1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.497 2007/03/21 15:10:39 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.
171 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
172 homing or because of multiple MX records with /mx) generated more than
173 one IP address, and the following item turned out to be the local host,
174 all the secondary addresses of the first item were incorrectly removed
175 from the list, along with the local host and any following hosts (which
176 is what is supposed to happen).
178 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
179 whoever called Exim into the -H file. In the case of the daemon it was
180 behaving confusingly. When first started, it used values for whoever
181 started the daemon, but after a SIGHUP it used the Exim user (because it
182 calls itself on a restart). I have changed the code so that it now always
185 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
186 message are rejected with the same error (e.g. no authentication or bad
187 sender address), and a DATA command is nevertheless sent (as can happen
188 with PIPELINING or a stupid MUA), the error message that was given to the
189 RCPT commands is included in the rejection of the DATA command. This is
190 intended to be helpful for MUAs that show only the final error to their
197 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
198 fixed by 4.65/MH/01 (is this a record?) are fixed:
200 (i) An empty string was always treated as zero by the numeric comparison
201 operators. This behaviour has been restored.
203 (ii) It is documented that the numeric comparison operators always treat
204 their arguments as decimal numbers. This was broken in that numbers
205 starting with 0 were being interpreted as octal.
207 While fixing these problems I realized that there was another issue that
208 hadn't been noticed. Values of message_size_limit (both the global option
209 and the transport option) were treated as octal if they started with 0.
210 The documentation was vague. These values are now always treated as
211 decimal, and I will make that clear in the documentation.
217 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
218 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
221 MH/01 Don't check that the operands of numeric comparison operators are
222 integers when their expansion is in "skipping" mode (fixes bug
223 introduced by 4.64-PH/07).
225 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
226 child addresses, Exim now panics and dies. Previously, because the count
227 is held in a short int, deliveries were likely to be lost. As such a
228 large number of recipients for a single message is ridiculous
229 (performance will be very, very poor), I have chosen to impose a limit
230 rather than extend the field.
236 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
237 leftover -K file (the existence of which was triggered by #402).
238 While we were at it, introduced process PID as part of the -K
239 filename. This should rule out race conditions when creating
242 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
243 processing considerably. Previous code took too long for large mails,
244 triggering a timeout which in turn triggers #401.
246 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
247 in the DK code in transports.c. sendfile() is not really portable,
248 hence the _LINUX specificness.
250 TF/01 In the add_headers option to the mail command in an Exim filter,
251 there was a bug that Exim would claim a syntax error in any
252 header after the first one which had an odd number of characters
255 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
256 callout verification, Exim cached a "reject" for the entire domain. This
257 is correct for most verifications, but it is not correct for a recipient
258 verification with use_sender or use_postmaster set, because in that case
259 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
260 case of MAIL FROM:<> rejection from other early rejections (e.g.
261 rejection of HELO). When verifying a recipient using a non-null MAIL
262 address, the cache is ignored if it shows MAIL FROM:<> rejection.
263 Whatever the result of the callout, the value of the domain cache is
264 left unchanged (for any other kind of callout, getting as far as trying
265 RCPT means that the domain itself is ok).
267 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
270 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
271 manifest itself as EPIPE rather than ECONNECT. When tidying away a
272 session, the daemon ignores ECONNECT errors and logs others; it now
273 ignores EPIPE as well.
275 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
276 (quoted-printable decoding).
278 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
279 later the small subsequent patch to fix an introduced bug.
281 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
283 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
285 PH/08 An error is now given if message_size_limit is specified negative.
287 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
288 to be given (somewhat) arbitrary names.
290 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
293 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
294 miscellaneous code fixes
296 PH/10 Added the log_reject_target ACL modifier to specify where to log
299 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
300 hostname. This is wrong, because it relates to the incoming message (and
301 probably the interface on which it is arriving) and not to the outgoing
302 callout (which could be using a different interface). This has been
303 changed to use the value of the helo_data option from the smtp transport
304 instead - this is what is used when a message is actually being sent. If
305 there is no remote transport (possible with a router that sets up host
306 addresses), $smtp_active_hostname is used.
308 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
309 tweaks were necessary in order to get it to work (see also 21 below):
310 (a) The code assumed that strncpy() returns a negative number on buffer
311 overflow, which isn't the case. Replaced with Exim's string_format()
313 (b) There were several signed/unsigned issues. I just did the minimum
314 hacking in of casts. There is scope for a larger refactoring.
315 (c) The code used strcasecmp() which is not a standard C function.
316 Replaced with Exim's strcmpic() function.
317 (d) The code set only $1; it now sets $auth1 as well.
318 (e) A simple test gave the error "authentication client didn't specify
319 service in request". It would seem that Dovecot has changed its
320 interface. Fortunately there's a specification; I followed it and
321 changed what the client sends and it appears to be working now.
323 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
326 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
327 address A is aliased to B and C, where B exists and C does not. Without
328 -v the output is "A verified" because verification stops after a
329 successful redirection if more than one address is generated. However,
330 with -v the child addresses are also verified. Exim was outputting "A
331 failed to verify" and then showing the successful verification for C,
332 with its parentage. It now outputs "B failed to verify", showing B's
333 parentage before showing the successful verification of C.
335 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
336 look up a TXT record in a specific list after matching in a combined
339 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
340 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
341 they consult the DNS. I had assumed they would set it the way they
342 wanted; and indeed my experiments on Linux seem to show that in some
343 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
344 To be on the safe side, however, I have now made the interface to
345 host_find_byname() similar to host_find_bydns(), with an argument
346 containing the DNS resolver options. The host_find_byname() function now
347 sets these options at its start, just as host_find_bydns() does. The smtp
348 transport options dns_qualify_single and dns_search_parents are passed to
349 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
350 of host_find_byname() use the default settings of RES_DEFNAMES
351 (qualify_single) but not RES_DNSRCH (search_parents).
353 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
354 spool_read_header() do less string testing, by means of a preliminary
355 switch on the second character of optional "-foo" lines. (This is
356 overdue, caused by the large number of possibilities that now exist.
357 Originally there were few.) While I was there, I also converted the
358 str(n)cmp tests so they don't re-test the leading "-" and the first
359 character, in the hope this might squeeze out yet more improvement.
361 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
362 flag allowing group syntax was set by the header_syntax check but not
363 turned off, possible causing trouble later; (2) The flag was not being
364 set at all for the header_verify test, causing "group"-style headers to
365 be rejected. I have now set it in this case, and also caused header_
366 verify to ignore an empty address taken from a group. While doing this, I
367 came across some other cases where the code for allowing group syntax
368 while scanning a header line wasn't quite right (mostly, not resetting
369 the flag correctly in the right place). These bugs could have caused
370 trouble for malformed header lines. I hope it is now all correct.
372 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
373 with the "reply" argument non-NULL. The code, however (which originally
374 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
375 but it didn't always do it. This confused somebody who was copying the
376 code for some other use. I have removed all the tests.
378 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
379 feature that was used to support insecure browsers during the U.S. crypto
380 embargo. It requires special client support, and Exim is probably the
381 only MTA that supported it -- and would never use it because real RSA is
382 always available. This code has been removed, because it had the bad
383 effect of slowing Exim down by computing (never used) parameters for the
384 RSA_EXPORT functionality.
386 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
387 authenticator to fail if there's a tab character in the incoming data
388 (there should never be unless someone is messing about, as it's supposed
389 to be base64-encoded). Also added, on Timo's advice, the "secured" option
390 if the connection is using TLS or if the remote IP is the same as the
391 local IP, and the "valid-client-cert option" if a client certificate has
394 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
395 authenticators. This can be used for authorization after authentication
396 succeeds. (In the case of plaintext, it servers for both authentication
399 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
400 if any retry times were supplied.
402 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
403 connection, where there is no client IP address to check. In this
404 situation, the verify now always succeeds.
406 PH/25 Applied John Jetmore's -Mset patch.
408 PH/26 Added -bem to be like -Mset, but loading a message from a file.
410 PH/27 In a string expansion for a processed (not raw) header when multiple
411 headers of the same name were present, leading whitespace was being
412 removed from all of them, but trailing whitespace was being removed only
413 from the last one. Now trailing whitespace is removed from each header
414 before concatenation. Completely empty headers in a concatenation (as
417 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
418 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
420 PH/29 [Removed. This was a change that I later backed out, and forgot to
421 correct the ChangeLog entry (that I had efficiently created) before
422 committing the later change.]
424 PH/30 Exim was sometimes attempting to deliver messages that had suffered
425 address errors (4xx response to RCPT) over the same connection as other
426 messages routed to the same hosts. Such deliveries are always "forced",
427 so retry times are not inspected. This resulted in far too many retries
428 for the affected addresses. The effect occurred only when there were more
429 hosts than the hosts_max_try setting in the smtp transport when it had
430 the 4xx errors. Those hosts that it had tried were not added to the list
431 of hosts for which the message was waiting, so if all were tried, there
432 was no problem. Two fixes have been applied:
434 (i) If there are any address or message errors in an SMTP delivery, none
435 of the hosts (tried or untried) are now added to the list of hosts
436 for which the message is waiting, so the message should not be a
437 candidate for sending over the same connection that was used for a
438 successful delivery of some other message. This seems entirely
439 reasonable: after all the message is NOT "waiting for some host".
440 This is so "obvious" that I'm not sure why it wasn't done
441 previously. Hope I haven't missed anything, but it can't do any
442 harm, as the worst effect is to miss an optimization.
444 (ii) If, despite (i), such a delivery is accidentally attempted, the
445 routing retry time is respected, so at least it doesn't keep
446 hammering the server.
448 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
449 in ${readsocket because some servers need this prod.
451 PH/32 Added some extra debug output when updating a wait-xxx database.
453 PH/33 The hint "could be header name not terminated by colon", which has been
454 given for certain expansion errors for a long time, was not being given
455 for the ${if def:h_colon_omitted{... case.
457 PH/34 The spec says: "With one important exception, whenever a domain list is
458 being scanned, $domain contains the subject domain." There was at least
459 one case where this was not true.
461 PH/35 The error "getsockname() failed: connection reset by peer" was being
462 written to the panic log as well as the main log, but it isn't really
463 panic-worthy as it just means the connection died rather early on. I have
464 removed the panic log writing for the ECONNRESET error when getsockname()
467 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
468 runs only) independently of the message's sender address. This meant
469 that, if the 4xx error was in fact related to the sender, a different
470 message to the same recipient with a different sender could confuse
471 things. In particualar, this can happen when sending to a greylisting
472 server, but other circumstances could also provoke similar problems.
473 I have changed the default so that the retry time for these errors is now
474 based a combination of the sender and recipient addresses. This change
475 can be overridden by setting address_retry_include_sender=false in the
478 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
479 remote server are returned as part of bounce messages. This was not
480 happening for LMTP over a pipe (the lmtp transport), but now it is the
481 same for both kinds of LMTP.
483 PH/38 Despite being documented as not happening, Exim was rewriting addresses
484 in header lines that were in fact CNAMEs. This is no longer the case.
486 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
487 and queue runs started by the daemon processed all messages. This has
488 been fixed so that -R and -S can now usefully be given with -q<time>.
490 PH/40 Import PCRE release 6.7 (fixes some bugs).
492 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
494 PH/42 Give an error if -q is specified more than once.
496 PH/43 Renamed the variables $interface_address and $interface_port as
497 $received_ip_address and $received_port, to make it clear that these
498 values apply to message reception, and not to the outgoing interface when
499 a message is delivered. (The old names remain recognized, of course.)
501 PH/44 There was no timeout on the connect() call when using a Unix domain
502 socket in the ${readsocket expansion. There now is.
504 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
505 be meaningful with "accept".
507 SC/01 Eximstats V1.43
508 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
510 SC/02 Eximstats V1.44
511 Use a glob alias rather than an array ref in the generated
512 parser. This improves both readability and performance.
514 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
515 Collect SpamAssassin and rejection statistics.
516 Don't display local sender or destination tables unless
517 there is data to show.
518 Added average volumes into the top table text output.
520 SC/04 Eximstats V1.46
521 Collect data on the number of addresses (recipients)
522 as well as the number of messages.
524 SC/05 Eximstats V1.47
525 Added 'Message too big' to the list of mail rejection
526 reasons (thanks to Marco Gaiarin).
528 SC/06 Eximstats V1.48
529 Mainlog lines which have GMT offsets and are too short to
530 have a flag are now skipped.
532 SC/07 Eximstats V1.49 (Alain Williams)
533 Added the -emptyok flag.
535 SC/08 Eximstats V1.50
536 Fixes for obtaining the IP address from reject messages.
538 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
539 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
540 whitesspace changes from 4.64-PH/27
542 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
545 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
546 are found, allow negative numbers in numeric criteria)
548 JJ/06 exipick.20061117.2, added new $message_body_missing variable
550 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
551 to match changes made in 4.64-PH/43
553 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
555 PH/47 Put in an explicit test for a DNS lookup of an address record where the
556 "domain" is actually an IP address, and force a failure. This locks out
557 those revolvers/nameservers that support "A-for-A" lookups, in
558 contravention of the specifications.
560 PH/48 When a host name was looked up from an IP address, and the subsequent
561 forward lookup of the name timed out, the host name was left in
562 $sender_host_name, contrary to the specification.
564 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
565 restricted to single-key lookups, Exim was not diagnosing an error if
566 * or *@ was used with a query-style lookup.
568 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
570 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
571 long ago, but noone interested enough thought of it. Let's just say that
572 the "1.1" means that there are some new functions that weren't there at
573 some point in the past.
575 PH/51 Error processing for expansion failure of helo_data from an smtp
576 transport during callout processing was broken.
578 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
579 tested/used via the -bh/-bhc/-bs options.
581 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
582 bug, fixed in subsequent PCRE releases).
584 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
585 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
587 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
593 SC/01 Use a glob alias rather than an array ref in eximstats generated
594 parser. This improves both readability and performance.
596 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
597 Don't display local sender or destination tables in eximstats unless
598 there is data to show.
599 Added average volumes into the eximstats top table text output.
601 SC/03 Collect data on the number of addresses (recipients) as well
602 as the number of messages in eximstats.
604 TF/01 Correct an error in the documentation for the redirect router. Exim
605 does (usually) call initgroups() when daemonizing.
607 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
608 with consistent privilege compared to when running as a daemon.
610 TF/03 Note in the spec that $authenticated_id is not set for local
611 submissions from trusted users.
613 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
614 Thanks to Dean Brooks <dean@iglou.com> for the patch.
616 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
617 by adding some example configuration directives to the default
618 configuration file. A little bit of work is required to uncomment the
619 directives and define how usernames and passwords are checked, but
620 there is now a framework to start from.
622 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
623 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
624 without this. I don't know how relevant this is to other LDAP libraries.
626 PH/02 Add the verb name to the "unknown ACL verb" error.
628 PH/03 Magnus Holmgren's patch for filter_prepend_home.
630 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
632 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
633 directory not expanded when it should be if an expanded home directory
634 was set for the address (which is overridden by the transport).
636 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
639 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
640 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
641 because it is too late at that time, and has no effect.
643 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
644 security issue with \' (bugzilla #107). I could not use the
645 PQescapeStringConn() function, because it needs a PGconn value as one of
648 PH/08 When testing addresses using -bt, indicate those final addresses that
649 are duplicates that would not cause an additional delivery. At least one
650 person was confused, thinking that -bt output corresponded to deliveries.
651 (Suppressing duplicates isn't a good idea as you lose the information
652 about possibly different redirections that led to the duplicates.)
654 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
655 systems where poll() doesn't work, in particular OS X.
657 PH/10 Added more information to debugging output for retry time not reached.
659 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
660 operations in malware.c.
662 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
665 PH/13 If write_rejectlog was set false when logging was sent to syslog with
666 syslog_duplication set false, log lines that would normally be written
667 both the the main log and to the reject log were not written to syslog at
670 PH/14 In the default configuration, change the use of "message" in ACL warn
671 statements to "add_header".
673 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
674 not followed by a command (e.g. "seen endif").
676 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
677 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
680 PH/17 Added extra conditions to the default value of delay_warning_condition
684 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
685 { match{$h_precedence:}{(?i)bulk|list|junk} } \
686 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
689 The Auto-Submitted: and various List- headers are standardised, whereas I
690 don't think Precedence: ever was.
692 PH/18 Refactored debugging code in route_finduser() to show more information,
693 in particular, the error code if getpwnam() issues one.
695 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
696 This is apparently needed in addition to the PH/07 change above to avoid
697 any possible encoding problems.
699 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
700 but not after initializing Perl.
702 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
703 output them only if debugging. By default they are written stderr,
704 apparently, which is not desirable.
706 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
709 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
712 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
714 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
715 authenticated or an ident call has been made. Suppress the default
716 values for $authenticated_id and $authenticated_sender (but permit -oMai
717 and -oMas) when testing with -bh.
719 PH/24 Re-jigged the order of the tests in the default configuration so that the
720 tests for valid domains and recipients precede the DNS black list and CSA
721 tests, on the grounds that those ones are more expensive.
723 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
724 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
725 This bug exists in every version of Exim that I still have, right back to
728 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
729 However, an attempt to turn on case-sensitivity in a regex key by
730 including (?-i) didn't work because the subject string was already
731 lowercased, and the effects were non-intuitive. It turns out that a
732 one-line patch can be used to allow (?-i) to work as expected.
738 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
739 other effects) broke the use of negated acl sub-conditions.
741 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
744 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
745 "Deny" causes Exim to reject the incoming connection with a 554 error.
746 Unfortunately, if there is a major crisis, such as a disk failure,
747 tcp-wrappers gives "deny", whereas what one would like would be some
748 kind of temporary error. A kludge has been added to help with this.
749 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
750 554 error is used if errno is still zero or contains ENOENT (which occurs
751 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
754 PH/03 Add -lutil to the default FreeBSD LIBS setting.
756 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
757 errors. Otherwise a message that provokes a temporary error (when other
758 messages do not) can cause a whole host to time out.
760 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
761 addresses were routed directly to files or pipes from a redirect router.
762 File deliveries just didn't batch; pipe deliveries might have suffered
765 PH/06 A failure to get a lock for a hints database would erroneously always say
766 "Failed to get write lock", even when it was really a read lock.
768 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
769 of 0600. This has been changed to use the value of the lockfile_mode
770 option (which defaults to 0600).
