
3703 actually sending 8xx at this point.)
3705 73. An unknown lookup type in a host list could cause Exim to panic-die when
3706 the list was checked. (An example that provoked this was putting <; in the
3707 middle of a list instead of at the start.) If this happened during a DATA
3708 ACL check, a -D file could be left lying around. This kind of configuration
3709 error no longer causes Exim to die; instead it causes a defer errror. The
3710 incident is still logged to the main and panic logs.
3712 74. Buglet left over from Exim 3 conversion. The message "too many messages
3713 in one connection" was written to the rejectlog but not the mainlog, except
3714 when address rewriting (yes!) was being logged.
3716 75. Added write_rejectlog option.
3718 76. When a system filter was run not as root (that is, when system_filter_user
3719 was set), the values of the $n variables were not being returned to the
3720 main process; thus, they were not subsequently available in the $sn
3723 77. Added +return_path_on_delivery log selector.
3725 78. A connection timeout was being treated differently from recipients deferred
3726 when testing hosts_max_try with a message that was older than the host's
3727 retry timeout. (The host should not be counted, thus allowing all hosts to
3728 be tried at least once before bouncing.) This may have been the cause of an
3729 occasionally reported bug whereby a message would remain on the queue
3730 longer than the retry timeout, but would be bounced if a delivery was
3731 forced. I say "may" because I never totally pinned down the problem;
3732 setting up timeout/retry tests is difficult. See also the next item.
3734 79. The ultimate address timeout was not being applied to errors that involved
3735 a combination of host plus message (for example, a timeout on a MAIL
3736 command). When an address resolved to a number of possible hosts, and they
3737 were not all tried for each delivery (e.g. because of hosts_max_try), a
3738 message could remain on the queue longer than the retry timeout.
3740 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3743 81. Fixed an obscure SMTP outgoing bug which required at least the following
3744 conditions: (a) there was another message waiting for the same server;
3745 (b) the server returned 5xx to all RCPT commands in the first message so
3746 that the message was not completed; (c) the server dropped the connection
3747 or gave a negative response to the RSET that Exim sends to abort the
3748 transaction. The observed case was a dropped connection after DATA that had
3749 been sent in pipelining mode. That is, the server had advertised PIPELINING
3750 but was not implementing it correctly. The effect of the bug was incorrect
3751 behaviour, such as trying another host, and this could lead to a crash.
3757 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3758 and daemon.c were passed as pointers to ints; they should have been
3759 pointers to socklen_t variables (which are typically unsigned ints).
3761 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3764 3. Fixed a really odd bug that affected only the testing scheme; patching a
3765 certain fixed string in the binary changed the value of another string that
3766 happened to be identical to the end of the original first string.
3768 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3769 name", it returns that address as the IP address. On some operating
3770 systems (e.g. Solaris), it also passes back the IP address string as the
3771 "host name". However, on others (e.g. Linux), it passes back an empty
3772 string. Exim wasn't checking for this, and was changing the host name to an
3773 empty string, assuming it had been canonicized.
3775 5. Although rare, it is permitted to have more than one PTR record for a given
3776 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3777 all the names associated with an address, because they do in Solaris.
3778 However, it seems that they do not in Linux for data that comes from the
3779 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3780 I found this out when I moved to a new Linux workstation and tried to run
3781 the Exim test suite.
3783 To get round this problem I have changed the code so that it now does its
3784 own call to the DNS to look up PTR records when searching for a host name.
3785 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3786 addresses that are only in /etc/hosts are still found.
3788 This behaviour is, however, controlled by an option called host_lookup_
3789 order, which defaults to "bydns:byaddr". If people want to use the other
3790 order, or indeed, just use one or the other means of lookup, they can
3791 specify it in this variable.
3793 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3794 some operating systems, this comes back from gethostbyaddr() as an empty
3795 string, and this is what Exim used to test for. However, it seems that in
3796 other systems, "." is yielded. Exim now tests for this case too.
3798 7. The values of check_spool_space and check_log_space are now held internally
3799 as a number of kilobytes instead of an absolute number of bytes. If a
3800 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3801 kilobyte. This means that much larger values can be stored.
3803 8. Exim monitor: an attempt to get the action menu when not actually pointing
3804 at a message produces an empty menu entitled "No message selected". This
3805 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3806 no entries in it ("Shell widget menu has zero width and/or height"). So I
3807 have added a single, blank menu entry in this case.