772 PH/08 Applied small patch from the Sieve maintainer.
774 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
775 folder from quota calculations, a direct delivery into this folder messed
776 up the contents of the maildirsize file. This was because the regex was
777 used only to exclude .Trash (or whatever) when the size of the mailbox
778 was calculated. There was no check that a delivery was happening into an
779 excluded directory. This bug has been fixed by ignoring all quota
780 processing for deliveries into excluded directories.
782 PH/10 Added the maildirfolder_create_regex option to appendfile.
788 PH/01 The code for finding all the local interface addresses on a FreeBSD
789 system running IPv6 was broken. This may well have applied to all BSD
790 systems, as well as to others that have similar system calls. The broken
791 code found IPv4 interfaces correctly, but gave incorrect values for the
792 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
793 that it would not match correctly against @[] and not recognize the IPv6
796 PH/02 The ipliteral router was not recognizing addresses of the form user@
797 [ipv6:....] because it didn't know about the "ipv6:" prefix.
799 PH/03 Added disable_ipv6.
801 PH/04 Changed $reply_address to use the raw form of the headers instead of the
802 decoded form, because it is most often used to construct To: headers
803 lines in autoreplies, and the decoded form may well be syntactically
804 invalid. However, $reply_address has leading white space removed, and all
805 newlines turned into spaces so that the autoreply transport does not
808 PH/05 If group was specified without a user on a router, and no group or user
809 was specified on a transport, the group from the router was ignored.
811 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
812 for visible compile-time settings that can be used to change these
813 numbers, for those that want even more. Backwards compatibility with old
814 spool files has been maintained. However, going back to a previous Exim
815 release will lost any variables that are in spool files.
817 PH/07 Two small changes when running in the test harness: increase delay when
818 passing a TCP/IP connection to a new process, in case the original
819 process has to generate a bounce, and remove special handling of
820 127.0.0.2 (sic), which is no longer necessary.
822 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
823 be the same on different OS.
825 PH/09 Moved a debug statement in filter processing to avoid a race problem when
828 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
829 whether --show-vars was specified or not
831 JJ/02 exipick: Added support for new ACL variable spool format introduced
834 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
835 syntactically invalid From: or Reply-to: line, and a filter used this to
836 generate an autoreply, and therefore failed to obtain an address for the
837 autoreply, Exim could try to deliver to a non-existent relative file
838 name, causing unrelated and misleading errors. What now happens is that
839 it logs this as a hard delivery error, but does not attempt to create a
842 PH/11 The exinext utility has a -C option for testing purposes, but although
843 the given file was scanned by exinext itself; it wasn't being passed on
844 when Exim was called.
846 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
847 an end-of-file indication when reading a command response.
849 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
850 compiled. In many other places in Exim, IPv6 addresses are always
851 recognized, so I have changed this. It also means that IPv4 domain
852 literals of the form [IPV4:n.n.n.n] are now always recognized.
854 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
855 used if the router is not running as root, for example, when verifying at
856 ACL time, or when using -bh. The debugging output from this situation was
857 non-existent - all you got was a failure to exec. I have made two
860 (a) Failures to set uid/gid, the current directory, or a process leader
861 in a subprocess such as that created by queryprogram now generate
862 suitable debugging ouput when -d is set.
864 (b) The queryprogram router detects when it is not running as root,
865 outputs suitable debugging information if -d is set, and then runs
866 the subprocess without attempting to change uid/gid.
868 PH/15 Minor change to Makefile for building test_host (undocumented testing
871 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
872 additional section of a DNS packet that returns MX or SRV records.
873 Instead, it always explicitly searches for A/AAAA records. This avoids
874 major problems that occur when a DNS server includes only records of one
875 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
876 fixed another bug: if SRV records were looked up and the corresponding
877 address records were *not* found in the additional section, the port
878 values from the SRV records were lost.
880 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
881 using the correct key (the original address) when searching the retry
882 rules in order to find which one to use for generating the retry hint.
884 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
885 adding the default one. Similarly, if it contains a Reply-To: header, the
886 errors_reply_to option, if set, is not used.
888 PH/19 When calculating a retry time, Exim used to measure the "time since
889 failure" by looking at the "first failed" field in the retry record. Now
890 it does not use this if it is later than than the arrival time of the
891 message. Instead it uses the arrival time. This makes for better
892 behaviour in cases where some deliveries succeed, thus re-setting the
893 "first failed" field. An example is a quota failure for a huge message
894 when small messages continue to be delivered. Without this change, the
895 "time since failure" will always be short, possible causing more frequent
896 delivery attempts for the huge message than are intended.
897 [Note: This change was subsequently modified - see PH/04 for 4.62.]
899 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
900 $1, $2, $3) because the numerical variables can be reset during some
901 expansion items (e.g. "match"), thereby losing the authentication data.
903 PH/21 Make -bV show the size of off_t variables so that the test suite can
904 decide whether to run tests for quotas > 2G.
906 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
907 mailbox_size, and mailbox_filecount in the appendfile transport. If a
908 filecount value is greater than 2G or if a quota value is greater than 2G
909 on a system where the size of off_t is not greater than 4, a panic error
912 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
913 never match. The debug and -bh output now contains an explicit error
914 message indicating a malformed IPv4 address or mask.
916 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
917 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
920 PH/25 Do not write to syslog when running in the test harness. The only
921 occasion when this arises is a failure to open the main or panic logs
922 (for which there is an explicit test).
924 PH/26 Added the /no_tell option to "control=freeze".
926 PH/27 If a host name lookup failed very early in a connection, for example, if
927 the IP address matched host_lookup and the reverse lookup yielded a name
928 that did not have a forward lookup, an error message of the form "no IP
929 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
930 could be logged. Now it outputs the IP address instead of "NULL".
932 PH/28 An enabling patch from MH: add new function child_open_exim2() which
933 allows the sender and the authenticated sender to be set when
934 submitting a message from within Exim. Since child_open_exim() is
935 documented for local_scan(), the new function should be too.
937 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
938 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
939 results in an empty string is now treated as unset.
941 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
943 PH/31 Added sender_verify_fail logging option.
945 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
946 needed by RFC 822 but not by RFC 2822 was commented out. I have now
947 tidied the source and removed it altogether.
949 PH/33 When a queue run was abandoned because the load average was too high, a
950 log line was always written; now it is written only if the queue_run log
951 selector is set. In addition, the log line for abandonment now contains
952 information about the queue run such as the pid. This is always present
953 in "start" and "stop" lines but was omitted from the "abandon" line.
955 PH/34 Omit spaces between a header name and the colon in the error message that
956 is given when verify = headers_syntax fails (if there are lots of them,
957 the message gets confusing).
959 PH/35 Change the default for dns_check_names_pattern to allow slashes within
960 names, as there are now some PTR records that contain slashes. This check
961 is only to protect against broken name servers that fall over on strange
962 characters, so the fact that it applies to all lookups doesn't matter.
964 PH/36 Now that the new test suite is complete, we can remove some of the
965 special code in Exim that was needed for the old test suite. For example,
966 sorting DNS records because real resolvers return them in an arbitrary
967 order. The new test suite's fake resolver always returns records in the
970 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
971 bounces) except when queue_only is set, to avoid logging races between
972 the different processes.
974 PH/38 Panic-die if .include specifies a non-absolute path.
976 PH/39 A tweak to the "H" retry rule from its user.
978 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
979 a label. They prevented compilation on older perls.
981 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
982 a warning to be raised on newish perls.
984 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
985 on queue. Changes to match documented behaviour of showing count of
986 messages matching specified criteria.
988 PH/40 Changed the default ident timeout from 30s to 5s.
990 PH/41 Added support for the use of login_cap features, on those BSD systems
991 that have them, for controlling the resources used by pipe deliveries.
993 PH/42 The content-scanning code uses fopen() to create files in which to put
994 message data. Previously it was not paying any attention to the mode of
995 the files. Exim runs with umask(0) because the rest of the code creates
996 files with open(), and sets the required mode explicitly. Thus, these
997 files were ending up world-writeable. This was not a big issue, because,
998 being within the spool directory, they were not world-accessible. I have
999 created a function called modefopen, which takes an additional mode
1000 argument. It sets umask(777), creates the file, chmods it to the required
1001 mode, then resets the umask. All the relevant calls to fopen() in the
1002 content scanning code have been changed to use this function.
1004 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1005 to 24 hours. This avoids potential overflow problems when processing G
1006 and H retry rules. I suspect nobody ever tinkers with this value.
1008 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1010 PH/45 When the plaintext authenticator is running as a client, the server's
1011 challenges are checked to ensure they are valid base64 strings. By
1012 default, the authentication attempt is cancelled if an invalid string is
1013 received. Setting client_ignore_invalid_base64 true ignores these errors.
1014 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1015 they are received. Thus, the responses can be made to depend on the
1016 challenges. If an invalid string is ignored, an empty string is placed in
1019 PH/46 Messages that are created by the autoreply transport now contains a
1020 References: header, in accordance with RFCs 2822 and 3834.
1022 PH/47 Added authenticated_sender_force to the smtp transport.
1024 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1026 PH/49 Installed latest patch from the Sieve maintainer.
1028 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1029 was also set, the appendfile transport was unnecessarily scanning a
1030 directory of message files (e.g. for maildir delivery) to find the count
1031 of files (along with the size), even though it did not need this
1032 information. It now does the scan only if it needs to find either the
1033 size of the count of files.
1035 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1037 PH/52 Two bugs concerned with error handling when the smtp transport is
1040 (i) Exim was not creating retry information for temporary errors given
1041 for individual recipients after the DATA command when the smtp transport
1042 was used in LMTP mode. This meant that they could be retried too
1043 frequently, and not timed out correctly.
1045 (ii) Exim was setting the flag that allows error details to be returned
1046 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1047 recipients that were returned after the DATA command.
1049 PH/53 This is related to PH/52, but is more general: for any failing address,
1050 when detailed error information was permitted to be returned to the
1051 sender, but the error was temporary, then after the final timeout, only
1052 "retry timeout exceeded" was returned. Now it returns the full error as
1053 well as "retry timeout exceeded".
1055 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1056 do this, and (what is worse) MTAs that accept it.
1058 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1059 will now be deprecated.
1061 PH/56 New os.c-cygwin from the Cygwin maintainer.
1063 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1064 formats (previously only available in exim formats via -bpr, -bpru,
1065 and -bpra. Now also available in native and exiqgrep formats)
1067 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1068 with very large, slow to parse queues
1070 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1072 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1074 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1075 responses to authentication challenges, though it was showing the
1076 challenges; (ii) I've removed the CR characters from the debug output for
1079 PH/58 Allow for the insertion of a newline as well as a space when a string
1080 is turned into more than one encoded-word during RFC 2047 encoding. The
1081 Sieve code now uses this.
1083 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1084 data_4xx, lost_connection, tls_required.
1086 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1087 message was being sent as an SMTP response.
1089 PH/61 Add -l and -k options to exicyclog.
1091 PH/62 When verifying, if an address was redirected to one new address, so that
1092 verification continued, and the new address failed or deferred after
1093 having set something in $address_data, the value of $address_data was not
1094 passed back to the ACL. This was different to the case when no
1095 redirection occurred. The value is now passed back in both cases.
1097 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1098 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1099 use login_cap.h, so on its own it isn't the distinguishing feature. The
1100 new name refers directly to the setclassresources() function.
1102 PH/65 Added configuration files for NetBSD3.
1104 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1106 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1107 is preferred over IPv4.
1109 PH/68 The bounce_return_message and bounce_return_body options were not being
1110 honoured for bounces generated during the reception of non-SMTP messages.
1111 In particular, this applied to messages rejected by the ACL. This bug has
1112 been fixed. However, if bounce_return_message is true and bounce_return_
1113 body is false, the headers that are returned for a non-SMTP message
1114 include only those that have been read before the error was detected.
1115 (In the case of an ACL rejection, they have all been read.)
1117 PH/69 The HTML version of the specification is now built in a directory called
1118 spec_html instead of spec.html, because the latter looks like a path with
1119 a MIME-type, and this confuses some software.
1121 PH/70 Catch two compiler warnings in sieve.c.
1123 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1124 function verify_get_ident() calls ip_connect() to connect a socket, but
1125 if the "connect()" function timed out, ip_connect() used to close the
1126 socket. However, verify_get_ident() also closes the socket later, and in
1127 between Exim writes to the log, which may get opened at this point. When
1128 the socket was closed in ip_connect(), the log could get the same file
1129 descriptor number as the socket. This naturally causes chaos. The fix is
1130 not to close the socket in ip_connect(); the socket should be closed by
1131 the function that creates it. There was only one place in the code where
1132 this was missing, in the iplookup router, which I don't think anybody now
1133 uses, but I've fixed it anyway.
1135 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1136 well as to direct DNS lookups. Otherwise the handling of names in host
1137 lists is inconsistent and therefore confusing.
1143 PH/01 Two changes to the default runtime configuration:
1145 (1) Move the checks for relay_from_hosts and authenticated clients from
1146 after to before the (commented out) DNS black list checks.
1148 (2) Add control=submission to the relay_from_hosts and authenticated
1149 clients checks, on the grounds that messages accepted by these
1150 statements are most likely to be submissions.
1152 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1154 (1) Generate an error if the third argument for the ${prvs expansion is
1157 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1160 (3) Reset the variables that are obtained from the first argument of
1161 ${prvscheck and used in the second argument before leaving the code,
1162 because their memory is reclaimed, so using them afterwards may do
1165 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1166 one (it's much easier than Tom thought :-).
1168 (5) Because of (4), we can now allow for the use of $prvscheck_result
1169 inside the third argument.
1171 PH/03 For some reason, the default setting of PATH when running a command from
1172 a pipe transport was just "/usr/bin". I have changed it to
1175 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1176 anything to be listed in the output from -bV.
1178 PH/05 When a filter generated an autoreply, the entire To: header line was
1179 quoted in the delivery log line, like this:
1181 => >A.N.Other <ano@some.domain> <original@ddress> ...
1183 This has been changed so that it extracts the operative address. There
1184 may be more than one such address. If so, they are comma-separated, like
1187 => >ano@some.domain,ona@other.domain <original@ddress> ...
1189 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1190 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1191 not being looked up in the rDNS to get a host name, Exim was showing the
1192 IP address twice in Received: lines, even though the IP addresses were
1193 identical. For example:
1195 Received: from [1.2.3.4] (helo=[1.2.3.4])
1197 However, if the real host name was known, it was omitting the HELO data
1198 if it matched the actual IP address. This has been tidied up so that it
1199 doesn't show the same IP address twice.
1201 PH/07 When both +timestamp and +memory debugging was on, the value given by
1202 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1203 called by the debug printing, thereby overwriting the timestamp buffer.
1204 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1206 PH/08 When the original message was included in an autoreply transport, it
1207 always said "this is a copy of the message, including all the headers",
1208 even if body_only or headers_only was set. It now gives an appropriate
1211 PH/09 Applied a patch from the Sieve maintainer which:
1213 o fixes some comments
1214 o adds the (disabled) notify extension core
1215 o adds some debug output for the result of if/elsif tests
1216 o points to the current vacation draft in the documentation
1217 and documents the missing references header update
1221 o fixes a bug in processing the envelope test (when testing
1222 multiple envelope elements, the last element determinted the
1225 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1226 Electronic Mail") by including:
1228 Auto-submitted: auto-generated
1230 in the messages that it generates (bounce messages and others, such as
1231 warnings). In the case of bounce messages for non-SMTP mesages, there was
1232 also a typo: it was using "Auto_submitted" (underscore instead of
1233 hyphen). Since every message generated by Exim is necessarily in response
1234 to another message, thes have all been changed to:
1236 Auto-Submitted: auto-replied
1238 in accordance with these statements in the RFC:
1240 The auto-replied keyword:
1242 - SHOULD be used on messages sent in direct response to another
1243 message by an automatic process,
1245 - MUST NOT be used on manually-generated messages,
1247 - MAY be used on Delivery Status Notifications (DSNs) and Message
1248 Disposition Notifications (MDNs),
1250 - MUST NOT be used on messages generated by automatic or periodic
1251 processes, except for messages which are automatic responses to
1254 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1255 to the default Received: header definition.
1257 PH/12 Added log selector acl_warn_skipped (default on).
1259 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1260 variables because (a) they are in the wrong storage pool and (b) even if
1261 they were copied, it wouldn't work properly because of the caching.
1263 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1264 checking when decoding. Apparently there are clients that generate
1265 overlong encoded strings. Why am I not surprised?
1267 PH/15 If the first argument of "${if match_address" was not empty, but did not
1268 contain an "@" character, Exim crashed. Now it writes a panic log message
1269 and treats the condition as false.
1271 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1273 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1274 extension "envelope-auth". The code is finished and in agreement with
1275 other implementations, but there is no documentation so far and in fact,
1276 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1277 not changing the active code.
1279 Print executed "if" and "elsif" statements when debugging is used. This
1280 helps a great deal to understand what a filter does.
1282 Document more things not specified clearly in RFC3028. I had all this
1283 sorted out, when out of a sudden new issues came to my mind. Oops."
1285 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1288 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1289 canonical form (as documented). However, after a host name lookup from
1290 the IP address, check_host() was doing a simple string comparison with
1291 addresses acquired from the DNS when checking that the found name did
1292 have the original IP as one of its addresses. Since any found IPv6
1293 addresses are likely to be in abbreviated form, the comparison could
1294 fail. Luckily, there already exists a function for doing the comparison
1295 by converting both addresses to binary, so now that is used instead of
1296 the text comparison.
1298 PH/20 There was another similar case to PH/19, when a complete host name was
1299 given in a host list; looking up its IP address could give an abbreviated
1300 form, whereas the current host's name might or might not be abbreviated.
1301 The same fix has been applied.
1307 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1308 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1311 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1312 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1314 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1316 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1317 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1318 Also, for "Auto-Submitted", treat anything other than "no" as
1319 non-personal, in accordance with RFC 3834. (Previously it treated
1320 anything starting "auto-" as non-personal.)
1322 TF/01 The control=submission/name=... option had a problem with syntax
1323 errors if the name included a slash character. The /name= option
1324 now slurps the rest of the string, so it can include any characters
1325 but it must come last in the list of options (after /sender_retain
1328 PH/05 Some modifications to the interface to the fake nameserver for the new
1336 TK/01 Added the "success_on_redirect" address verification option. See
1337 NewStuff for rationale and an example.
1339 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1341 PH/02 Patch to exigrep to allow it to work on syslog lines.