3809 9. Added ${quote_local_part.
3811 10. MIME decoding is now applied to the contents of Subject: header lines when
3814 11. Now that a reference to $sender_host_address automatically causes a reverse
3815 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3816 host lookup before query-style lookups in lists that might use this
3817 variable. This has therefore been abolished, and the "net-" prefix is no
3818 longer necessary for query-style lookups.
3820 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3821 have been a typo for LFLAGS, so it has been changed.
3823 13. The install script calls Exim with "-C /dev/null" in order to find the
3824 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3825 to be output. Howeve, since Exim outputs its version number before the
3826 error, it didn't break the script. It just looked ugly. I fixed this by
3827 always allowing "-C /dev/null" if the caller is root.
3829 14. Ignore overlarge ACL variable number when reading spool file - insurance
3830 against a later release with more variables having written the file.
3832 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3833 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3834 this, as well as the form without the "IPv6" on the front.
3836 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3837 OS/Makefile-Darwin file.
3839 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3840 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3842 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3843 variable. However, some OS use ints for inodes. I've added cast to long int
3844 to get rid of the compiler warning.
3846 19. I had forgotten to lock out "/../" in configuration file names when
3847 ALT_CONFIG_PREFIX was set.
3849 20. Routers used for verification do not need to specify transports. However,
3850 if such a router generated a host list, and callout was configured, Exim
3851 crashed, because it could not find a port number from the (non-existent)
3852 transport. It now assumes port 25 in this circumstance.
3854 21. Added the -t option to exigrep.
3856 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3857 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3858 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3859 them is set, LOOKUP_LSEARCH is forced.
3861 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3863 24. Added sender and host information to the "rejected by local_scan()" log
3864 line; previously there was no indication of these.
3866 25. Added .include_if_exists.
3868 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3869 Linux. It turns out that not all file systems support this. Apparently some
3870 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3871 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3872 not supported on the file descriptor, is now ignored when Exim is trying to
3873 sync a directory. This applies only to Linux.
3875 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3877 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3878 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3881 29. Added support for maildirsize files from supplied patch (modified a bit).
3883 30. The use of :fail: followed by an empty string could lead Exim to respond to
3884 sender verification failures with (e.g.):
3886 550 Verification failed for <xxx>
3887 550 Sender verify failed
3889 where the first response line was missing the '-' that indicates it is not
3890 the final line of the response.
3892 31. The loop for finding the name of the user that called Exim had a hardwired
3893 limit of 10; it now uses the value of finduser_retries, which is used for
3894 all other user lookups.
3896 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3899 33. Exim was neglecting to zero errno before one call of strtol() when
3900 expanding a string and expecting an integer value. On some systems this
3901 resulted in spurious "integer overflow" errors. Also, it was casting the
3902 result into an int without checking.
3904 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3905 did not work. The code looks as if it has *never* worked, though it appears
3906 to have been documented since at least releast 1.62. I have made it work.
3908 35. The "timeout_DNS" error in retry rules, also documented since at least
3909 1.62, also never worked. As it isn't clear exactly what this means, and
3910 clearly it isn't a major issue, I have abolished the feature by treating it
3911 as "timeout", and writing a warning to the main and panic logs.
3913 36. The display of retry rules for -brt wasn't always showing the error code
3916 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3917 timeout_connect_A, timeout_connect_MX.
3919 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3920 to the empty sender.
3922 39. The daemon was not analysing the content of -oX till after it had closed
3923 stderr and disconnected from the controlling terminal. This meant that any
3924 syntax errors were only noted on the panic log, and the return code from
3925 the command was 0. By re-arranging the code a little, I've made the
3926 decoding happen first, so such errors now appear on stderr, and the return
3927 code is 1. However, the actual setting up of the sockets still happens in
3928 the disconnected process, so errors there are still only recorded on the
3931 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3932 connections (as happens on some IP stacks) was logged at start up time as
3933 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3934 it from "IPv6 and IPv4", which means that two separate sockets are being
3937 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3938 says whether AF_INET or AF_INET6 was passed as an argument.
3940 42. Exiwhat output was messed up when time zones were included in log
3943 43. Exiwhat now gives more information about the daemon's listening ports,
3944 and whether -tls-on-connect was used.
3946 44. The "port" option of the smtp transport is now expanded.