1343 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1344 fread() to skip over the body file's header line, because in Cygwin the
1345 header line is locked and is inaccessible.
1347 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1348 co-exist for some time) to make it clear that it is the Exim ID that is
1349 referenced, not the Message-ID: header line.
1351 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1352 string_format() function, because snprintf() does not exist on all
1355 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1356 ${stat: expansion item.
1358 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1359 protocol synchronization error", to keep the pedants happy.
1361 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1362 well as for IRIX systems, when gcc is being used. See the host.c source
1365 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1367 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1370 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1371 transport and to the smtp transport in LMTP mode.
1373 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1375 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1376 run a filter in a subprocess. This could lead to confusion in subsequent
1377 lookups in the parent process. There should also be a search_tidyup() at
1378 the end of the subprocess.
1380 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1381 only if the host matched helo_try_verify_hosts, which caused the
1382 verification to occur when the EHLO/HELO command was issued. The ACL just
1383 tested the remembered result. Now, if a previous verification attempt has
1384 not happened, "verify = helo" does it there and then.
1386 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1388 TK/03 Fix log output including CR from clamd.
1390 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1391 exist provoked a memory error which could cause a segfault.
1393 PH/15 Installed PCRE 6.2
1395 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1397 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1398 of the problem. Specifically, suggested +O2 rather than +O1 for the
1401 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1403 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1404 record, the port was not correctly logged when the outgoing_port log
1405 selector was set (it logged the transort's default port).
1407 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1408 fallback_hosts, and "hosts" in the smtp transport.
1410 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1411 host errors such as "Connection refused".
1413 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1414 authentication with radiusclient 0.4.9:
1416 - Error returned from rc_read_config was caught wrongly
1417 - Username/password not passed on to radius server due to wrong length.
1419 The presumption is that some radiusclient API changes for 4.51/PH/17
1420 were not taken care of correctly. The code is still untested by me (my
1421 Linux distribution still has 0.3.2 of radiusclient), but it was
1422 contributed by a Radius user.
1424 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1425 expanding the "port" option of the smtp transport.
1427 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1428 while reading a MIME header. Thanks to Tom Hughes for a patch.
1430 PH/24 Include config.h inside local_scan.h so that configuration settings are
1433 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1434 that in an ACL for RCPT (for example), you can examine exactly what was
1437 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1438 commands, but it was not correctly comparing the address with the actual
1439 client host address. Thus, it would show the EHLO address in Received:
1440 header lines when this was not necessary.
1442 PH/27 Added the % operator to ${eval:}.
1444 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1445 it should be ignoring failures (because with -C, for example, it has lost
1446 privilege). It wasn't ignoring creation failures other than "already
1449 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1452 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1453 because an input line was too long, either on its own, or by virtue of
1454 too many continuations, the temporary file was not being removed, and the
1455 return code was incorrect.
1457 PH/31 Missing "BOOL" in function definition in filtertest.c.
1459 PH/32 Applied Sieve patches from the maintainer.
1461 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1463 PH/33 Added "verify = not_blind".
1465 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1466 Local/Makefile (with some defaults set). These are used in built scripts
1467 such as exicyclog, but they have never been used in the exim_install
1468 script (though there are many overriding facilities there). I have
1469 arranged that the exim_install script now takes note of these two
1472 PH/35 Installed configuration files for Dragonfly.
1474 PH/36 When a locally submitted message by a trusted user did not contain a
1475 From: header, and the sender address was obtained from -f or from an SMTP
1476 MAIL command, and the trusted user did not use -F to supply a sender
1477 name, $originator_name was incorrectly used when constructing a From:
1478 header. Furthermore, $originator_name was used for submission mode
1479 messages from external hosts without From: headers in a similar way,
1480 which is clearly wrong.
1482 PH/37 Added control=suppress_local_fixups.
1484 PH/38 When log_selector = +received_sender was set, and the addition of the
1485 sender made the log line's construction buffer exactly full, or one byte
1486 less than full, an overflow happened when the terminating "\n" was
1489 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1490 when the result of a list match is failure because a DNS lookup failed.
1492 PH/40 RM_COMMAND is now used in the building process.
1494 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1495 the "build-* directories that it finds.
1497 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1498 address in a domain literal was a prefix of an interface address.
1500 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1501 when verifying a sender address, unless rewrite_headers is false.
1503 PH/44 Wrote a long comment about why errors_to addresses are verified as
1504 recipients, not senders.
1506 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1507 the ratelimit ACL was added.
1509 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1511 PH/46 Added extra information about PostgreSQL errors to the error string.
1513 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1514 avoiding the need to install special zones in a real server. This is
1515 backwards compatible; if it can't find the fake resolver, it drops back.
1516 Thus, both old and new test suites can be run.
1518 TF/02 Added util/ratelimit.pl
1520 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1521 clock is set back in time.
1523 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1524 Candler <B.Candler@pobox.com>.
1526 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1527 OK for addresses that are the result of redirections.
1529 PH/48 A number of further additions for the benefit of the new test suite,
1530 including a fake gethostbyname() that interfaces to the fake DNS resolver
1533 TF/06 The fix for widen_domains has also been applied to qualify_single and
1534 search_parents which are the other dnslookup options that can cause
1537 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1540 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1542 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1543 service" if it encountered an MX record with an empty target hostname.
1544 The message is now "an MX or SRV record indicated no SMTP service".
1546 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1547 if the DNS of the sending site is misconfigured. This is quite a
1548 common situation. This change restores the behaviour of treating a
1549 helo verification defer as a failure.
1551 PH/51 If self=fail was set on a router, the bounce message did not include the
1552 actual error message.
1558 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1560 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1561 command itself ended in error, the underlying message about the transport
1562 filter timeout was being overwritten with the pipe command error. Now the
1563 underlying error message should be appended to the second error message.
1565 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1567 PH/02 Reduce the amount of output that "make" produces by default. Full output
1568 can still be requested.
1570 PH/03 The warning log line about a condition test deferring for a "warn" verb
1571 was being output only once per connection, rather than after each
1572 occurrence (because it was using the same function as for successful
1573 "warn" verbs). This seems wrong, so I have changed it.
1575 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1576 it should not have, which might have caused a crash in the right
1577 circumstances, but probably never did.
1579 PH/04 Installed a modified version of Tony Finch's patch to make submission
1580 mode fix the return path as well as the Sender: header line, and to
1581 add a /name= option so that you can make the user's friendly name appear
1584 TF/03 Added the control = fakedefer ACL modifier.
1586 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1587 Mark Lowes for thorough testing.
1589 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1591 TK/03 Merged latest SRS patch from Miles Wilton.
1593 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1594 with the definition in sysexits.h (which is #included earlier).
1595 Fortunately, Exim does not actually use EX_OK. The code used to try to
1596 preserve the sysexits.h value, by assumimg that macro definitions were
1597 scanned for macro replacements. I have been disabused of this notion,
1598 so now the code just undefines EX_OK before #including unistd.h.
1600 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1601 in the smtp transport. When a block could not be written in a single
1602 write() function, the timeout was being re-applied to each part-write.
1603 This seems wrong - if the receiver was accepting one byte at a time it
1604 would take for ever. The timeout is now adjusted when this happens. It
1605 doesn't have to be particularly precise.
1607 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1608 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1610 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1611 without a domain if the check to <postmaster@domain> fails.
1613 SC/01 Eximstats: added -xls and the ability to specify output files
1614 (patch written by Frank Heydlauf).
1616 SC/02 Eximstats: use FileHandles for outputing results.
1618 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1620 SC/04 Eximstats: fixed display of large numbers with -nvr option
1622 SC/05 Eximstats: fixed merging of reports with empty tables.
1624 SC/06 Eximstats: added the -include_original_destination flag
1626 SC/07 Eximstats: removed tabs and trailing whitespace.
1628 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1630 TK/06 MBOX spool code: Add real "From " MBOX separator line
1631 so the .eml file is really in mbox format (even though
1632 most programs do not really care). Patch from Alex Miller.
1634 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1635 The latter is generated from $received_to and is only set if the
1636 message has one envelope recipient. SA can use these headers,
1637 obviously out-of-the-box. Patch from Alex Miller.
1639 PH/08 The ${def test on a variable was returning false if the variable's
1640 value was "0", contrary to what the specification has always said!
1641 The result should be true unless the variable is empty.
1643 PH/09 The syntax error of a character other than { following "${if
1644 def:variable_name" (after optional whitespace) was not being diagnosed.
1645 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1646 accidental colon was present, for example, could give incorrect results.
1648 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1649 result is used (not including appendfile, where other changes are about
1652 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1653 This involved changing a lot of size variables from int to off_t. It
1654 should work with maildirs and everything.
1656 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1657 spamd dying while we are connected to it.
1659 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1662 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1663 function for BDB 4.3.
1665 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1667 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1668 the sources for PCRE have been re-arranged and more files are now
1671 PH/15 The code I had for printing potentially long long variables in PH/11
1672 above was not the best (it lost precision). The length of off_t variables
1673 is now inspected at build time, and an appropriate printing format (%ld
1674 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1675 to be "long long int" or "long int". This is needed for the internal
1676 formatting function string_vformat().
1678 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1679 the configuration file to be ":syslog", then the script "guesses" where
1680 the logs files are, rather than using the compiled in default. In our
1681 case the guess is not the same as the compiled default, so the script
1682 suddenly stopped working when I started to use syslog. The patch checks
1683 to see if log_file_path is "". If so, it attempts to read it from exim
1684 with no configuration file to get the compiled in version, before it
1685 falls back to the previous guessing code."
1687 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1688 implementing BATV in an Exim configuration. See NewStuff for the gory
1691 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1692 Makefile that are specific to HP-UX.
1694 PH/18 If the "use_postmaster" option was set for a recipient callout together
1695 with the "random" option, the postmaster address was used as the MAIL
1696 FROM address for the random test, but not for the subsequent recipient
1697 test. It is now used for both.
1699 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1700 patch removes a few documentation additions to RFC 3028, because the
1701 latest draft now contains them. It adds the new en;ascii-case comparator
1702 and a new error check for 8bit text in MIME parts. Comparator and
1703 require names are now matched exactly. I enabled the subaddress
1704 extension, but it is not well tested yet (read: it works for me)."
1706 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1707 rework some of the code of TK/09 above to avoid the hardwired use of
1708 "%lld" and "long long". Replaced the call to snprintf() with a call to
1711 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1712 records point to non-existent hosts", "retry timeout exceeded", and
1713 "retry time not reached for any host after a long failure period".
1715 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1716 experimental DomainKeys support:
1718 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1719 (2) On an error such as an illegally used "control", the wrong name for
1720 the control was given.
1722 These problems did NOT occur unless DomainKeys support was compiled.
1724 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1726 PH/24 Added ${if match_ip condition.
1728 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1729 databases so that it will be absolutely obvious if a crash occurs in the
1730 DB library. This is a regular occurrence (often caused by mis-matched
1733 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1734 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1735 picked up on a user's system that detects such things. There doesn't seem
1736 to be a gcc warning option for this - only an attribute that has to be
1737 put on the function's prototype. It seems that in Fedora Core 4 they have
1738 set this on a number of new functions. No doubt there will be more in due
1741 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1742 specify a transport. However, if an address that was verified by such a
1743 router was the subject of a callout, Exim crashed because it tried to
1744 read the rcpt_include_affixes from the non-existent transport. Now it
1745 just assumes that the setting of that option is false. This bug was
1746 introduced by 4.51/PH/31.
1748 PH/28 Changed -d+all to exclude +memory, because that information is very
1749 rarely of interest, but it makes the output a lot bigger. People tend to
1750 do -d+all out of habit.
1752 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1753 code in os-type was giving problems when libc.so lives in lib64, like on
1756 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1757 aren't the modern standard, and it seems that some systems' include files
1758 don't always have them. Exim was already checking for some of the newer
1759 ones like T_AAAA, and defining it itself. I've added checks for all the
1760 record types that Exim uses.
1762 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1763 not automatically generating a new one, as it is supposed to. This
1764 prevented TLS from working. If the file did exist, but contained invalid
1765 data, a new version was generated, as expected. It was only the case of a
1766 non-existent file that was broken.
1768 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1769 with a change in libdomainkeys > 0.64.
1771 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1772 from DNS. If the selector record carries the flag, it now has
1773 precedence over the domain-wide flag.
1775 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1777 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1778 the use of a port name that isn't defined in /etc/services) occurred, the
1779 message was deferred as in a normal delivery, and thus remained on the
1780 spool, instead of being failed because of the mua_wrapper setting. This
1781 is now fixed, and I tidied up some of the mua_wrapper messages at the
1784 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1785 the messages in a hash of arrays rather than using individual hashes.
1786 This is a bit cleaner and results in dramatic memory savings, albeit
1787 at a slight CPU cost.
1789 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1790 as requested by Marc Sherman.
1792 SC/10 Eximstats: added histograms for user specified patterns as requested
1795 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1797 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1798 fopen() in the content-scanning modules that did not already have it.
1804 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1805 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1807 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1809 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1811 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1812 to test Sieve filters that use "vacation".
1814 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1815 that changes the way the GnuTLS parameters are stored in the cache file.
1816 The new format can be generated externally. For backward compatibility,
1817 if the data in the cache doesn't make sense, Exim assumes it has read an
1818 old-format file, and it generates new data and writes a new file. This
1819 means that you can't go back to an older release without removing the
1822 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1823 work if there are any delivery delays because "one_time" forces the
1824 parent to be marked "delivered", so its unseen clone is never tried
1825 again. For this reason, Exim now forbids the simultaneous setting of
1828 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1829 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1830 entry if you want to know the details. The fix, however, neglected to
1831 consider the case when local delivery batching is involved. The test for
1832 "previously delivered" was not happening when checking to see if an
1833 address could be batched with a previous (undelivered) one; under
1834 certain circumstances this could lead to multiple deliveries to the same
1837 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1838 in its include files, and this causes problems building Exim.
1840 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1841 header_syntax) but Exim was just ignoring anything given after a slash.
1842 In particular, this caused confusion with an attempt to use "verify =
1843 reverse_host_lookup/defer_ok". An error is now given when options are
1844 supplied for verify items that do not have them. (Maybe reverse_host_
1845 lookup should have a defer_ok option, but that's a different point.)
1847 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1848 defined by RFC 821) to 2048, because there were problems with some AUTH
1849 commands, and RFC 1869 says the size should be increased for extended
1850 SMTP commands that take arguments.
1852 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1855 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1856 "unknown" error; now it says that the functionality isn't in the binary.
1858 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1859 an address' error message when a string expansion fails (syntax or
1860 whatever). Otherwise the password may appear in the log. Following change
1861 PH/42 below, there is no longer a chance of it appearing in a bounce
1864 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1866 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1867 of its addresses was ever tried. (Bugzilla bug #2.)
1869 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1870 the result incorrectly in the debug output. (It correctly added a newline
1871 to what was transported.)
1873 TF/01 Added $received_time.
1875 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1876 commented out examples of how to interface to a virus scanner and to
1877 SpamAssassin. Also added commented examples of av_scanner and
1878 spamd_address settings.
1880 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1881 and controls are allowed in which ACLs. There were a couple of minor
1882 errors. Some of the entries in the conditions table (which is a table of
1883 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1884 negation of where the condition IS allowed.
1886 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1888 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1889 header file does not have a version number, so I've had to invent a new
1890 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1891 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1892 radiusclient), but it was contributed by a Radius user.
1894 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1895 files or directories, for OpenSSL.
1897 PH/19 When an Exim process that is running as root has to create an Exim log
1898 file, it does so in a subprocess that runs as exim:exim so as to get the
1899 ownership right at creation (otherwise, other Exim processes might see
1900 the file with the wrong ownership). There was no test for failure of this
1901 fork() call, which would lead to the process getting stuck as it waited
1902 for a non-existent subprocess. Forks do occasionally fail when resources
1903 run out. I reviewed all the other calls to fork(); they all seem to check
1906 PH/20 When checking for unexpected SMTP input at connect time (before writing
1907 the banner), Exim was not dealing correctly with a non-positive return
1908 from the read() function. If the client had disconnected by this time,
1909 the result was a log entry for a synchronization error with an empty
1910 string after "input=" when read() returned zero. If read() returned -1
1911 (an event I could not check), uninitialized data bytes were printed.
1912 There were reports of junk text (parts of files, etc) appearing after
1915 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1917 PH/22 Added support for macro redefinition, and (re)definition in between
1918 driver and ACL definitions.
1920 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1921 forgetting to use the resulting value; it was using the unexpanded value.
1923 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1924 hadn't been configured. The fix is from Juergen Kreileder, who
1925 understands it better than I do:
1927 "Here's what I see happening with three configured cyrus_sasl
1928 authenticators configured (plain, login, cram-md5):
1930 On startup auth_cyrus_sasl_init() gets called for each of these.
1931 This means three calls to sasl_listmech() without a specified mech_list.
1932 => SASL tests which mechs of all available mechs actually work
1933 => three warnings about OTP not working
1934 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1936 With the patch, sasl_listmech() also gets called three times. But now
1937 SASL's mech_list option is set to the server_mech specified in the the
1938 authenticator. Or in other words, the answer from sasl_listmech()
1939 gets limited to just the mech you're testing for (which is different
1941 => the return list contains just 'plain' or 'login', 'cram-md5' or
1942 nothing depending on the value of ob->server_mech.
1944 I've just tested the patch: Authentication still works fine,
1945 unavailable mechs specified in the exim configuration are still
1946 caught, and the auth.log warnings about OTP are gone."
1948 PH/25 When debugging is enabled, the contents of the command line are added
1949 to the debugging output, even when log_selector=+arguments is not
1952 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1953 answer is "GNU", and only if the return is "GNU/something" is the answer
1956 PH/27 $acl_verify_message is now set immediately after the failure of a
1957 verification in an ACL, and so is available in subsequent modifiers. In
1958 particular, the message can be preserved by coding like this:
1960 warn !verify = sender
1961 set acl_m0 = $acl_verify_message
1963 Previously, $acl_verify_message was set only while expanding "message"
1964 and "log_message" when a very denied access.
1966 PH/28 Modified OS/os.c-Linux with
1968 -#ifndef OS_LOAD_AVERAGE
1969 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1971 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1972 nomenclature these days.)