3948 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3949 silently ignored. Now an error message is written to the main and panic
3952 46. There's a new ACL modifier called "logwrite" which writes to a log file
3953 as soon as it is encountered.
3955 47. Added $local_user_uid and $local_user_gid at routing time.
3957 48. Exim crashed when trying to verify a sender address that was being
3960 49. Exim was recognizing only a space character after ".include". It now also
3961 recognizes a tab character.
3963 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3964 extracting the relevant information from the specification. The man page no
3965 longer contains scrambled data for the -d option, and I've added a section
3966 at the front about calling Exim under different names.
3968 51. Added "extra_headers" argument to the "mail" command in filter files.
3970 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3973 53. Installed eximstats 1.29.
3975 54. Added transport_filter_timeout as a generic transport option.
3977 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3978 Cc: header lines. This was required by RFC 822, but it not required by RFC
3981 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3982 incoming messages that did not have them. Now it does so only if the
3983 message originates locally, that is, if there is no associated remote host
3984 address. When Resent- header lines are present, this applies to the Resent-
3985 lines rather than the non-Resent- lines.
3987 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3988 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3990 58. Messages for configuration errors now include the name of the main
3991 configuration file - useful now that there may be more than one file in a
3992 list (.included file names were always shown).
3994 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3995 for those rare installations that do not start the daemon as root or run it
3996 setuid root. I've cut out the call to initgroups() if the daemon is not
3999 60. The Exim user and group can now be bound into the binary as text strings
4000 that are looked up at the start of Exim's processing.
4002 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4004 62. Added $mailstore_basename variable.
4006 63. Installed patch to sieve.c from Michael Haardt.
4008 64. When Exim failed to open the panic log after failing to open the main log,
4009 the original message it was trying to log was written to stderr and debug
4010 output, but if they were not available (the usual case in production), it
4011 was lost. Now it is written to syslog before the two lines that record the
4012 failures to open the logs.
4014 65. Users' Exim filters run in subprocesses under the user's uid. It is
4015 possible for a "deliver" command or an alias in a "personal" command to
4016 provoke an address rewrite. If logging of address rewriting is configured,
4017 this fails because the process is not running as root or exim. There may be
4018 a better way of dealing with this, but for the moment (because 4.30 needs
4019 to be released), I have disabled address rewrite logging when running a
4020 filter in a non-root, non-exim process.
4026 1. The buildconfig auxiliary program wasn't quoting the value set for
4027 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4028 not defined. This bug was masked in 4.22 by the effect that was fixed in
4031 2. Some messages that were rejected after a message id was allocated were
4032 shown as "incomplete" by exigrep. It no longer does this for messages that
4033 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4035 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4036 have allow_domain_literals set, the ID did not get logged in the <= line.
4037 Domain literals are now always recognized in Message-ID: header lines.
4039 4. The first argument for a ${extract expansion item is the key name or field
4040 number. Leading and trailing spaces in this item were not being ignored,
4041 causing some misleading effects.
4043 5. When deliver_drop_privilege was set, single queue runner processes started
4044 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4045 same command in the process it spins off) were not dropping privilege.
4047 6. When the daemon running as "exim" started a queue runner, it always
4048 re-executed Exim in the spun-off process. This is a waste of effort when
4049 deliver_drop_privilege is set. The new process now just calls the
4050 queue-runner function directly.
4056 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4059 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4060 ignored. Though the use of -oP was forcing the writing of a pid file, it
4061 was always written to the default place.
4063 3. If the message "no IP address found for host xxxx" is generated during
4064 incoming verification, it is now followed by identification of the incoming
4065 connection (so you can more easily find what provoked it).
4067 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4069 5. Added some features to "harden" Exim a bit more against certain attacks:
4071 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4072 be put in Local/Makefile. This is like the never_users runtime option,
4073 but it cannot be overridden. The default setting is "root".
4075 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4076 prefix string with which any file named in a -C command line option
4079 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4080 is retained for -C and -D only if the caller of Exim is root. Without
4081 it, the exim user may also use -C and -D and retain privilege.
4083 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4084 command line option is disabled.
4086 6. Macro names set by the -D option must start with an upper case letter, just
4087 like macro names defined in the configuration file.
4089 7. Added "dereference=" facility to LDAP.
4091 8. Two instances of the typo "uknown" in the source files are fixed.
4093 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4094 the Configure-Makefile script screwed up while processing it.