1974 PH/29 Installed patch from the Sieve maintainer that adds the options
1975 sieve_useraddress and sieve_subaddress to the redirect router.
1977 PH/30 In these circumstances:
1978 . Two addresses routed to the same list of hosts;
1979 . First host does not offer TLS;
1980 . First host accepts first address;
1981 . First host gives temporary error to second address;
1982 . Second host offers TLS and a TLS session is established;
1983 . Second host accepts second address.
1984 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1985 and peerdn, if requested) that were in fact used only for the second
1988 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1989 attention to any local part prefix or suffix that was matched by the
1990 router that accepted the address. It now behaves in the same way as it
1991 does for delivery: the affixes are removed from the local part unless
1992 rcpt_include_affixes is set on the transport.
1994 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1995 timeout during the DATA phase of an incoming message.
1997 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1998 applied a patch sanctioned by the Sieve maintainer.
2000 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2001 the uid or gid is negative. A case of a negative gid caused this to be
2002 noticed. The fix allows for either to be negative.
2004 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2005 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2008 PH/36 The change PH/12 above was broken. Fixed it.
2010 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2011 the grounds that routing the same address twice would always produce the
2012 same answer. This might have been true once, but it is certainly no
2013 longer true now. Routing a child address may depend on the previous
2014 routing that produced that child. Some complicated redirection strategies
2015 went wrong when messages had multiple recipients, and made Exim's
2016 behaviour dependent on the order in which the addresses were given.
2018 I have moved the duplicate checking until after the routing is complete.
2019 Exim scans the addresses that are assigned to local and remote
2020 transports, and removes any duplicates. This means that more work will be
2021 done, as duplicates will always all be routed, but duplicates are
2022 presumably rare, so I don't expect this is of any significance.
2024 For deliveries to pipes, files, and autoreplies, the duplicate checking
2025 still happens during the routing process, since they are not going to be
2028 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2029 It corrects a timeout issue with spamd. This is Ian's comment: "The
2030 background is that sometimes spamd either never reads data from a
2031 connection it has accepted, or it never writes response data. The exiscan
2032 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2033 blindly assumes that writes won't block so it may never time out."
2035 PH/39 Allow G after quota size as well as K and M.
2037 PH/40 The value set for $authenticated_id in an authenticator may not contain
2038 binary zeroes or newlines because the value is written to log lines and
2039 to spool files. There was no check on this. Now the value is run through
2040 the string_printing() function so that such characters are converted to
2041 printable escape sequences.
2043 PH/41 $message_linecount is a new variable that contains the total number of
2044 lines in the message. Compare $body_linecount, which is the count for the
2047 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2048 bounce and delay warning messages, except in certain special cases, which
2051 (a) An SMTP error message from a remote host;
2052 (b) A message specified in a :fail: redirection;
2053 (c) A message specified in a "fail" command in a system filter;
2054 (d) A message specified in a FAIL return from the queryprogram router;
2055 (e) A message specified by the cannot_route_message router option.
2057 In these cases only, Exim does include the error details in bounce and
2058 warning messages. There are also a few cases where bland messages such
2059 as "unrouteable address" or "local delivery error" are given.
2061 PH/43 $value is now also set for the "else" part of a ${run expansion.
2063 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2064 being worked on, but at least Exim now implements the latest version to
2067 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2068 process to complete was treated as a delivery failure, a timeout while
2069 writing the message to the pipe was logged, but erroneously treated as a
2070 successful delivery. Such timeouts include transport filter timeouts. For
2071 consistency with the overall process timeout, these timeouts are now
2072 treated as errors, giving rise to delivery failures by default. However,
2073 there is now a new Boolean option for the pipe transport called
2074 timeout_defer, which, if set TRUE, converts the failures into defers for
2075 both kinds of timeout. A transport filter timeout is now identified in
2078 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2079 systems where "make" and "gmake" are different, calling "gmake" at top
2080 level broke things. I've arranged for the value of $(MAKE) to be passed
2081 from the Makefile to this script so that it can call the same version of
2085 A note about Exim versions 4.44 and 4.50
2086 ----------------------------------------
2088 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2089 changes of various kinds. As a consequence, a big documentation update was
2090 needed. This delayed the release for rather longer than seemed good, especially
2091 in the light of a couple of (minor) security issues. Therefore, the changes
2092 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2093 release. So 4.44 and 4.50 are in effect two different branches that both start
2096 I have left the 4.50 change log unchanged; it contains all the changes since
2097 4.43. The change log for 4.44 is below; many of its items are identical to
2098 those for 4.50. This seems to be the most sensible way to preserve the
2099 historical information.
2105 1. Minor wording change to the doc/README.SIEVE file.
2107 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2108 computation of the current number of files was incorrect.
2110 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2111 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2114 4. Give more explanation in the error message when the command for a transport
2115 filter fails to execute.
2117 5. There are several places where Exim runs a non-Exim command in a
2118 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2119 was being done only for the command run by the queryprogram router. It is
2120 now done for all such subprocesses. The other cases are: ${run, transport
2121 filters, and the commands run by the lmtp and pipe transports.
2123 6. Added CONFIGURE_GROUP build-time option.
2125 7. Some older OS have a limit of 256 on the maximum number of file
2126 descriptors. Exim was using setrlimit() to set 1000 as a large value
2127 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2128 systems. I've change it so that if it can't get 1000, it tries for 256.
2130 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2131 was an oversight, and furthermore, ever since the addition of extra
2132 controls (e.g. 4.43/32), the checks on when to allow different forms of
2133 "control" were broken. There should now be diagnostics for all cases when a
2134 control that does not make sense is encountered.
2136 9. Added the /retain_sender option to "control=submission".
2138 10. $recipients is now available in the predata ACL (oversight).
2140 11. Tidy the search cache before the fork to do a delivery from a message
2141 received from the command line. Otherwise the child will trigger a lookup
2142 failure and thereby defer the delivery if it tries to use (for example) a
2143 cached ldap connection that the parent has called unbind on.
2145 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2146 of $address_data from the recipient verification was clobbered by the
2147 sender verification.
2149 13. The value of address_data from a sender verification is now available in
2150 $sender_address_data in subsequent conditions in the ACL statement.
2152 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2154 15. Added a new option "connect=<time>" to callout options, to set a different
2157 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2158 was its contents. (It was OK if the option was not defined at all.)
2160 17. A "Completed" log line is now written for messages that are removed from
2161 the spool by the -Mrm option.
2163 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2164 information about exactly what failed.
2166 19. Added -dd to debug only the daemon process.
2168 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2169 handles timeouts, both on the server side and network timeouts. Renamed the
2170 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2172 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2173 It is now set to "smtps".
2175 22. $host_address is now set to the target address during the checking of
2176 ignore_target_hosts.
2178 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2179 being passed; this would have caused $sender_host_name to have been used if
2180 matching the list had actually called for a host name (not very likely,
2181 since this list is usually IP addresses). A host name is now passed as
2184 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2185 code that specifies a non-restarting handler (typically sigaction() in
2186 modern systems) in an attempt to fix a rare and obscure crash bug.
2188 25. Narrowed the window for a race in the daemon that could cause it to ignore
2189 SIGCHLD signals. This is not a major problem, because they are used only to
2190 wake it up if nothing else does.
2192 26. A malformed maildirsize file could cause Exim to calculate negative values
2193 for the mailbox size or file count. Odd effects could occur as a result.
2194 The maildirsize information is now recalculated if the size or filecount
2197 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2198 support for a long time. Removed HAVE_SYS_VFS_H.
2200 28. Installed the latest version of exipick from John Jetmore.
2202 29. In an address list, if the pattern was not a regular expression, an empty
2203 subject address (from a bounce message) matched only if the pattern was an
2204 empty string. Non-empty patterns were not even tested. This was the wrong
2205 because it is perfectly reasonable to use an empty address as part of a
2206 database query. An empty address is now tested by patterns that are
2207 lookups. However, all the other forms of pattern expect the subject to
2208 contain a local part and a domain, and therefore, for them, an empty
2209 address still always fails if the pattern is not itself empty.
2211 30. Exim went into a mad DNS loop when attempting to do a callout where the
2212 host was specified on an smtp transport, and looking it up yielded more
2213 than one IP address.
2215 31. Re-factored the code for checking spool and log partition space into a
2216 function that finds that data and another that does the check. The former
2217 is then used to implement four new variables: $spool_space, $log_space,
2218 $spool_inodes, and $log_inodes.
2220 32. The RFC2047 encoding function was originally intended for short strings
2221 such as real names; it was not keeping to the 75-character limit for
2222 encoded words that the RFC imposes. It now respects the limit, and
2223 generates multiple encoded words if necessary. To be on the safe side, I
2224 have increased the buffer size for the ${rfc2047: expansion operator from
2227 33. It is now permitted to omit both strings after an "if" condition; if the
2228 condition is true, the result is "true". As before, when the second string
2229 is omitted, a false condition yields an empty string. This makes it less
2230 cumbersome to write custom ACL and router conditions.
2232 34. Failure to deliver a bounce message always caused it to be frozen, even if
2233 there was an errors_to setting on the router. The errors_to setting is now
2236 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2237 canonical form (fully expanded) before being placed in
2238 $sender_host_address.
2240 36. The table in the code that translates DNS record types into text (T_A to
2241 "A" for instance) was missing entries for NS and CNAME. It is just possible
2242 that this could have caused confusion if both these types were looked up
2243 for the same domain, because the text type is used as part of Exim's
2244 per-process caching. But the chance of anyone hitting this buglet seems
2247 37. The dnsdb lookup has been extended in a number of ways.
2249 (1) There is a new type, "zns", which walks up the domain tree until it
2250 finds some nameserver records. It should be used with care.
2252 (2) There is a new type, "mxh", which is like "mx" except that it returns
2253 just the host names, not the priorities.
2255 (3) It is now possible to give a list of domains (or IP addresses) to be
2256 looked up. The behaviour when one of the lookups defers can be
2257 controlled by a keyword.
2259 (4) It is now possible to specify the separator character for use when
2260 multiple records are returned.
2262 38. The dnslists ACL condition has been extended: it is now possible to supply
2263 a list of IP addresses and/or domains to be looked up in a particular DNS
2266 39. Added log_selector=+queue_time_overall.
2268 40. When running the queue in the test harness, wait just a tad after forking a
2269 delivery process, to get repeatability of debugging output.
2271 41. Include certificate and key file names in error message when GnuTLS fails
2272 to set them up, because the GnuTLS error message doesn't include the name
2273 of the failing file when there is a problem reading it.
2275 42. Allow both -bf and -bF in the same test run.
2277 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2279 44. The "Exiscan patch" is now merged into the mainline Exim source.
2281 45. Sometimes the final signoff response after QUIT could fail to get
2282 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2283 < 0 before doing a fflush(). This bug looks as though it goes back to the
2284 introduction of TLS in release 3.20, but "sometimes" must have been rare
2285 because the tests only now provoked it.
2287 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2288 (this can affect the format of dates).
2290 47. exim_tidydb, when checking for the continued existence of a message for
2291 which it has found a message-specific retry record, was not finding
2292 messages that were in split spool directories. Consequently, it was
2293 deleting retry records that should have stayed in existence.
2295 48. Steve fixed some bugs in eximstats.
2297 49. The SPA authentication driver was not abandoning authentication and moving
2298 on to the next authenticator when an expansion was forced to fail,
2299 contradicting the general specification for all authenticators. Instead it
2300 was generating a temporary error. It now behaves as specified.
2302 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2303 (the order specifies the preference for clients). The order is now AES256,
2304 AES128, 3DES, ARCFOUR128.
2306 51. Small patch to Sieve code - explicitly set From: when generating an
2309 52. Exim crashed if a remote delivery caused a very long error message to be
2310 recorded - for instance if somebody sent an entire SpamAssassin report back
2311 as a large number of 550 error lines. This bug was coincidentally fixed by
2312 increasing the size of one of Exim's internal buffers (big_buffer) that
2313 happened as part of the Exiscan merge. However, to be on the safe side, I
2314 have made the code more robust (and fixed the comments that describe what
2317 53. Now that there can be additional text after "Completed" in log lines (if
2318 the queue_time_overall log selector is set), a one-byte patch to exigrep
2319 was needed to allow it to recognize "Completed" as not the last thing in
2322 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2323 patch that reportedly fixes this has been added. I am not expert enough to
2324 create a test for it. This is what the patch creator wrote:
2326 "I found a little strange behaviour of ldap code when working with
2327 Windows 2003 AD Domain, where users was placed in more than one
2328 Organization Units. When I tried to give exim partial DN, the exit code
2329 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2330 But simultaneously result of request was absolutely normal ldap result,
2331 so I produce this patch..."
2333 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2334 REFERENCE, so I have modified the code to exclude the patch when that macro
2337 55. Some experimental protocols are using DNS PTR records for new purposes. The
2338 keys for these records are domain names, not reversed IP addresses. The
2339 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2340 leaves it alone. Component reversal etc. now happens only for IP addresses.
2343 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2345 57. Double the size of the debug message buffer (to 2048) so that more of very
2346 long debug lines gets shown.
2348 58. The exicyclog utility now does better if the number of log files to keep
2349 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2351 59. Two changes related to the smtp_active_hostname option:
2353 (1) $smtp_active_hostname is now available as a variable.
2354 (2) The default for smtp_banner uses $smtp_active_hostname instead
2355 of $primary_hostname.
2357 60. The host_aton() function is supposed to be passed a string that is known
2358 to be a valid IP address. However, in the case of IPv6 addresses, it was
2359 not checking this. This is a hostage to fortune. Exim now panics and dies
2360 if the condition is not met. A case was found where this could be provoked
2361 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2362 components; fortuitously, this particular loophole had already been fixed
2363 by change 4.50/55 above.
2365 If there are any other similar loopholes, the new check in host_aton()
2366 itself should stop them being exploited. The report I received stated that
2367 data on the command line could provoke the exploit when Exim was running as
2368 exim, but did not say which command line option was involved. All I could
2369 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2370 running as the user.
2373 61. There was a buffer overflow vulnerability in the SPA authentication code
2374 (which came originally from the Samba project). I have added a test to the
2375 spa_base64_to_bits() function which I hope fixes it.
2378 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2379 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2381 63. The daemon start-up calls getloadavg() while still root for those OS that
2382 need the first call to be done as root, but it missed one case: when
2383 deliver_queue_load_max is set with deliver_drop_privilege. This is
2384 necessary for the benefit of the queue runner, because there is no re-exec
2385 when deliver_drop_privilege is set.
2387 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2388 This has been fixed.
2390 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2391 was in use, was not putting the data itself into the right store pool;
2392 consequently, it could be overwritten for a subsequent message in the same
2393 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2396 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2398 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2399 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2400 the function were treating the return as a boolean value, which happened to
2401 work because 0=false and not-0=true, but is not correct code.
2403 68. The host_aton() function was not handling scoped IPv6 addresses (those
2404 with, for example, "%eth0" on the end) correctly.
2406 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2407 negated items (that is, ~something) in unsigned ints. Some compilers
2408 apparently mutter when there is no cast.
2410 70. If an address verification called from an ACL failed, and did not produce a
2411 user-specific message (i.e. there was only a "system" message), nothing was
2412 put in $acl_verify_message. In this situation, it now puts the system
2415 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2416 session; change 4.31/43 added the unwanted input to the log line - except
2417 that it did not do this in the start of session case. It now does.
2419 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2420 This is wrong and can cause the other end to generate a synchronization
2421 error if it is another Exim or anything else that does the synchronization
2422 check. A QUIT command is no longer sent after a timeout.
2424 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2425 during host lookups.
2427 74. The defer_ok option of callout verification was not working if it was used
2428 when verifying addresses in header lines, that is, for this case:
2430 verify = header_sender/callout=defer_ok
2432 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2433 those file descriptors could be used for SMTP connections. If anything
2434 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2435 could be sent to the SMTP client, causing chaos. The daemon now opens
2436 stdin, stdout, and stderr to /dev/null when it puts itself into the
2439 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2440 log by default. The user can override this with suitable Perl magic.
2442 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2443 the log message when discard triggers, was not working for the DATA ACL or
2444 for the non-SMTP ACL.
2446 78. Error message wording change in sieve.c.
2448 79. If smtp_accept_max_per_host was set, the number of connections could be
2449 restricted to fewer than expected, because the daemon was trying to set up
2450 a new connection before checking whether the processes handling previous
2451 connections had finished. The check for completed processes is now done
2452 earlier. On busy systems, this bug wouldn't be noticed because something
2453 else would have woken the daemon, and it would have reaped the completed
2456 80. If a message was submitted locally by a user whose login name contained one
2457 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2458 It caused a spool format error. I have fixed the spool reading code. A
2459 related problem was that the "from" clause in the Received: line became
2460 illegal because of the space(s). It is now covered by ${quote_local_part.
2462 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2465 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2467 83. Patch from Sieve maintainer to fix unterminated string problem in
2468 "vacation" handling.
2470 84. Some minor changes to the Linux configuration files to help with other
2471 OS variants using glibc.
2473 85. One more patch for Sieve to update vacation handling to latest spec.
2476 ----------------------------------------------------
2477 See the note above about the 4.44 and 4.50 releases.
2478 ----------------------------------------------------
2484 1. Change 4.43/35 introduced a bug that caused file counts to be
2485 incorrectly computed when quota_filecount was set in an appendfile
2488 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2489 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2492 3. Give more explanation in the error message when the command for a transport
2493 filter fails to execute.
2495 4. There are several places where Exim runs a non-Exim command in a
2496 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2497 was being done only for the command run by the queryprogram router. It is
2498 now done for all such subprocesses. The other cases are: ${run, transport
2499 filters, and the commands run by the lmtp and pipe transports.
2501 5. Some older OS have a limit of 256 on the maximum number of file
2502 descriptors. Exim was using setrlimit() to set 1000 as a large value
2503 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2504 systems. I've change it so that if it can't get 1000, it tries for 256.
2506 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2507 was an oversight, and furthermore, ever since the addition of extra
2508 controls (e.g. 4.43/32), the checks on when to allow different forms of
2509 "control" were broken. There should now be diagnostics for all cases when a
2510 control that does not make sense is encountered.
2512 7. $recipients is now available in the predata ACL (oversight).
2514 8. Tidy the search cache before the fork to do a delivery from a message
2515 received from the command line. Otherwise the child will trigger a lookup
2516 failure and thereby defer the delivery if it tries to use (for example) a
2517 cached ldap connection that the parent has called unbind on.
2519 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2520 of $address_data from the recipient verification was clobbered by the
2521 sender verification.
2523 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2524 was its contents. (It was OK if the option was not defined at all.)
2526 11. A "Completed" log line is now written for messages that are removed from
2527 the spool by the -Mrm option.
2529 12. $host_address is now set to the target address during the checking of
2530 ignore_target_hosts.
2532 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2533 being passed; this would have caused $sender_host_name to have been used if
2534 matching the list had actually called for a host name (not very likely,
2535 since this list is usually IP addresses). A host name is now passed as
2538 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2539 code that specifies a non-restarting handler (typically sigaction() in
2540 modern systems) in an attempt to fix a rare and obscure crash bug.
2542 15. Narrowed the window for a race in the daemon that could cause it to ignore
2543 SIGCHLD signals. This is not a major problem, because they are used only to
2544 wake it up if nothing else does.
2546 16. A malformed maildirsize file could cause Exim to calculate negative values
2547 for the mailbox size or file count. Odd effects could occur as a result.
2548 The maildirsize information is now recalculated if the size or filecount
2551 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2552 support for a long time. Removed HAVE_SYS_VFS_H.
2554 18. Updated exipick to current release from John Jetmore.
2556 19. Allow an empty sender to be matched against a lookup in an address list.
2557 Previously the only cases considered were a regular expression, or an
2560 20. Exim went into a mad DNS lookup loop when doing a callout where the
2561 host was specified on the transport, if the DNS lookup yielded more than
2564 21. The RFC2047 encoding function was originally intended for short strings
2565 such as real names; it was not keeping to the 75-character limit for
2566 encoded words that the RFC imposes. It now respects the limit, and
2567 generates multiple encoded words if necessary. To be on the safe side, I
2568 have increased the buffer size for the ${rfc2047: expansion operator from
2571 22. Failure to deliver a bounce message always caused it to be frozen, even if
2572 there was an errors_to setting on the router. The errors_to setting is now
2575 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2576 canonical form (fully expanded) before being placed in
2577 $sender_host_address.
2579 24. Updated eximstats to version 1.33
2581 25. Include certificate and key file names in error message when GnuTLS fails
2582 to set them up, because the GnuTLS error message doesn't include the name
2583 of the failing file when there is a problem reading it.
2585 26. Expand error message when OpenSSL has problems setting up cert/key files.
2588 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2589 (this can affect the format of dates).
2591 28. exim_tidydb, when checking for the continued existence of a message for
2592 which it has found a message-specific retry record, was not finding
2593 messages that were in split spool directories. Consequently, it was
2594 deleting retry records that should have stayed in existence.
2596 29. eximstats updated to version 1.35
2597 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2598 1.35 - bugfix such that pie charts by volume are generated correctly
2600 30. The SPA authentication driver was not abandoning authentication and moving
2601 on to the next authenticator when an expansion was forced to fail,
2602 contradicting the general specification for all authenticators. Instead it
2603 was generating a temporary error. It now behaves as specified.
2605 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2606 (the order specifies the preference for clients). The order is now AES256,
2607 AES128, 3DES, ARCFOUR128.
2609 31. Small patch to Sieve code - explicitly set From: when generating an
2612 32. Exim crashed if a remote delivery caused a very long error message to be
2613 recorded - for instance if somebody sent an entire SpamAssassin report back
2614 as a large number of 550 error lines. This bug was coincidentally fixed by
2615 increasing the size of one of Exim's internal buffers (big_buffer) that
2616 happened as part of the Exiscan merge. However, to be on the safe side, I
2617 have made the code more robust (and fixed the comments that describe what
2620 33. Some experimental protocols are using DNS PTR records for new purposes. The
2621 keys for these records are domain names, not reversed IP addresses. The
2622 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2623 leaves it alone. Component reversal etc. now happens only for IP addresses.
2626 34. The host_aton() function is supposed to be passed a string that is known
2627 to be a valid IP address. However, in the case of IPv6 addresses, it was
2628 not checking this. This is a hostage to fortune. Exim now panics and dies
2629 if the condition is not met. A case was found where this could be provoked
2630 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2631 components; fortuitously, this particular loophole had already been fixed
2632 by change 4.50/55 or 4.44/33 above.
2634 If there are any other similar loopholes, the new check in host_aton()
2635 itself should stop them being exploited. The report I received stated that
2636 data on the command line could provoke the exploit when Exim was running as
2637 exim, but did not say which command line option was involved. All I could
2638 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2639 running as the user.
2642 35. There was a buffer overflow vulnerability in the SPA authentication code
2643 (which came originally from the Samba project). I have added a test to the
2644 spa_base64_to_bits() function which I hope fixes it.
2647 36. The daemon start-up calls getloadavg() while still root for those OS that
2648 need the first call to be done as root, but it missed one case: when
2649 deliver_queue_load_max is set with deliver_drop_privilege. This is
2650 necessary for the benefit of the queue runner, because there is no re-exec
2651 when deliver_drop_privilege is set.
2653 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2654 was in use, was not putting the data itself into the right store pool;
2655 consequently, it could be overwritten for a subsequent message in the same
2656 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2659 38. Sometimes the final signoff response after QUIT could fail to get
2660 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2661 < 0 before doing a fflush(). This bug looks as though it goes back to the
2662 introduction of TLS in release 3.20, but "sometimes" must have been rare
2663 because the tests only now provoked it.
2669 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2670 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2671 Now it returns an integer. A number of calls were still expecting a T/F
2672 return. Fortuitously, in all cases, the tests worked in OK situations,
2673 which is the norm. However, things would have gone wrong on any write
2674 failures on the smtp file descriptor. This function is used when sending
2675 messages over SMTP and also when doing verify callouts.
2677 2. When Exim is called to do synchronous delivery of a locally submitted
2678 message (the -odf or -odi options), it no longer closes stderr before doing
2681 3. Implemented the mua_wrapper option.
2683 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2685 5. Implemented the functions header_remove(), header_testname(),
2686 header_add_at_position(), and receive_remove_recipient(), and exported them
2689 6. If an ACL "warn" statement specified the addition of headers, Exim already
2690 inserted X-ACL-Warn: at the start if there was no header name. However, it
2691 was not making this test for the second and subsequent header lines if
2692 there were newlines in the string. This meant that an invalid header could
2693 be inserted if Exim was badly configured.
2695 7. Allow an ACL "warn" statement to add header lines at the start or after all
2696 the Received: headers, as well as at the end.
2698 8. Added the rcpt_4xx retry error code.
2700 9. Added postmaster_mailfrom=xxx to callout verification option.
2702 10. Added mailfrom=xxxx to the callout verification option, for verify=
2705 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2706 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2708 12. Inserted some casts to stop certain compilers warning when using pointer
2709 differences as field lengths or precisions in printf-type calls (mostly
2710 affecting debugging statements).
2712 13. Added optional readline() support for -be (dynamically loaded).
2714 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2715 same clock tick as a message's arrival, so that its received time was the
2716 same as the "first fail" time on the retry record, and that message
2717 remained on the queue past the ultimate address timeout, every queue runner
2718 would try a delivery (because it was past the ultimate address timeout) but
2719 after another failure, the ultimate address timeout, which should have then
2720 bounced the address, did not kick in. This was a "< instead of <=" error;
2721 in most cases the first failure would have been in the next clock tick
2722 after the received time, and all would be well.
2724 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2725 being recognized when the domain list was tested by the match_domain
2726 condition in an expansion string.
2728 16. Added the ${str2b64: operator.
2730 17. Exim was always calling setrlimit() to set a large limit for the number of
2731 processes, without checking whether the existing limit was already
2732 adequate. (It did check for the limit on file descriptors.) Furthermore,
2733 errors from getrlimit() and setrlimit() were being ignored. Now they are
2734 logged to the main and panic logs, but Exim does carry on, to try to do its
2735 job under whatever limits there are.
2737 18. Imported PCRE 5.0.
2739 19. Trivial typo in log message " temporarily refused connection" (the leading
2742 20. If the log selector return_path_on_delivery was set and an address was
2743 redirected to /dev/null, the delivery process crashed because it assumed
2744 that a return path would always be set for a "successful" delivery. In this
2745 case, the whole delivery is bypassed as an optimization, and therefore no
2748 21. Internal re-arrangement: the function for sending a challenge and reading
2749 a response while authentication was assuming a zero-terminated challenge
2750 string. It's now changed to take a pointer and a length, to allow for
2751 binary data in such strings.
2753 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2755 23. Exim was not respecting finduser_retries when seeking the login of the
2756 uid under which it was called; it was always trying 10 times. (The default
2757 setting of finduser_retries is zero.) Also, it was sleeping after the final
2758 failure, which is pointless.
2760 24. Implemented tls_on_connect_ports.
2762 25. Implemented acl_smtp_predata.
2764 26. If the domain in control=submission is set empty, Exim assumes that the
2765 authenticated id is a complete email address when it generates From: or
2766 Sender: header lines.
2768 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2769 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2770 chown and chgrp in /bin and hostname in /usr/bin.
2772 28. Exim was keeping the "process log" file open after each use, just as it
2773 does for the main log. This opens the possibility of it remaining open for
2774 long periods when the USR1 signal hits a daemon. Occasional processlog
2775 errors were reported, that could have been caused by this. Anyway, it seems
2776 much more sensible not to leave this file open at all, so that is what now
2779 29. The long-running daemon process does not normally write to the log once it
2780 has entered its main loop, and it closes the log before doing so. This is
2781 so that log files can straightforwardly be renamed and moved. However,
2782 there are a couple of unusual error situations where the daemon does write
2783 log entries, and I had neglected to close the log afterwards.
2785 30. The text of an SMTP error response that was received during a remote
2786 delivery was being truncated at 512 bytes. This is too short for some of
2787 the long messages that one sometimes sees. I've increased the limit to
2790 31. It is now possible to make retry rules that apply only when a message has a
2791 specific sender, in particular, an empty sender.
2793 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2794 it possible to be selective about when SMTP synchronization is enforced.
2796 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2798 32. Implemented hosts_connection_nolog.
2800 33. Added an ACL for QUIT.
2802 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2805 35. Added mailbox_size and mailbox_filecount to appendfile.
2807 36. Added control = no_multiline_responses to ACLs.
2809 37. There was a bug in the logic of the code that waits for the clock to tick
2810 in the case where the clock went backwards by a substantial amount such
2811 that the microsecond fraction of "now" was more than the microsecond
2812 fraction of "then" (but the whole seconds number was less).
2814 38. Added support for the libradius Radius client library this is found on
2815 FreeBSD (previously only the radiusclient library was supported).
2821 1. When certain lookups returned multiple values in the form name=value, the
2822 quoting of the values was not always being done properly. Specifically:
2823 (a) If the value started with a double quote, but contained no whitespace,
2825 (b) If the value contained whitespace other than a space character (i.e.
2826 tabs or newlines or carriage returns) it was not quoted.
2827 This fix has been applied to the mysql and pgsql lookups by writing a
2828 separate quoting function and calling it from the lookup code. The fix
2829 should probably also be applied to nisplus, ibase and oracle lookups, but
2830 since I cannot test any of those, I have not disturbed their existing code.
2832 2. A hit in the callout cache for a specific address caused a log line with no
2833 reason for rejecting RCPT. Now it says "Previous (cached) callout
2834 verification failure".
2836 3. There was an off-by-one bug in the queryprogram router. An over-long
2837 return line was truncated at 256 instead of 255 characters, thereby
2838 overflowing its buffer with the terminating zero. As well as fixing this, I
2839 have increased the buffer size to 1024 (and made a note to document this).
2841 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2842 when Exim is waiting for an SMTP response from a remote server, Exim
2843 restarts its select() call on the socket, thereby resetting its timeout.
2844 This is not a problem when such interrupts are rare. Somebody set up a cron
2845 job to run exiwhat every 2 minutes, which is less than the normal select()
2846 timeout (5 or 10 minutes). This meant that the select() timeout never
2847 kicked in because it was always reset. I have fixed this by comparing the
2848 time when an interrupt arrives with the time at the start of the first call
2849 to select(). If more time than the timeout has elapsed, the interrupt is
2850 treated as a timeout.
2852 5. Some internal re-factoring in preparation for the addition of Sieve
2853 extensions (by MH). In particular, the "personal" test is moved to a
2854 separate function, and given an option for scanning Cc: and Bcc: (which is
2855 not set for Exim filters).
2857 6. When Exim created an email address using the login of the caller as the
2858 local part (e.g. when creating a From: or Sender: header line), it was not
2859 quoting the local part when it contained special characters such as @.
2861 7. Installed new OpenBSD configuration files.
2863 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2864 try to make them clearer.
2866 9. Callout options, other than the timeout value, were being ignored when
2867 verifying sender addresses in header lines. For example, when using
2869 verify = header_sender/callout=no_cache
2871 the cache was (incorrectly) being used.
2873 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2874 only the Cygwin environment.
2876 11. When return_path_on_delivery was set as a log selector, if different remote
2877 addresses in the same message used different return paths and parallel
2878 remote delivery occurred, the wrong values would sometimes be logged.
2879 (Whenever a remote delivery process finished, the return path value from
2880 the most recently started remote delivery process was logged.)
2882 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2883 lines when AUTH and/or TLS are in use. This is the "received protocol"
2884 field. Exim used to use "asmtp" for authenticated SMTP, without any
2885 indication (in the protocol name) for TLS use. Now it follows the RFC and
2886 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2887 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2888 names appear in log lines as well as in Received: header lines.
2890 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2891 extensions, and comparison tests, and to fix some bugs.
2893 14. Changes to the "personal" filter test:
2895 (1) The test was buggy in that it was just doing the equivalent of
2896 "contains" tests on header lines. For example, if a user's address was
2897 anne@some.where, the "personal" test would incorrectly be true for
2899 To: susanne@some.where
2901 This test is now done by extracting each address from the header in turn,
2902 and checking the entire address. Other tests that are part of "personal"
2903 are now done using regular expressions (for example, to check local parts
2904 of addresses in From: header lines).
2906 (2) The list of non-personal local parts in From: addresses has been
2907 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2908 taken from the Sieve specification recommendations.
2910 (3) If the message contains any header line starting with "List-" it is
2911 treated as non-personal.
2913 (4) The test for "circular" in the Subject: header line has been removed
2914 because it now seems ill-conceived.
2916 15. Minor typos in src/EDITME comments corrected.
2918 16. Installed latest exipick from John Jetmore.
2920 17. If headers_add on a router specified a text string that was too long for
2921 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2922 of string_sprintf() is now avoided.
2924 18. $message_body_size was not set (it was always zero) when running the DATA
2925 ACL and the local_scan() function.
2927 19. For the "mail" command in an Exim filter, no default was being set for
2928 the once_repeat time, causing a random time value to be used if "once" was
2929 specified. (If the value happened to be <= 0, no repeat happened.) The
2930 default is now 0s, meaning "never repeat". The "vacation" command was OK
2931 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2932 (I found it when inspecting the code).
2934 20. There is now an overall timeout for performing a callout verification. It
2935 defaults to 4 times the callout timeout, which applies to individual SMTP
2936 commands during the callout. The overall timeout applies when there is more
2937 than one host that can be tried. The timeout is checked before trying the
2938 next host. This prevents very long delays if there are a large number of
2939 hosts and all are timing out (e.g. when the network connections are timing
2940 out). The value of the overall timeout can be changed by specifying an
2941 additional sub-option for "callout", called "maxwait". For example:
2943 verify = sender/callout=5s,maxwait=20s
2945 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2946 to the end before writing, but this should make it even safer).
2948 22. Exim was forgetting that it had advertised PIPELINING for the second and
2949 subsequent messages on an SMTP connection. It was also not resetting its
2950 memory on STARTTLS and an internal HELO.
2952 23. When Exim logs an SMTP synchronization error within a session, it now
2953 records whether PIPELINING has been advertised or not.
2955 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2956 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2957 rather than long int.
2959 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2961 26. Added the never_mail option to autoreply.
2967 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2968 crash if the getsockname() call failed; this can happen if a connection is
2969 closed very soon after it is established. The problem was simply in the
2970 order in which certain operations were done, causing Exim to try to write
2971 to the SMTP stream before it had set up the file descriptor. The bug has
2972 been fixed by making things happen in the correct order.
2978 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2979 before the connection was closed, thus losing the rejection response.
2981 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2982 some early Solaris releases, but causes trouble in current releases where
2983 socklen_t is defined.
2985 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2988 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2991 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2992 that releasing the top of it at the end releases what was used for sub-
2993 expansions (unless the block got too big). However, discard this block if
2994 the first thing is a variable or header, so that we can use its block when
2995 it is dynamic (useful for very large $message_headers, for example).
2997 6. Lookups now cache *every* query, not just the most recent. A new, separate
2998 store pool is used for this. It can be recovered when all lookup caches are
2999 flushed. Lookups now release memory at the end of their result strings.
3000 This has involved some general refactoring of the lookup sources.
3002 7. Some code has been added to the store_xxx() functions to reduce the amount
3003 of flapping under certain conditions.
3005 8. log_incoming_interface used to affect only the <= reception log lines. Now
3006 it causes the local interface and port to be added to several more SMTP log
3007 lines, for example "SMTP connection from", and rejection lines.
3009 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3011 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3013 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3014 could be overwritten at the end of the current message (or the start of a
3015 new message if it was set in a HELO ACL). The value is now preserved for
3016 the duration of the SMTP connection.
3018 12. If a transport had a headers_rewrite setting, and a matching header line
3019 contained an unqualified address, that address was qualified, even if it
3020 did not match any rewriting rules. The underlying bug was that the values
3021 of the flags that permit the existence of unqualified sender and recipient
3022 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3023 non-local messages, and by -bnq for local messages) were not being
3024 preserved with the message after it was received.
3026 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3027 "next input=" as part of the text comprising the host identity instead of
3028 the correct text. The code was using the same buffer for two different
3029 strings. However, depending on which order the printing function evaluated
3030 its arguments, the bug did not always show up. Under Linux, for example, my
3031 test suite worked just fine.
3033 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3034 doesn't work with some older versions of Perl. It has been changed to "my",
3035 which in any case is probably the better facility to use.