4096 10. Incorporated PCRE 4.4.
4098 11. The SMTP synchronization check was not operating right at the start of an
4099 SMTP session. For example, it could not catch a HELO sent before the client
4100 waited for the greeting. There is now a check for outstanding input at the
4101 point when the greeting is written. Because of the duplex, asynchronous
4102 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4103 way, but not yet received, when the check is performed.
4105 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4106 on TCP/IP sockets, because this apparently causes some broken clients to
4109 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4110 unchanged) from the Cygwin maintainer.
4112 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4113 was supported instead of testing for mbx. Effectively a typo.
4115 15. The spa authenticator server code was not checking that the input it
4116 received was valid base64.
4118 16. The debug output line for the "set" modifier in ACLs was not showing the
4119 name of the variable that was being set.
4121 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4123 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4124 lookup on its own. Something else was needed to trigger the lookup. For
4125 example, a match in host_lookup or the need for a host name in a host list.
4126 Now, if $sender_host_name is referenced and the host name has not yet been
4127 looked up, a lookup is performed. If the lookup fails, the variable remains
4128 empty, and $host_lookup_failed is set to "1".
4130 19. Added "eqi" as a case-independent comparison operator.
4132 20. The saslauthd authentication condition could segfault if neither service
4133 nor realm was specified.
4135 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4136 error message that was logged was misleading, and incoming SMTP
4137 connections were dropped. The message is now more accurate, and temporary
4138 errors are given to SMTP connections.
4140 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4141 (except RSET) with a 421 error, until QUIT is received. However, it was
4142 failing to send a response to QUIT.
4144 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4145 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4146 ACL is now run after the helo_try_verify_hosts code.
4148 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4149 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4150 case-independent, so other case variants are also recognized). Apparently
4151 some systems use these upper case variants.
4153 25. If more than two messages were waiting for the same host, and a transport
4154 filter was specified for the transport, Exim sent two messages over the
4155 same TCP/IP connection, and then failed with "socket operation on non-
4156 socket" when it tried to send the third.
4158 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4160 27. The extern definition of crypt16() in expand.c was not being excluded when
4161 the OS had its own crypt16() function.
4163 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4164 as a preferred synonym for return_size_limit, both as an option and as an
4167 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4169 30. Changed the default configuration ACL to relax the local part checking rule
4170 for addresses that are not in any local domains. For these addresses,
4171 slashes and pipe symbols are allowed within local parts, but the sequence
4172 /../ is explicitly forbidden.
4174 31. SPA server authentication was not clearing the challenge buffer before
4177 32. log_message in a "warn" ACL statement was writing to the reject log as
4178 well as to the main log, which contradicts the documentation and doesn't
4179 seem right (because no rejection is happening). So I have stopped it.
4181 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4182 However, I am unable to do any testing of this.
4184 34. Fixed an infelicity in the appendfile transport. When checking directories
4185 for a mailbox, to see if any needed to be created, it was accidentally
4186 using path names with one or more superfluous leading slashes; tracing
4187 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4189 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4190 contents are added to the log line that is written for every discarded
4191 recipient. (Previously a log_message setting was ignored.)
4193 36. The ${quote: operator now quotes the string if it is empty.
4195 37. The install script runs exim in order to find its version number. If for
4196 some reason other than non-existence or emptiness, which it checks, it
4197 could not run './exim', it was installing it with an empty version number,
4198 i.e. as "exim-". This error state is now caught, and the installation is
4201 38. An argument was missing from the function that creates an error message
4202 when Exim fails to connect to the socket for saslauthd authentication.
4203 This could cause Exim to crash, or give a corrupted message.
4205 39. Added isip, isip4, and isip6 to ${if conditions.
4207 40. The ACL variables $acl_xx are now saved with the message, and can be
4208 accessed later in routers, transports, and filters.
4210 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4211 strings in the file are not string-expanded.
4213 42. If a MAIL command specified a SIZE value that was too large to fit into an
4214 int variable, the check against message_size_limit failed. Such values are
4215 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4216 day this will have to be increased, but I don't think I want to be around
4217 when emails are that large.
4224 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4225 iconv() is not standard in FreeBSD.
4227 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4228 IPv6 enabled. The observed symptom was a segmentation fault on return from
4229 the function os_common_find_running_interfaces() in src/os.c.