3037 15. A really picky compiler found some instances of statements for creating
3038 error messages that either had too many or two few arguments for the format
3041 16. The size of the buffer for calls to the DNS resolver has been increased
3042 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3043 for addresses that have a lot of PTR records. This alleviates a problem; it
3044 does not fully solve it.
3046 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3047 buffer now truncates the list and logs the incident, which is the same
3048 action as happens when Exim is looking up a host name and its aliases.
3049 Previously in this situation something unpredictable would happen;
3050 sometimes it was "internal error: store_reset failed".
3052 18. If a server dropped the connection unexpectedly when an Exim client was
3053 using GnuTLS and trying to read a response, the client delivery process
3054 crashed while trying to generate an error log message.
3056 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3057 string, for example:
3059 warn message = H1: something\nH2: something
3061 the text was added as a single header line from Exim's point of view
3062 though it ended up OK in the delivered message. However, searching for the
3063 second and subsequent header lines using $h_h2: did not work. This has been
3064 fixed. Similarly, if a system filter added multiple headers in this way,
3065 the routers could not see them.
3067 20. Expanded the error message when iplsearch is called with an invalid key to
3068 suggest using net-iplsearch in a host list.
3070 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3071 ACLs are no longer actually imposed (and a message to that effect is
3074 22. If a "gecos" field in a passwd entry contained escaped characters, in
3075 particular, if it contained a \" sequence, Exim got it wrong when building
3076 a From: or a Sender: header from that name. A second bug also caused
3077 incorrect handling when an unquoted " was present following a character
3078 that needed quoting.
3080 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3081 was not being matched caselessly.
3083 24. Arranged for all hyphens in the exim.8 source to be escaped with
3086 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3087 properly. Recipient callout cache records were still being keyed to include
3088 the sender, even when use_sender was set false. This led to far more
3089 callouts that were necessary. The sender is no longer included in the key
3090 when use_sender is false.
3092 26. Added "control = submission" modifier to ACLs.
3094 27. Added the ${base62d: operator to decode base 62 numbers.
3096 28. dnsdb lookups can now access SRV records.
3098 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3099 the configuration file.
3101 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3102 (-v) mode. This makes the output for a verbose queue run more intelligible.
3104 31. Added a use_postmaster feature to recipient callouts.
3106 32. Added the $body_zerocount variable, containing the number of binary zero
3107 bytes in the message body.
3109 33. The time of last modification of the "new" subdirectory is now used as the
3110 "mailbox time last read" when there is a quota error for a maildir
3113 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3115 35. Added +ignore_unknown as a special item in host lists.
3117 36. Code for decoding IPv6 addresses in host lists is now included, even if
3118 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3119 address was recognized as an IP address, but was then not correctly decoded
3120 into binary, causing unexpected and incorrect effects when compared with
3127 1. Very minor rewording of debugging text in manualroute to say "list of
3128 hosts" instead of "hostlist".
3130 2. If verify=header_syntax was set, and a header line with an unqualified
3131 address (no domain) and a large number of spaces between the end of the
3132 name and the colon was received, the reception process suffered a buffer
3133 overflow, and (when I tested it) crashed. This was caused by some obsolete
3134 code that should have been removed. The fix is to remove it!
3136 3. When running in the test harness, delay a bit after writing a bounce
3137 message to get a bit more predictability in the log output.
3139 4. Added a call to search_tidyup() just before forking a reception process. In
3140 theory, someone could use a lookup in the expansion of smtp_accept_max_
3141 per_host which, without the tidyup, could leave open a database connection.
3143 5. Added the variables $recipient_data and $sender_data which get set from a
3144 lookup success in an ACL "recipients" or "senders" condition, or a router
3145 "senders" option, similar to $domain_data and $local_part_data.
3147 6. Moved the writing of debug_print from before to after the "senders" test
3150 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3151 problems for message scanning, either using a data ACL, or using
3152 local_scan() because the Received: header was not generated till after they
3153 were called (in order to set the time as the time of reception completion).
3154 I have revised the way this works. The header is now generated after the
3155 body is received, but before the ACL or local_scan() are called. After they
3156 are run, the timestamp in the header is updated.
3162 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3163 before starting a queue runner without re-exec. This happened only when
3164 deliver_drop_privilege was set or when the Exim user was set to root. The
3165 effect of the bug was that timeouts during subsequent deliveries caused
3166 crashes instead of being properly handled. The handler is now left at its
3167 default (and expected) setting.
3169 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3170 message, again when deliver_drop_privilege is set or Exim is run as root.
3171 The bug described in (1) was not present in this case, but the tidying up
3172 of the other signals was missing. I have made the two cases consistent.
3174 3. The ignore_target_hosts setting on a manualroute router was being ignored
3175 for hosts that were looked up using the /MX notation.
3177 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3180 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3181 operated on the sender address. After changing the $sender_address to <>
3182 for the sender address verify, Exim was re-instated it as the original
3183 (before rewriting) address, but remembering that it had rewritten it, so it
3184 wasn't rewriting it again. This bug also had the effect of breaking the
3185 sender address verification caching when the sender address was rewritten.
3187 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3188 This has been changed so that if the ip literal address matches
3189 ignore_target_hosts, the router declines.
3191 7. Added expansion conditions match_domain, match_address, and match_local_
3192 part (NOT match_host).
3194 8. The placeholder for the Received: header didn't have a length field set.
3196 9. Added code to Exim itself and to exim_lock to test for a specific race
3197 condition that could lead to file corruption when using MBX delivery. The
3198 issue is with the lockfile that is created in /tmp. If this file is removed
3199 after a process has opened it but before that process has acquired a lock,
3200 there is the potential for a second process to recreate the file and also
3201 acquire a lock. This could lead to two Exim processes writing to the file
3202 at the same time. The added code performs the same test as UW imapd; it
3203 checks after acquiring the lock that its file descriptor still refers to
3204 the same named file.
3206 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3207 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3210 11. Added the smtp_active_hostname option. If used, this will typically be made
3211 to depend on the incoming interface address. Because $interface_address is
3212 not set up until the daemon has forked a reception process, error responses
3213 that can happen earlier (such as "too many connections") no longer contain
3216 12. If an expansion in a condition on a "warn" statement fails because a lookup
3217 defers, the "warn" statement is abandoned, and the next ACL statement is
3218 processed. Previously this caused the whole ACL to be aborted.
3220 13. Added the iplsearch lookup type.
3222 14. Added ident_timeout as a log selector.
3224 15. Added tls_certificate_verified as a log selector.
3226 16. Added a global option tls_require_ciphers (compare the smtp transport
3227 option of the same name). This controls incoming TLS connections.
3229 17. I finally figured out how to make tls_require_ciphers do a similar thing
3230 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3231 before starting the TLS session.
3233 18. Tabs are now shown as \t in -bP output.
3235 19. If the log selector return_path_on_delivery was set, Exim crashed when
3236 bouncing a message because it had too many Received: header lines.
3238 20. If two routers both had headers_remove settings, and the first one included
3239 a superfluous trailing colon, the final name in the first list and the
3240 first name in the second list were incorrectly joined into one item (with a
3241 colon in the middle).
3247 1. Added -C and -D options to the exinext utility, mainly to make it easier
3248 to include in the automated testing, but these could be helpful when
3249 multiple configurations are in use.
3251 2. The exinext utility was not formatting the output nicely when there was
3252 an alternate port involved in the retry record key, nor when there was a
3253 message id as well (for retries that were specific to a specific message
3254 and a specific host). It was also confused by IPv6 addresses, because of
3255 the additional colons they contain. I have fixed the IPv4 problem, and
3256 patched it up to do a reasonable job for IPv6.
3258 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3259 delivery, the log line now contains "pipelined" if PIPELINING was used.
3261 4. An SMTP transport process used to panic and die if the bind() call to set
3262 an explicit outgoing interface failed. This has been changed; it is now
3263 treated in the same way as a connect() failure.
3265 5. A reference to $sender_host_name in the part of a conditional expansion
3266 that was being skipped was still causing a DNS lookup. This no longer
3269 6. The def: expansion condition was not recognizing references to header lines
3270 that used bh_ and bheader_.
3272 7. Added the _cache feature to named lists.
3274 8. The code for checking quota_filecount in the appendfile transport was
3275 allowing one more file than it should have been.
3277 9. For compatibility with Sendmail, the command line option
3285 and sets the incoming protocol and host name (for trusted callers). The
3286 host name and its colon can be omitted when only the protocol is to be set.
3287 Note the Exim already has two private options, -pd and -ps, that refer to
3288 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3289 "s", but I don't think that's a major issue.
3291 10. A number of refactoring changes to the code, none of which should affect
3294 (a) The number of logging options was getting close to filling up the
3295 32-bit word that was used as a bit map. I have split them into two classes:
3296 those that are passed in the argument to log_write(), and those that are
3297 only ever tested independently outside of that function. These are now in
3298 separate 32-bit words, so there is plenty of room for expansion again.
3299 There is no change in the user interface or the logging behaviour.
3301 (b) When building, for example, log lines, the code previously used a
3302 macro that called string_cat() twice, in order to add two strings. This is
3303 not really sufficiently general. Furthermore, there was one instance where
3304 it was actually wrong because one of the argument was used twice, and in
3305 one call a function was used. (As it happened, calling the function twice
3306 did not affect the overall behaviour.) The macro has been replaced by a
3307 function that can join an arbitrary number of extra strings onto a growing
3310 (c) The code for expansion conditions now uses a table and a binary chop
3311 instead of a serial search (which was left over from when there were very
3312 few conditions). Also, it now recognizes conditions like "pam" even when
3313 the relevant support is not compiled in: a suitably worded error message is
3314 given if an attempt is made to use such a condition.
3316 11. Added ${time_interval:xxxxx}.
3318 12. A bug was causing one of the ddress fields not to be passed back correctly
3319 from remote delivery subprocesses. The field in question was not being
3320 subsequently used, so this caused to problems in practice.
3322 13. Added new log selectors queue_time and deliver_time.
3324 14. Might have fixed a bug in maildirsizefile handling that threw up
3325 "unexpected character" debug warnings, and recalculated the data
3326 unnecessarily. In any case, I expanded the warning message to give more
3329 15. Added the message "Restricted characters in address" to the statements in
3330 the default ACL that block characters like @ and % in local parts.
3332 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3333 Three changes have been made:
3335 (a) There was a serious bug; a negative response to MAIL caused the whole
3336 recipient domain to be cached as invalid, thereby blocking all messages
3337 to all local parts at the same domain, from all senders. This bug has
3338 been fixed. The domain is no longer cached after a negative response to
3339 MAIL if the sender used is not empty.
3341 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3344 (c) A new callout option, "use_sender" has been added for people who want
3345 the modified behaviour.
3351 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3354 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3355 indeed breaks things for older releases.
3357 3. Added additional logging to the case where there is a problem reading data
3358 from a filter that is running in a subprocess using a pipe, in order to
3359 try to track down a specific problem.
3361 4. Testing facility fudge: when running in the test harness and attempting
3362 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3363 getting "No route to host". Convert this to a timeout.
3365 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3368 6. Some OS don't have socklen_t but use size_t instead. This affects the
3369 fifth argument of getsockopt() amongst other things. This is now
3370 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3371 can be set for individual OS. I have set it for SunOS5, OSF1, and
3372 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3373 some earlier ones do not.
3375 7. Change 4.30/15 was not doing the test caselessly.
3377 8. The standard form for an IPv6 address literal was being rejected by address
3378 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3379 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3380 this, as well as the form without the "IPv6" on the front (but only when
3381 address literals are enabled, of course).
3383 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3385 10. Exim crashed if a message with an empty sender address specified by -f
3386 encountered a router with an errors_to setting. This could be provoked only
3387 by a command such as
3391 where an empty string was supplied; "<>" did not hit this bug.
3393 11. Installed PCRE release 4.5.
3395 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3396 remained set. It is now erased.
3398 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3399 times from message ids (which are base 36 rather than the normal 62).
3401 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3402 were being counted as actual protocol errors, and logged if the log
3403 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3404 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3405 and DATA following a set of rejected RCPTs do not count as protocol errors.
3406 In other words, Exim assumes they were pipelined, though this may not
3407 actually be the case. Of course, in all cases the client gets an
3408 appropriate error code.
3410 15. If a lookup fails in an ACL condition, a message about the failure may
3411 be available; it is used if testing the ACL cannot continue, because most
3412 such messages specify what the cause of the deferral is. However, some
3413 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3414 that caused an old message to be retained and used if a later statement
3415 caused a defer, replacing the real cause of the deferral.
3417 16. If an IP address had so many PTR records that the DNS lookup buffer
3418 was not large enough to hold them, Exim could crash while trying to process
3419 the truncated data. It now detects and logs this case.
3421 17. Further to 4.21/58, another change has been made: if (and only if) the
3422 first line of a message (the first header line) ends with CRLF, a bare LF
3423 in a subsequent header line has a space inserted after it, so as not to
3424 terminate the header.
3426 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3427 unnecessarily, used atoi() instead of strtol(), and didn't check the
3428 termination when getting file sizes from file names by regex.
3430 19. Completely re-implemented the support for maildirsize files, in the light
3431 of a number of problems with the previous contributed implementation
3432 (4.30/29). In particular:
3434 . If the quota is zero, the maildirsize file is maintained, but no quota is
3437 . If the maildir directory does not exist, it is created before any attempt
3438 to write a maildirsize file.
3440 . The quota value in the file is just a cache; if the quota is changed in
3441 the transport, the new value overrides.
3443 . A regular expression is available for excluding directories from the
3446 20. The autoreply transport checks the characters in options that define the
3447 message's headers; it allows continued headers, but it was checking with
3448 isspace() after an embedded newline instead of explicitly looking for a
3451 21. If all the "regular" hosts to which an address was routed had passed their
3452 expiry times, and had not reached their retry times, the address was
3453 bounced, even if fallback hosts were defined. Now Exim should go on to try
3456 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3457 equivalent code in the SMTP transport. Some hosts send humungous responses
3458 to HELO/EHLO, more than 1024 it seems.
3460 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3461 gives compiler warnings in some environments. I've now done it "properly",
3464 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3465 (because of problems with the built-in one) was declared to return uschar *
3466 instead of char *, causing compiler failure.
3468 25. Fixed a file descriptor leak when processing alias/forward files.
3470 26. Fixed a minor format string issue in dbfn.c.
3472 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3474 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3475 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3478 29. When a sender address is verified, it is cached, to save repeating the test
3479 when there is more than one recipient in a message. However, when the
3480 verification involves a callout, it is possible for different callout
3481 options to be set for different recipients. It is too complicated to keep
3482 track of this in the cache, so now Exim always runs a verification when a
3483 callout is required, relying on the callout cache for the optimization.
3484 The overhead is duplication of the address routing, but this should not be
3487 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3488 to be verified with callout=postmaster, and the main callout worked but the
3489 postmaster check failed, the verification correctly failed. However, if a
3490 subsequent RCPT command asked for sender verification *without* the
3491 postmaster check, incorrect caching caused this verification also to fail,
3494 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3495 it was not caching the DNS options (qualify_single, search_parents) that
3496 were used when the lookup failed. A subsequent lookup with different
3497 options therefore always gave the same answer, though there were cases
3498 where it should not have. (Example: a "domains = !$mx_any" option on a
3499 dnslookup router: the "domains" option is always processed without any
3500 widening, but the router might have qualify_single set.) Now Exim uses the
3501 cached value only when the same options are set.
3503 32. Added John Jetmore's "exipick" utility to the distribution.
3505 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3506 than a timeout (e.g. a certificate is required, and is not provided), an
3507 Exim server now closes the connection immediately. Previously it waited for
3508 the client to close - but if the client is SSL, it seems that they each
3509 wait for each other, leading to a delay before one of them times out.
3511 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3512 maintained 0.8.x compatibility because I don't think many are using it, and
3513 it is clearly obsolete.
3515 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3518 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3519 client certificate was expired. A simple patch fixes this, though I don't
3520 understand the full logic of why the verify callback is called multiple
3523 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3524 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3525 which causes problems with some clients (such as the Certicom SSL Plus
3526 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3527 disables the coutermeasure allowing Eudora to connect."
3529 38. Exim was not checking that a write() to a log file succeeded. This could
3530 lead to Bad Things if a log got too big, in particular if it hit a file
3531 size limit. Exim now panics and dies if it cannot write to a log file, just
3532 as it does if it cannot open a log file.
3534 39. Modified OS/Makefile-Linux so that it now contains
3536 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3538 The two -D definitions ensure that Exim is compiled with large file
3539 support, which makes it possible to handle log files that are bigger than
3542 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3543 instance) a domain was checked against a named list that involved a lookup,
3544 causing $domain_data to be set, then another domain was checked against the
3545 same list, then the first domain was re-checked, the value of $domain_data
3546 after the final check could be wrong. In particular, if the second check
3547 failed, it could be set empty. This bug probably also applied to
3550 41. The strip_trailing_dot option was not being applied to the address given
3551 with the -f command-line option.
3553 42. The code for reading a message's header from the spool was incrementing
3554 $received_count, but never initializing it. This meant that the value was
3555 incorrect (doubled) while delivering a message in the same process in which
3556 it was received. In the most common configuration of Exim, this never
3557 happens - a fresh exec is done - but it can happen when
3558 deliver_drop_privilege is set.
3560 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3561 it now includes up to 150 characters of the unexpected data in the log
3564 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3565 and building data strings. The size of both of these buffers was 10 000
3566 bytes - far larger than anybody would *ever* want, thought I. Needless to
3567 say, somebody hit the limit. I have increased the maximum line length to
3568 20 000 and the maximum data length of concatenated lines to 100 000. I have
3569 also fixed two bugs, because there was no checking on these buffers. Tsk,
3570 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3571 buffer is too small.
3573 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3574 lsearch lookups. Now it does.
3576 46. When parsing a route_list item in a manualroute router, a fixed-length
3577 buffer was used for the list of hosts. I made this 1024 bytes long,
3578 thinking that nobody would ever have a list of hosts that long. Wrong.
3579 Somebody had a whole pile of complicated expansion conditions, and the
3580 string was silently truncated, leading to an expansion error. It turns out
3581 that it is easier to change to an unlimited length (owing to other changes
3582 that have happened since this code was originally written) than to build
3583 structure for giving a limitation error. The length of the item that
3584 expands into the list of hosts is now unlimited.