4231 3. In the check_special_case() function in daemon.c I had used "errno" as an
4232 argument name, which causes warnings on some systems. This was basically a
4233 typo, since it was named "eno" in the comments!
4235 4. The code that waits for the clock to tick (at a resolution of some fraction
4236 of a second) so as to ensure message-id uniqueness was always waiting for
4237 at least one whole tick, when it could have waited for less. [This is
4238 almost certainly not relevant at current processor speeds, where it is
4239 unlikely to ever wait at all. But we try to future-proof.]
4241 5. The function that sleeps for a time interval that includes fractions of a
4242 second contained a race. It did not block SIGALRM between setting the
4243 timer, and suspending (a couple of lines later). If the interval was short
4244 and the sigsuspend() was delayed until after it had expired, the suspension
4245 never ended. On busy systems this could lead to processes getting stuck for
4248 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4249 process, before it forks any delivery processes. The open lookup caching
4250 mechanism meant that the open file or database connection was passed into
4251 the delivery process. The problem was that delivery processes always tidy
4252 up cached lookup data. This could cause a problem for the next delivery
4253 process started by the queue runner, because the external queue runner
4254 process does not know about the closure. So the next delivery process
4255 still has data in the lookup cache. In the case of a file lookup, there was
4256 no problem because closing a file descriptor in a subprocess doesn't affect
4257 the parent. However, if the lookup was caching a connection to a database,
4258 the connection was closed, and the second delivery process was likely to
4259 see errors such as "PGSQL: query failed: server closed the connection
4260 unexpectedly". The problem has been fixed by closing all cached lookups
4261 in a queue runner before running a delivery process.
4263 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4264 seem to have the "const" qualifier which it has on other OS. I've
4267 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4268 *of the same type* (client or server) with the same public name that an
4269 error should be diagnosed.
4271 9. When Exim looked up a host name for an IP address, but failed to find the
4272 original IP address when looking up the host name (a safety check), it
4273 output the message "<ip address> does not match any IP for NULL", which was
4274 confusing, to say the least. The bug was that the host name should have
4275 appeared instead of "NULL".
4277 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4278 user that is built into the binary, and the -C or -D options is used, root
4279 privilege is dropped before the configuration file is read. In addition,
4280 logging is switched to stderr instead of the normal log files. If the
4281 configuration then re-defines the Exim user, the unprivileged environment
4282 is probably not what is expected, so Exim logs a panic warning message (but
4285 However, if deliver_drop_privilege is set, the unprivileged state may well
4286 be exactly what is intended, so the warning has been cut out in that case,
4287 and Exim is allowed to try to write to its normal log files.
4293 1. smtp_return_error_details was not giving details for temporary sender
4294 or receiver verification errors.
4296 2. Diagnose a configuration error if two authenticators have the same public
4299 3. Exim used not to create the message log file for a message until the first
4300 delivery attempt. This could be confusing when incoming messages were held
4301 for policy or load reasons. The message log file is now created at the time
4302 the message is received, and an initial "Received" line is written to it.
4304 4. The automatically generated man page for command line options had a minor
4305 bug that caused no ill effects; however, a more serious problem was that
4306 the procedure for building the man page automatically didn't always
4307 operate. Consequently, release 4.20 contains an out-of-date version. This
4308 shouldn't happen again.
4310 5. When building Exim with embedded Perl support, the script that builds the
4311 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4312 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4314 6. The freeze_tell option was not being used for messages that were frozen on
4315 arrival, either by an ACL or by local_scan().
4317 7. Added the smtp_incomplete_transaction log selector.
4319 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4320 was accepting AUTH without a new EHLO.
4322 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4323 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4326 10. Logging of TCP/IP connections (when configured) now happens in the main
4327 daemon process instead of the child process, so that the TCP/IP connection
4328 count is more accurate (but it can never be perfect).
4330 11. The use of "drop" in a nested ACL was not being handled correctly in the
4331 outer ACL. Now, if condition failure induced by the nested "drop" causes
4332 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4333 or "require"), the connection is dropped.
4335 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4336 that yield "discard" can now be used with an "accept" or a "discard" verb,
4337 but an error is generated for any others (because I can't see a useful way
4338 to define what should happen).
4340 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4341 are now processed in the same way as lines in the Exim configuration file.
4342 In particular, continuation lines are supported.
4344 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4346 15. Added -ti meaning -t -i.
4348 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4349 domains, and warn by logging if others are found.