3586 47. The lsearch lookup could not handle data where the length of text line was
3587 more than 4095 characters. Such lines were truncated, leading to shortened
3588 data being returned. It should now handle lines of any length.
3590 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3591 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3594 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3595 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3596 no longer happens in comment lines. A list of replacements is now placed
3597 at the head of all of the source files, except those whose only change is
3598 to replace PERL_COMMAND in the very first #! line.
3600 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3601 messages on the queue, with a bottom-up merge sort, using code contributed
3602 by Michael Haardt. This should make operations like -bp somewhat faster on
3603 large queues. It won't affect queue runners, except when queue_run_in_order
3606 51. Installed eximstats 1.31 in the distribution.
3608 52. Added support for SRV lookups to the dnslookup router.
3610 53. If an ACL referred to $message_body or $message_body_end, the value was not
3611 reset for any messages that followed in the same SMTP session.
3613 54. The store-handling optimization for building very long strings was not
3614 differentiating between the different store pools. I don't think this
3615 actually made any difference in practice, but I've tidied it.
3617 55. While running the routers to verify a sender address, $sender_address
3618 was still set to the sender address. This is wrong, because when routing to
3619 send a bounce to the sender, it would be empty. Therefore, I have changed
3620 it so that, while verifying a sender address, $sender_address is set to <>.
3621 (There is no change to what happens when verifying a recipient address.)
3623 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3624 target A or AAAA records (if not already returned) without resetting the
3625 qualify_single or search_parents options of the DNS resolver. These are
3626 inappropriate in this case because the targets of MX and SRV records must
3627 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3628 target that, when qualified, matched something in the local domain. These
3629 two options are now turned off when doing these lookups.
3631 57. It seems that at least some releases of Reiserfs (which does not have the
3632 concept of a fixed number of inodes) returns zero and not -1 for the
3633 number of available inodes. This interacted badly with check_spool_inodes,
3634 which assumed that -1 was the "no such thing" setting. What I have done is
3635 to check that the total number of inodes is greater than zero before doing
3636 the test of how many are available.
3638 58. When a "warn" ACL statement has a log_message modifier, the message is
3639 remembered, and not repeated. This is to avoid a lot of repetition when a
3640 message has many recipients that cause the same warning to be written.
3641 Howewer, Exim was preserving the list of already written lines for an
3642 entire SMTP session, which doesn't seem right. The memory is now reset if a
3643 new message is started.
3645 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3646 debugging output unless log_rewrite was also set.
3648 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3649 of (int)(handle) when we know that handle contains (void *)(-1).
3651 61. The Exim daemon panic-logs an error return when it closes the incoming
3652 connection. However "connection reset by peer" seems to be common, and
3653 isn't really an error worthy of noting specially, so that particular error
3656 62. When Exim is trying to find all the local interfaces, it used to panic and
3657 die if the ioctl to get the interface flags failed. However, it seems that
3658 on at least one OS (Solaris 9) it is possible to have an interface that is
3659 included in the list of interfaces, but for which you get a failure error
3660 for this call. This happens when the interface is not "plumbed" into a
3661 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3662 failure of the "get flags" call assumes that the interface is down.
3664 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3665 makes life easier for people who are doing arithmetic on fields extracted
3666 from dates, where you often get leading zeros that should not be
3667 interpreted as octal.
3669 64. Added qualify_domain to the redirect router, to override the global
3672 65. If a pathologically long header line contained very many addresses (the
3673 report of this problem mentioned 10 000) and each of them was rewritten,
3674 Exim could use up a very large amount of memory. (It kept on making new
3675 copies of the header line as it rewrote, and never released the old ones.)
3676 At the expense of a bit more processing, the header rewriting function has
3677 been changed so that it no longer eats memory in this way.
3679 66. The generation of the Received: header has been moved from the time that a
3680 message starts to be received, to the time that it finishes. The timestamp
3681 in the Received: header should now be very close to that of the <= log
3682 line. There are two side-effects of this change:
3684 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3685 logged header lines no longer include the local Received: line, because
3686 it has not yet been created. The same applies to a copy of the message
3687 that is returned to a non-SMTP sender when a message is rejected.
3689 (b) When a filter file is tested using -bf, no additional Received: header
3690 is added to the test message. After some thought, I decided that this
3693 This change does not affect the value of $received_for. It is still set
3694 after address rewriting, but before local_scan() is called.
3696 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3698 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3699 gave an unhelpful panic error message, and a defer error. I have managed to
3700 change this behaviour so that it now rejects any supplied certificate,
3701 which seems right, as the list of acceptable certificates is empty.
3703 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3704 gave an unhelpful defer error. I have not managed to make this reject any
3705 supplied certificates, but the error message it gives is "no certificate
3706 supplied", which is not helpful.
3708 70. exigrep's output now also includes lines that are not associated with any
3709 message, but which match the given pattern. Implemented by a patch from
3710 Martin Sluka, which also tidied up the Perl a bit.
3712 71. Recipient callout verification, like sender verification, was using <> in
3713 the MAIL FROM command. This isn't really the right thing, since the actual
3714 sender may affect whether the remote host accepts the recipient or not. I
3715 have changed it to use the actual sender in the callout; this means that
3716 the cache record is now keyed on a recipient/sender pair, not just the
3717 recipient address. There doesn't seem to be a real danger of callout loops,
3718 since a callout by the remote host to check the sender would use <>.
3719 [SEE ABOVE: changed after hitting problems.]
3721 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3722 temporary errors. However, in the case of such a code being given after
3723 the end of a data transmission (i.e. after ".") Exim was failing to write
3724 a retry record for the message. (Yes, there was some broken host that was
3725 actually sending 8xx at this point.)
3727 73. An unknown lookup type in a host list could cause Exim to panic-die when
3728 the list was checked. (An example that provoked this was putting <; in the
3729 middle of a list instead of at the start.) If this happened during a DATA
3730 ACL check, a -D file could be left lying around. This kind of configuration
3731 error no longer causes Exim to die; instead it causes a defer errror. The
3732 incident is still logged to the main and panic logs.
3734 74. Buglet left over from Exim 3 conversion. The message "too many messages
3735 in one connection" was written to the rejectlog but not the mainlog, except
3736 when address rewriting (yes!) was being logged.
3738 75. Added write_rejectlog option.
3740 76. When a system filter was run not as root (that is, when system_filter_user
3741 was set), the values of the $n variables were not being returned to the
3742 main process; thus, they were not subsequently available in the $sn
3745 77. Added +return_path_on_delivery log selector.
3747 78. A connection timeout was being treated differently from recipients deferred
3748 when testing hosts_max_try with a message that was older than the host's
3749 retry timeout. (The host should not be counted, thus allowing all hosts to
3750 be tried at least once before bouncing.) This may have been the cause of an
3751 occasionally reported bug whereby a message would remain on the queue
3752 longer than the retry timeout, but would be bounced if a delivery was
3753 forced. I say "may" because I never totally pinned down the problem;
3754 setting up timeout/retry tests is difficult. See also the next item.
3756 79. The ultimate address timeout was not being applied to errors that involved
3757 a combination of host plus message (for example, a timeout on a MAIL
3758 command). When an address resolved to a number of possible hosts, and they
3759 were not all tried for each delivery (e.g. because of hosts_max_try), a
3760 message could remain on the queue longer than the retry timeout.
3762 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3765 81. Fixed an obscure SMTP outgoing bug which required at least the following
3766 conditions: (a) there was another message waiting for the same server;
3767 (b) the server returned 5xx to all RCPT commands in the first message so
3768 that the message was not completed; (c) the server dropped the connection
3769 or gave a negative response to the RSET that Exim sends to abort the
3770 transaction. The observed case was a dropped connection after DATA that had
3771 been sent in pipelining mode. That is, the server had advertised PIPELINING
3772 but was not implementing it correctly. The effect of the bug was incorrect
3773 behaviour, such as trying another host, and this could lead to a crash.
3779 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3780 and daemon.c were passed as pointers to ints; they should have been
3781 pointers to socklen_t variables (which are typically unsigned ints).
3783 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3786 3. Fixed a really odd bug that affected only the testing scheme; patching a
3787 certain fixed string in the binary changed the value of another string that
3788 happened to be identical to the end of the original first string.
3790 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3791 name", it returns that address as the IP address. On some operating
3792 systems (e.g. Solaris), it also passes back the IP address string as the
3793 "host name". However, on others (e.g. Linux), it passes back an empty
3794 string. Exim wasn't checking for this, and was changing the host name to an
3795 empty string, assuming it had been canonicized.
3797 5. Although rare, it is permitted to have more than one PTR record for a given
3798 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3799 all the names associated with an address, because they do in Solaris.
3800 However, it seems that they do not in Linux for data that comes from the
3801 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3802 I found this out when I moved to a new Linux workstation and tried to run
3803 the Exim test suite.
3805 To get round this problem I have changed the code so that it now does its
3806 own call to the DNS to look up PTR records when searching for a host name.
3807 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3808 addresses that are only in /etc/hosts are still found.
3810 This behaviour is, however, controlled by an option called host_lookup_
3811 order, which defaults to "bydns:byaddr". If people want to use the other
3812 order, or indeed, just use one or the other means of lookup, they can
3813 specify it in this variable.
3815 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3816 some operating systems, this comes back from gethostbyaddr() as an empty
3817 string, and this is what Exim used to test for. However, it seems that in
3818 other systems, "." is yielded. Exim now tests for this case too.
3820 7. The values of check_spool_space and check_log_space are now held internally
3821 as a number of kilobytes instead of an absolute number of bytes. If a
3822 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3823 kilobyte. This means that much larger values can be stored.
3825 8. Exim monitor: an attempt to get the action menu when not actually pointing
3826 at a message produces an empty menu entitled "No message selected". This
3827 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3828 no entries in it ("Shell widget menu has zero width and/or height"). So I
3829 have added a single, blank menu entry in this case.
3831 9. Added ${quote_local_part.
3833 10. MIME decoding is now applied to the contents of Subject: header lines when
3836 11. Now that a reference to $sender_host_address automatically causes a reverse
3837 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3838 host lookup before query-style lookups in lists that might use this
3839 variable. This has therefore been abolished, and the "net-" prefix is no
3840 longer necessary for query-style lookups.
3842 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3843 have been a typo for LFLAGS, so it has been changed.
3845 13. The install script calls Exim with "-C /dev/null" in order to find the
3846 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3847 to be output. Howeve, since Exim outputs its version number before the
3848 error, it didn't break the script. It just looked ugly. I fixed this by
3849 always allowing "-C /dev/null" if the caller is root.
3851 14. Ignore overlarge ACL variable number when reading spool file - insurance
3852 against a later release with more variables having written the file.
3854 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3855 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3856 this, as well as the form without the "IPv6" on the front.
3858 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3859 OS/Makefile-Darwin file.
3861 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3862 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3864 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3865 variable. However, some OS use ints for inodes. I've added cast to long int
3866 to get rid of the compiler warning.
3868 19. I had forgotten to lock out "/../" in configuration file names when
3869 ALT_CONFIG_PREFIX was set.
3871 20. Routers used for verification do not need to specify transports. However,
3872 if such a router generated a host list, and callout was configured, Exim
3873 crashed, because it could not find a port number from the (non-existent)
3874 transport. It now assumes port 25 in this circumstance.
3876 21. Added the -t option to exigrep.
3878 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3879 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3880 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3881 them is set, LOOKUP_LSEARCH is forced.
3883 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3885 24. Added sender and host information to the "rejected by local_scan()" log
3886 line; previously there was no indication of these.
3888 25. Added .include_if_exists.
3890 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3891 Linux. It turns out that not all file systems support this. Apparently some
3892 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3893 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3894 not supported on the file descriptor, is now ignored when Exim is trying to
3895 sync a directory. This applies only to Linux.
3897 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3899 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3900 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3903 29. Added support for maildirsize files from supplied patch (modified a bit).
3905 30. The use of :fail: followed by an empty string could lead Exim to respond to
3906 sender verification failures with (e.g.):
3908 550 Verification failed for <xxx>
3909 550 Sender verify failed
3911 where the first response line was missing the '-' that indicates it is not
3912 the final line of the response.
3914 31. The loop for finding the name of the user that called Exim had a hardwired
3915 limit of 10; it now uses the value of finduser_retries, which is used for
3916 all other user lookups.
3918 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3921 33. Exim was neglecting to zero errno before one call of strtol() when
3922 expanding a string and expecting an integer value. On some systems this
3923 resulted in spurious "integer overflow" errors. Also, it was casting the
3924 result into an int without checking.
3926 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3927 did not work. The code looks as if it has *never* worked, though it appears
3928 to have been documented since at least releast 1.62. I have made it work.
3930 35. The "timeout_DNS" error in retry rules, also documented since at least
3931 1.62, also never worked. As it isn't clear exactly what this means, and
3932 clearly it isn't a major issue, I have abolished the feature by treating it
3933 as "timeout", and writing a warning to the main and panic logs.
3935 36. The display of retry rules for -brt wasn't always showing the error code
3938 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3939 timeout_connect_A, timeout_connect_MX.
3941 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3942 to the empty sender.
3944 39. The daemon was not analysing the content of -oX till after it had closed
3945 stderr and disconnected from the controlling terminal. This meant that any
3946 syntax errors were only noted on the panic log, and the return code from
3947 the command was 0. By re-arranging the code a little, I've made the
3948 decoding happen first, so such errors now appear on stderr, and the return
3949 code is 1. However, the actual setting up of the sockets still happens in
3950 the disconnected process, so errors there are still only recorded on the
3953 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3954 connections (as happens on some IP stacks) was logged at start up time as
3955 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3956 it from "IPv6 and IPv4", which means that two separate sockets are being
3959 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3960 says whether AF_INET or AF_INET6 was passed as an argument.
3962 42. Exiwhat output was messed up when time zones were included in log
3965 43. Exiwhat now gives more information about the daemon's listening ports,
3966 and whether -tls-on-connect was used.
3968 44. The "port" option of the smtp transport is now expanded.
3970 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3971 silently ignored. Now an error message is written to the main and panic
3974 46. There's a new ACL modifier called "logwrite" which writes to a log file
3975 as soon as it is encountered.
3977 47. Added $local_user_uid and $local_user_gid at routing time.
3979 48. Exim crashed when trying to verify a sender address that was being
3982 49. Exim was recognizing only a space character after ".include". It now also
3983 recognizes a tab character.
3985 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3986 extracting the relevant information from the specification. The man page no
3987 longer contains scrambled data for the -d option, and I've added a section
3988 at the front about calling Exim under different names.
3990 51. Added "extra_headers" argument to the "mail" command in filter files.
3992 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3995 53. Installed eximstats 1.29.
3997 54. Added transport_filter_timeout as a generic transport option.
3999 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4000 Cc: header lines. This was required by RFC 822, but it not required by RFC
4003 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4004 incoming messages that did not have them. Now it does so only if the
4005 message originates locally, that is, if there is no associated remote host
4006 address. When Resent- header lines are present, this applies to the Resent-
4007 lines rather than the non-Resent- lines.
4009 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4010 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4012 58. Messages for configuration errors now include the name of the main
4013 configuration file - useful now that there may be more than one file in a
4014 list (.included file names were always shown).
4016 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4017 for those rare installations that do not start the daemon as root or run it
4018 setuid root. I've cut out the call to initgroups() if the daemon is not
4021 60. The Exim user and group can now be bound into the binary as text strings
4022 that are looked up at the start of Exim's processing.
4024 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4026 62. Added $mailstore_basename variable.
4028 63. Installed patch to sieve.c from Michael Haardt.
4030 64. When Exim failed to open the panic log after failing to open the main log,
4031 the original message it was trying to log was written to stderr and debug
4032 output, but if they were not available (the usual case in production), it
4033 was lost. Now it is written to syslog before the two lines that record the
4034 failures to open the logs.
4036 65. Users' Exim filters run in subprocesses under the user's uid. It is
4037 possible for a "deliver" command or an alias in a "personal" command to
4038 provoke an address rewrite. If logging of address rewriting is configured,
4039 this fails because the process is not running as root or exim. There may be
4040 a better way of dealing with this, but for the moment (because 4.30 needs
4041 to be released), I have disabled address rewrite logging when running a
4042 filter in a non-root, non-exim process.
4048 1. The buildconfig auxiliary program wasn't quoting the value set for
4049 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4050 not defined. This bug was masked in 4.22 by the effect that was fixed in
4053 2. Some messages that were rejected after a message id was allocated were
4054 shown as "incomplete" by exigrep. It no longer does this for messages that
4055 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4057 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4058 have allow_domain_literals set, the ID did not get logged in the <= line.
4059 Domain literals are now always recognized in Message-ID: header lines.
4061 4. The first argument for a ${extract expansion item is the key name or field
4062 number. Leading and trailing spaces in this item were not being ignored,
4063 causing some misleading effects.
4065 5. When deliver_drop_privilege was set, single queue runner processes started
4066 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4067 same command in the process it spins off) were not dropping privilege.
4069 6. When the daemon running as "exim" started a queue runner, it always
4070 re-executed Exim in the spun-off process. This is a waste of effort when
4071 deliver_drop_privilege is set. The new process now just calls the
4072 queue-runner function directly.
4078 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4081 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4082 ignored. Though the use of -oP was forcing the writing of a pid file, it
4083 was always written to the default place.
4085 3. If the message "no IP address found for host xxxx" is generated during
4086 incoming verification, it is now followed by identification of the incoming
4087 connection (so you can more easily find what provoked it).
4089 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4091 5. Added some features to "harden" Exim a bit more against certain attacks:
4093 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4094 be put in Local/Makefile. This is like the never_users runtime option,
4095 but it cannot be overridden. The default setting is "root".
4097 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4098 prefix string with which any file named in a -C command line option
4101 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4102 is retained for -C and -D only if the caller of Exim is root. Without
4103 it, the exim user may also use -C and -D and retain privilege.
4105 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4106 command line option is disabled.
4108 6. Macro names set by the -D option must start with an upper case letter, just
4109 like macro names defined in the configuration file.
4111 7. Added "dereference=" facility to LDAP.
4113 8. Two instances of the typo "uknown" in the source files are fixed.
4115 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4116 the Configure-Makefile script screwed up while processing it.
4118 10. Incorporated PCRE 4.4.