4351 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4352 returned from GIFCONF when Exim is trying to find the list of interfaces on
4353 a host. The code in os.c has been modified to copy each ifreq to an aligned
4354 structure in all cases.
4356 Also, in some cases, the returned ifreq's were being copied to a 'struct
4357 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4358 means the last couple of bytes of an IPv6 address could be chopped if the
4359 ifreq contained only a normal sockaddr (14 bytes storage).
4361 18. Named domain lists were not supported in the hosts_treat_as_local option.
4362 An entry such as +xxxx was not recognized, and was treated as a literal
4365 19. Ensure that header lines added by a DATA ACL are included in the reject log
4366 if the ACL subsequently rejects the message.
4368 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4369 MD5 (which is deprecated).
4371 21. When testing a filter file using -bf, Exim was writing a message when it
4372 took the sender from a "From " line in the message, but it was not doing so
4373 when it took $return_path from a Return-Path: header line. It now does.
4375 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4376 with a valid header line field name (a series of printing characters
4377 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4379 23. Changed "disc" in the source to "disk" to conform to the documentation and
4380 the book and for uniformity.
4382 24. Ignore Sendmail's -Ooption=value command line item.
4384 25. When execve() failed while trying to run a command in a pipe transport,
4385 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4386 could be confused with a return value of 69 from the command itself. This
4387 has been changed to 127, the value the shell returns if it is asked to run
4388 a non-existent command. The wording for the related log line suggests a
4389 non-existent command as the problem.
4391 26. If received_header_text expands to an empty string, do not add a Received:
4392 header line to the message. (Well, it adds a token one on the spool, but
4393 marks it "old" so that it doesn't get used or transmitted.)
4395 27. Installed eximstats 1.28 (addition of -nt option).
4397 28. There was no check for failure on the call to getsockname() in the daemon
4398 code. This can fail if there is a shortage of resources on the system, with
4399 ENOMEM, for example. A temporary error is now given on failure.
4401 29. Contrary to the C standard, it seems that in some environments, the
4402 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4403 program. Exim now does this explicitly; it affects the formatting of
4404 timestamps using strftime().
4406 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4407 complaints. I've now initialized all the variables, to avoid this.
4409 32. Header lines added by a system filter were not being "seen" during
4410 transport-time rewrites.
4412 33. The info_callback() function passed to OpenSSL is set up with type void
4413 (*)(SSL *, int, int), as described somewhere. However, when calling the
4414 function (actually a macro) that sets it up, the type void(*)() is
4415 expected. I've put in a cast to prevent warnings from picky compilers.
4417 34. If a DNS black list lookup found a CNAME record, but there were no A
4418 records associated with the domain it pointed at, Exim crashed.
4420 35. If a DNS black list lookup returned more than one A record, Exim ignored
4421 all but the first. It now scans all returned addresses if a particular IP
4422 value is being sought. In this situation, the contents of the
4423 $dnslist_value variable are a list of all the addresses, separated by a
4426 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4427 to accept a host name and all its aliases if the forward lookup for any of
4428 them yielded the IP address of the incoming connection. Now it accepts only
4429 those names whose forward lookup yields the correct IP address. Any other
4430 names are discarded. This closes a loophole whereby a rogue DNS
4431 administrator could create reverse DNS records to break through a
4432 wildcarded host restriction in an ACL.
4434 37. If a user filter or a system filter that ran in a subprocess used any of
4435 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4436 the wrong values were passed to the pipe command ($thisaddress had the
4437 value of $0, $0 had the value of $1, etc). This bug was introduced by
4438 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4440 38. Improved the line breaking for long SMTP error messages from ACLs.
4441 Previously, if there was no break point between 40 and 75 characters, Exim
4442 left the rest of the message alone. Two changes have been made: (a) I've
4443 reduced the minimum length to 35 characters; (b) if it can't find a break
4444 point between 35 and 75 characters, it looks ahead and uses the first one
4445 that it finds. This may give the occasional overlong line, but at least the
4446 remaining text gets split now.
4448 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4449 file descriptors might be low, and that setting 1000 would always raise it.
4450 It turns out that in some environments, the limit is already over 1000 and
4451 that lowering it causes trouble. So now Exim takes care not to decrease it.
4453 40. When delivering a message, the value of $return_path is set to $sender_
4454 address at the start of routing (routers may change the value). By an
4455 oversight, this default was not being set up when an address was tested by
4456 -bt or -bv, which affected the outcome if any router or filter referred to
4459 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4460 log only when "message" or "log_message" are set. However, if one of the
4461 conditions was an address verification, or a call to a nested ACL, the
4462 messages generated by the underlying test were being passed through. This
4463 no longer happens. The underlying message is available in $acl_verify_
4464 message for both "message" and "log_message" expansions, so it can be
4465 passed through if needed.
4467 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4468 new expansion $bh_ to give the encoded byte string without charset
4469 translation. Translation happens only if iconv() is available; HAVE_ICONV
4470 indicates this at build time. HEADERS_CHARSET gives the charset to
4471 translate to; headers_charset can change it in the configuration, and
4472 "headers charset" can change it in an individual filter file.
4474 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4475 that creates RFC 2047 encoded "words" labels them as that charset instead
4476 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4477 expansion operator; (ii) when Exim creates a From: line for a local
4478 message; (iii) when a header line is rewritten to include a "phrase" part.
4480 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4481 buggy, causing it to skip the first lines of messages whose message ID
4482 ended in 'D'. This would not have bitten before Exim release 4.14, because
4483 message IDs were unlikely to end in 'D' before then. The effect was to have
4484 incorrect size information for certain domains.
4486 45. #include "config.h" was missing at the start of the crypt16.c module. This
4487 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4490 46. If there was a timeout during a "random" callout check, Exim treated it as
4491 a failure of the random address, and carried on sending RSET and the real
4492 address. If the delay was just some slowness somewhere, the response to the
4493 original RCPT would be taken as a response to RSET and so on, causing
4494 mayhem of various kinds.
4496 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4497 when I implemented it. It didn't allow for the fact that some option values
4498 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4499 the right test for positive values.
4501 48. Domain names in DNS records are case-independent. Exim always looks them up
4502 in lower case. Some resolvers return domain names in exactly the case they
4503 appear in the zone file, that is, they may contain uppercase letters. Not
4504 all resolvers do this - some return always lower case. Exim was treating a
4505 change of case by a resolver as a change of domain, similar to a widening
4506 of a domain abbreviation. This triggered its re-routing code and so it was
4507 trying to route what was effectively the same domain again. This normally
4508 caused routing to fail (because the router wouldn't handle the domain
4509 twice). Now Exim checks for this case specially, and just changes the
4510 casing of the domain that it ultimately uses when it transmits the message
4513 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4516 50. If a filter generated a file delivery with a non-absolute name (possible if
4517 no home directory exists for the router), the forbid_file option was not
4520 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4521 the existing equality matching.
4523 52. Exim was using ints instead of ino_t variables in some places where it was
4524 dealing with inode numbers.
4526 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4527 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4528 called TMPDIR, and if it finds it is different, it changes its value.
4530 54. The smtp_printf() function is now made available to local_scan() so
4531 additional output lines can be written before returning. There is also an
4532 smtp_fflush() function to enable the detection of a dropped connection.
4533 The variables smtp_input and smtp_batched_input are exported to
4536 55. Changed the default runtime configuration: the message "Unknown user"
4537 has been removed from the ACL, and instead placed on the localuser router,
4538 using the cannot_route_message feature. This means that any verification
4539 failures that generate their own messages won't get overridden. Similarly,
4540 the "Unrouteable address" message that was in the ACL for unverifiable
4541 relay addresses has also been removed.
4543 56. Added hosts_avoid_esmtp to the smtp transport.
4545 57. The exicyclog script was not checking for the esoteric option
4546 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4547 will work only if exicyclog is run under the appropriate euid.
4549 58. Following a discussion on the list, the rules by which Exim recognises line
4550 endings on incoming messages have been changed. The -dropcr and drop_cr
4551 options are now no-ops, retained only for backwards compatibility. The
4552 following line terminators are recognized: LF CRLF CR. However, special
4553 processing applies to CR:
4555 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4556 nor a local message in the state where . is a terminator.
4558 (ii) If a bare CR is encountered in a header line, an extra space is added
4559 after the line terminator so as not to end the header. The reasoning
4560 behind this is that bare CRs in header lines are most likely either
4561 to be mistakes, or people trying to play silly games.
4563 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4564 was being incorrectly given as 18 bytes larger than it should have been.