4120 11. The SMTP synchronization check was not operating right at the start of an
4121 SMTP session. For example, it could not catch a HELO sent before the client
4122 waited for the greeting. There is now a check for outstanding input at the
4123 point when the greeting is written. Because of the duplex, asynchronous
4124 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4125 way, but not yet received, when the check is performed.
4127 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4128 on TCP/IP sockets, because this apparently causes some broken clients to
4131 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4132 unchanged) from the Cygwin maintainer.
4134 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4135 was supported instead of testing for mbx. Effectively a typo.
4137 15. The spa authenticator server code was not checking that the input it
4138 received was valid base64.
4140 16. The debug output line for the "set" modifier in ACLs was not showing the
4141 name of the variable that was being set.
4143 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4145 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4146 lookup on its own. Something else was needed to trigger the lookup. For
4147 example, a match in host_lookup or the need for a host name in a host list.
4148 Now, if $sender_host_name is referenced and the host name has not yet been
4149 looked up, a lookup is performed. If the lookup fails, the variable remains
4150 empty, and $host_lookup_failed is set to "1".
4152 19. Added "eqi" as a case-independent comparison operator.
4154 20. The saslauthd authentication condition could segfault if neither service
4155 nor realm was specified.
4157 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4158 error message that was logged was misleading, and incoming SMTP
4159 connections were dropped. The message is now more accurate, and temporary
4160 errors are given to SMTP connections.
4162 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4163 (except RSET) with a 421 error, until QUIT is received. However, it was
4164 failing to send a response to QUIT.
4166 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4167 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4168 ACL is now run after the helo_try_verify_hosts code.
4170 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4171 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4172 case-independent, so other case variants are also recognized). Apparently
4173 some systems use these upper case variants.
4175 25. If more than two messages were waiting for the same host, and a transport
4176 filter was specified for the transport, Exim sent two messages over the
4177 same TCP/IP connection, and then failed with "socket operation on non-
4178 socket" when it tried to send the third.
4180 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4182 27. The extern definition of crypt16() in expand.c was not being excluded when
4183 the OS had its own crypt16() function.
4185 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4186 as a preferred synonym for return_size_limit, both as an option and as an
4189 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4191 30. Changed the default configuration ACL to relax the local part checking rule
4192 for addresses that are not in any local domains. For these addresses,
4193 slashes and pipe symbols are allowed within local parts, but the sequence
4194 /../ is explicitly forbidden.
4196 31. SPA server authentication was not clearing the challenge buffer before
4199 32. log_message in a "warn" ACL statement was writing to the reject log as
4200 well as to the main log, which contradicts the documentation and doesn't
4201 seem right (because no rejection is happening). So I have stopped it.
4203 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4204 However, I am unable to do any testing of this.
4206 34. Fixed an infelicity in the appendfile transport. When checking directories
4207 for a mailbox, to see if any needed to be created, it was accidentally
4208 using path names with one or more superfluous leading slashes; tracing
4209 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4211 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4212 contents are added to the log line that is written for every discarded
4213 recipient. (Previously a log_message setting was ignored.)
4215 36. The ${quote: operator now quotes the string if it is empty.
4217 37. The install script runs exim in order to find its version number. If for
4218 some reason other than non-existence or emptiness, which it checks, it
4219 could not run './exim', it was installing it with an empty version number,
4220 i.e. as "exim-". This error state is now caught, and the installation is
4223 38. An argument was missing from the function that creates an error message
4224 when Exim fails to connect to the socket for saslauthd authentication.
4225 This could cause Exim to crash, or give a corrupted message.
4227 39. Added isip, isip4, and isip6 to ${if conditions.
4229 40. The ACL variables $acl_xx are now saved with the message, and can be
4230 accessed later in routers, transports, and filters.
4232 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4233 strings in the file are not string-expanded.
4235 42. If a MAIL command specified a SIZE value that was too large to fit into an
4236 int variable, the check against message_size_limit failed. Such values are
4237 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4238 day this will have to be increased, but I don't think I want to be around
4239 when emails are that large.
4246 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4247 iconv() is not standard in FreeBSD.
4249 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4250 IPv6 enabled. The observed symptom was a segmentation fault on return from
4251 the function os_common_find_running_interfaces() in src/os.c.
4253 3. In the check_special_case() function in daemon.c I had used "errno" as an
4254 argument name, which causes warnings on some systems. This was basically a
4255 typo, since it was named "eno" in the comments!
4257 4. The code that waits for the clock to tick (at a resolution of some fraction
4258 of a second) so as to ensure message-id uniqueness was always waiting for
4259 at least one whole tick, when it could have waited for less. [This is
4260 almost certainly not relevant at current processor speeds, where it is
4261 unlikely to ever wait at all. But we try to future-proof.]
4263 5. The function that sleeps for a time interval that includes fractions of a
4264 second contained a race. It did not block SIGALRM between setting the
4265 timer, and suspending (a couple of lines later). If the interval was short
4266 and the sigsuspend() was delayed until after it had expired, the suspension
4267 never ended. On busy systems this could lead to processes getting stuck for
4270 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4271 process, before it forks any delivery processes. The open lookup caching
4272 mechanism meant that the open file or database connection was passed into
4273 the delivery process. The problem was that delivery processes always tidy
4274 up cached lookup data. This could cause a problem for the next delivery
4275 process started by the queue runner, because the external queue runner
4276 process does not know about the closure. So the next delivery process
4277 still has data in the lookup cache. In the case of a file lookup, there was
4278 no problem because closing a file descriptor in a subprocess doesn't affect
4279 the parent. However, if the lookup was caching a connection to a database,
4280 the connection was closed, and the second delivery process was likely to
4281 see errors such as "PGSQL: query failed: server closed the connection
4282 unexpectedly". The problem has been fixed by closing all cached lookups
4283 in a queue runner before running a delivery process.
4285 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4286 seem to have the "const" qualifier which it has on other OS. I've
4289 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4290 *of the same type* (client or server) with the same public name that an
4291 error should be diagnosed.
4293 9. When Exim looked up a host name for an IP address, but failed to find the
4294 original IP address when looking up the host name (a safety check), it
4295 output the message "<ip address> does not match any IP for NULL", which was
4296 confusing, to say the least. The bug was that the host name should have
4297 appeared instead of "NULL".
4299 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4300 user that is built into the binary, and the -C or -D options is used, root
4301 privilege is dropped before the configuration file is read. In addition,
4302 logging is switched to stderr instead of the normal log files. If the
4303 configuration then re-defines the Exim user, the unprivileged environment
4304 is probably not what is expected, so Exim logs a panic warning message (but
4307 However, if deliver_drop_privilege is set, the unprivileged state may well
4308 be exactly what is intended, so the warning has been cut out in that case,
4309 and Exim is allowed to try to write to its normal log files.
4315 1. smtp_return_error_details was not giving details for temporary sender
4316 or receiver verification errors.
4318 2. Diagnose a configuration error if two authenticators have the same public
4321 3. Exim used not to create the message log file for a message until the first
4322 delivery attempt. This could be confusing when incoming messages were held
4323 for policy or load reasons. The message log file is now created at the time
4324 the message is received, and an initial "Received" line is written to it.
4326 4. The automatically generated man page for command line options had a minor
4327 bug that caused no ill effects; however, a more serious problem was that
4328 the procedure for building the man page automatically didn't always
4329 operate. Consequently, release 4.20 contains an out-of-date version. This
4330 shouldn't happen again.
4332 5. When building Exim with embedded Perl support, the script that builds the
4333 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4334 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4336 6. The freeze_tell option was not being used for messages that were frozen on
4337 arrival, either by an ACL or by local_scan().
4339 7. Added the smtp_incomplete_transaction log selector.
4341 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4342 was accepting AUTH without a new EHLO.
4344 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4345 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4348 10. Logging of TCP/IP connections (when configured) now happens in the main
4349 daemon process instead of the child process, so that the TCP/IP connection
4350 count is more accurate (but it can never be perfect).
4352 11. The use of "drop" in a nested ACL was not being handled correctly in the
4353 outer ACL. Now, if condition failure induced by the nested "drop" causes
4354 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4355 or "require"), the connection is dropped.
4357 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4358 that yield "discard" can now be used with an "accept" or a "discard" verb,
4359 but an error is generated for any others (because I can't see a useful way
4360 to define what should happen).
4362 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4363 are now processed in the same way as lines in the Exim configuration file.
4364 In particular, continuation lines are supported.
4366 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4368 15. Added -ti meaning -t -i.
4370 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4371 domains, and warn by logging if others are found.
4373 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4374 returned from GIFCONF when Exim is trying to find the list of interfaces on
4375 a host. The code in os.c has been modified to copy each ifreq to an aligned
4376 structure in all cases.
4378 Also, in some cases, the returned ifreq's were being copied to a 'struct
4379 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4380 means the last couple of bytes of an IPv6 address could be chopped if the
4381 ifreq contained only a normal sockaddr (14 bytes storage).
4383 18. Named domain lists were not supported in the hosts_treat_as_local option.
4384 An entry such as +xxxx was not recognized, and was treated as a literal
4387 19. Ensure that header lines added by a DATA ACL are included in the reject log
4388 if the ACL subsequently rejects the message.
4390 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4391 MD5 (which is deprecated).
4393 21. When testing a filter file using -bf, Exim was writing a message when it
4394 took the sender from a "From " line in the message, but it was not doing so
4395 when it took $return_path from a Return-Path: header line. It now does.
4397 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4398 with a valid header line field name (a series of printing characters
4399 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4401 23. Changed "disc" in the source to "disk" to conform to the documentation and
4402 the book and for uniformity.
4404 24. Ignore Sendmail's -Ooption=value command line item.
4406 25. When execve() failed while trying to run a command in a pipe transport,
4407 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4408 could be confused with a return value of 69 from the command itself. This
4409 has been changed to 127, the value the shell returns if it is asked to run
4410 a non-existent command. The wording for the related log line suggests a
4411 non-existent command as the problem.
4413 26. If received_header_text expands to an empty string, do not add a Received:
4414 header line to the message. (Well, it adds a token one on the spool, but
4415 marks it "old" so that it doesn't get used or transmitted.)
4417 27. Installed eximstats 1.28 (addition of -nt option).
4419 28. There was no check for failure on the call to getsockname() in the daemon
4420 code. This can fail if there is a shortage of resources on the system, with
4421 ENOMEM, for example. A temporary error is now given on failure.
4423 29. Contrary to the C standard, it seems that in some environments, the
4424 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4425 program. Exim now does this explicitly; it affects the formatting of
4426 timestamps using strftime().
4428 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4429 complaints. I've now initialized all the variables, to avoid this.
4431 32. Header lines added by a system filter were not being "seen" during
4432 transport-time rewrites.
4434 33. The info_callback() function passed to OpenSSL is set up with type void
4435 (*)(SSL *, int, int), as described somewhere. However, when calling the
4436 function (actually a macro) that sets it up, the type void(*)() is
4437 expected. I've put in a cast to prevent warnings from picky compilers.
4439 34. If a DNS black list lookup found a CNAME record, but there were no A
4440 records associated with the domain it pointed at, Exim crashed.
4442 35. If a DNS black list lookup returned more than one A record, Exim ignored
4443 all but the first. It now scans all returned addresses if a particular IP
4444 value is being sought. In this situation, the contents of the
4445 $dnslist_value variable are a list of all the addresses, separated by a
4448 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4449 to accept a host name and all its aliases if the forward lookup for any of
4450 them yielded the IP address of the incoming connection. Now it accepts only
4451 those names whose forward lookup yields the correct IP address. Any other
4452 names are discarded. This closes a loophole whereby a rogue DNS
4453 administrator could create reverse DNS records to break through a
4454 wildcarded host restriction in an ACL.
4456 37. If a user filter or a system filter that ran in a subprocess used any of
4457 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4458 the wrong values were passed to the pipe command ($thisaddress had the
4459 value of $0, $0 had the value of $1, etc). This bug was introduced by
4460 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4462 38. Improved the line breaking for long SMTP error messages from ACLs.
4463 Previously, if there was no break point between 40 and 75 characters, Exim
4464 left the rest of the message alone. Two changes have been made: (a) I've
4465 reduced the minimum length to 35 characters; (b) if it can't find a break
4466 point between 35 and 75 characters, it looks ahead and uses the first one
4467 that it finds. This may give the occasional overlong line, but at least the
4468 remaining text gets split now.
4470 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4471 file descriptors might be low, and that setting 1000 would always raise it.
4472 It turns out that in some environments, the limit is already over 1000 and
4473 that lowering it causes trouble. So now Exim takes care not to decrease it.
4475 40. When delivering a message, the value of $return_path is set to $sender_
4476 address at the start of routing (routers may change the value). By an
4477 oversight, this default was not being set up when an address was tested by
4478 -bt or -bv, which affected the outcome if any router or filter referred to
4481 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4482 log only when "message" or "log_message" are set. However, if one of the
4483 conditions was an address verification, or a call to a nested ACL, the
4484 messages generated by the underlying test were being passed through. This
4485 no longer happens. The underlying message is available in $acl_verify_
4486 message for both "message" and "log_message" expansions, so it can be
4487 passed through if needed.
4489 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4490 new expansion $bh_ to give the encoded byte string without charset
4491 translation. Translation happens only if iconv() is available; HAVE_ICONV
4492 indicates this at build time. HEADERS_CHARSET gives the charset to
4493 translate to; headers_charset can change it in the configuration, and
4494 "headers charset" can change it in an individual filter file.
4496 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4497 that creates RFC 2047 encoded "words" labels them as that charset instead
4498 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4499 expansion operator; (ii) when Exim creates a From: line for a local
4500 message; (iii) when a header line is rewritten to include a "phrase" part.
4502 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4503 buggy, causing it to skip the first lines of messages whose message ID
4504 ended in 'D'. This would not have bitten before Exim release 4.14, because
4505 message IDs were unlikely to end in 'D' before then. The effect was to have
4506 incorrect size information for certain domains.
4508 45. #include "config.h" was missing at the start of the crypt16.c module. This
4509 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4512 46. If there was a timeout during a "random" callout check, Exim treated it as
4513 a failure of the random address, and carried on sending RSET and the real
4514 address. If the delay was just some slowness somewhere, the response to the
4515 original RCPT would be taken as a response to RSET and so on, causing
4516 mayhem of various kinds.
4518 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4519 when I implemented it. It didn't allow for the fact that some option values
4520 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4521 the right test for positive values.
4523 48. Domain names in DNS records are case-independent. Exim always looks them up
4524 in lower case. Some resolvers return domain names in exactly the case they
4525 appear in the zone file, that is, they may contain uppercase letters. Not
4526 all resolvers do this - some return always lower case. Exim was treating a
4527 change of case by a resolver as a change of domain, similar to a widening
4528 of a domain abbreviation. This triggered its re-routing code and so it was
4529 trying to route what was effectively the same domain again. This normally
4530 caused routing to fail (because the router wouldn't handle the domain
4531 twice). Now Exim checks for this case specially, and just changes the
4532 casing of the domain that it ultimately uses when it transmits the message
4535 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4538 50. If a filter generated a file delivery with a non-absolute name (possible if
4539 no home directory exists for the router), the forbid_file option was not
4542 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4543 the existing equality matching.
4545 52. Exim was using ints instead of ino_t variables in some places where it was
4546 dealing with inode numbers.
4548 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4549 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4550 called TMPDIR, and if it finds it is different, it changes its value.
4552 54. The smtp_printf() function is now made available to local_scan() so
4553 additional output lines can be written before returning. There is also an
4554 smtp_fflush() function to enable the detection of a dropped connection.
4555 The variables smtp_input and smtp_batched_input are exported to
4558 55. Changed the default runtime configuration: the message "Unknown user"
4559 has been removed from the ACL, and instead placed on the localuser router,
4560 using the cannot_route_message feature. This means that any verification
4561 failures that generate their own messages won't get overridden. Similarly,
4562 the "Unrouteable address" message that was in the ACL for unverifiable
4563 relay addresses has also been removed.
4565 56. Added hosts_avoid_esmtp to the smtp transport.
4567 57. The exicyclog script was not checking for the esoteric option
4568 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4569 will work only if exicyclog is run under the appropriate euid.
4571 58. Following a discussion on the list, the rules by which Exim recognises line
4572 endings on incoming messages have been changed. The -dropcr and drop_cr
4573 options are now no-ops, retained only for backwards compatibility. The
4574 following line terminators are recognized: LF CRLF CR. However, special
4575 processing applies to CR:
4577 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4578 nor a local message in the state where . is a terminator.
4580 (ii) If a bare CR is encountered in a header line, an extra space is added
4581 after the line terminator so as not to end the header. The reasoning
4582 behind this is that bare CRs in header lines are most likely either
4583 to be mistakes, or people trying to play silly games.
4585 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4586 was being incorrectly given as 18 bytes larger than it should have been.
4587 This is a VOB (very old bug).
4589 60. This may never have affected anything current, but just in case it has:
4590 When the local host is found other than at the start of a list of hosts,
4591 the local host, those with the same MX, and any that follow, are discarded.
4592 When the list in question was part of a longer list of hosts, the following
4593 hosts (not currently being processed) were also being discarded. This no
4594 longer happens. I'm not sure if this situation could ever has previously
4597 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4600 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4601 header. This is something that is recommended in a new Internet Draft, and
4602 is something that is documented as being done by Sendmail. There are two
4603 possible values. For messages generated by the autoreply transport, Exim
4606 Auto-Submitted: auto-replied
4608 whereas for all other generated messages (e.g. bounces) it adds
4610 Auto-Submitted: auto-generated
4612 63. The "personal" condition in filters now includes a test for the
4613 Auto-Submitted: header. If it contains the string "auto-" the message it
4614 not considered personal.
4616 64. Added rcpt_include_affixes as a generic transport option.
4618 65. Added queue_only_override (default true).
4620 66. Added the syslog_duplication option.
4622 67. If what should have been the first header line of a message consisted of
4623 a space followed by a colon, Exim was mis-interpreting it as a header line.
4624 It isn't of course - it is syntactically invalid and should therefore be
4625 treated as the start of the message body. The misbehaviour could have
4626 caused a number of strange effects, including loss of data in subsequent
4627 header lines, and spool format errors.
4629 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4630 client host had authenticated. This control can now be exercised by an ACL
4631 for more flexibility.
4633 69. By default, callouts do not happen when testing with -bh. There is