4565 This is a VOB (very old bug).
4567 60. This may never have affected anything current, but just in case it has:
4568 When the local host is found other than at the start of a list of hosts,
4569 the local host, those with the same MX, and any that follow, are discarded.
4570 When the list in question was part of a longer list of hosts, the following
4571 hosts (not currently being processed) were also being discarded. This no
4572 longer happens. I'm not sure if this situation could ever has previously
4575 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4578 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4579 header. This is something that is recommended in a new Internet Draft, and
4580 is something that is documented as being done by Sendmail. There are two
4581 possible values. For messages generated by the autoreply transport, Exim
4584 Auto-Submitted: auto-replied
4586 whereas for all other generated messages (e.g. bounces) it adds
4588 Auto-Submitted: auto-generated
4590 63. The "personal" condition in filters now includes a test for the
4591 Auto-Submitted: header. If it contains the string "auto-" the message it
4592 not considered personal.
4594 64. Added rcpt_include_affixes as a generic transport option.
4596 65. Added queue_only_override (default true).
4598 66. Added the syslog_duplication option.
4600 67. If what should have been the first header line of a message consisted of
4601 a space followed by a colon, Exim was mis-interpreting it as a header line.
4602 It isn't of course - it is syntactically invalid and should therefore be
4603 treated as the start of the message body. The misbehaviour could have
4604 caused a number of strange effects, including loss of data in subsequent
4605 header lines, and spool format errors.
4607 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4608 client host had authenticated. This control can now be exercised by an ACL
4609 for more flexibility.
4611 69. By default, callouts do not happen when testing with -bh. There is now a
4612 variant, -bhc, which does actually run the callout code, including
4613 consulting and updating the callout cache.
4615 70. Added support for saslauthd authentication, courtesy of Alexander
4618 71. If statvfs() failed on the spool or log directories while checking their
4619 size for availability, Exim confusingly gave the error "space shortage".
4620 Furthermore, in debugging mode it crashed with a floating point exception.
4621 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4622 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4623 serious problem, Exim now writes to the main and panic logs when this
4624 happens, with details of the failure. It then refuses to accept the
4625 incoming message, giving the message "spool directory problem" or "log
4626 directory problem" with a 421 code for SMTP messages.
4628 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4629 0, 1, and 2 exist, because some OS complain for execs without them (see
4630 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4631 descriptors. However, the code omitted to check that the open succeeded,
4632 causing mysterious errors if for some reason the permissions on /dev/null
4633 got screwed. Now Exim writes a message to the main and panic logs, and
4634 bombs out if it can't open /dev/null.
4636 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4637 interact so that it is all more flexible. It is supposed to remain
4638 backwards compatible. Also added extra_local_interfaces.
4640 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4641 to bomb out with an assertion failure - to the client this appears as a
4642 connection drop. This problem occurs in the part of the code that was taken
4643 from the Samba project. Fortunately, the assertion is in a very simple
4644 function, so I have fixed this by reproducing the function inline in the
4645 one place where it is called, and arranging for authentication to fail
4646 instead of killing the process with assert().
4648 75. The SPA client code was not working when the server requested OEM rather
4649 than Unicode encoding.
4651 76. Added code to make require_files with a specific uid setting more usable in
4652 the case where statting the file as root fails - usually a non-root-mounted
4653 NFS file system. When this happens and the failure is EACCES, Exim now
4654 forks a subprocess and does the per-uid checking as the relevant uid.
4656 77. Added process_log_path.
4658 78. If log_file_path was not explicitly set, a setting of check_log_space or
4659 check_log_inodes was ignored.
4661 79. If a space check for the spool or log partitions fails, the incident is now
4662 logged. Of course, in the latter case the data may get lost...
4664 80. Added the %p formatting code to string_format() so that it can be used to
4665 print addresses in debug_print(). Adjusted all the address printing in the
4666 debugging in store.c to use %p rather than %d.
4668 81. There was a concern that a line of code in smtp_in.c could overflow a
4669 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4670 initially expressed, the concern was not well-founded, because trailing
4671 spaces are removed early. However, if the trailing spaces were followed by
4672 a NULL, they did not get removed, so the overflow was possible. Two fixes
4675 (a) I re-wrote the offending code in a cleaner fashion.
4676 (b) If an incoming SMTP command contains a NULL character, it is rejected
4679 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4680 runs initgroups(), so that if the Exim user is in any additional groups,
4681 they will be used during message reception.
4687 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob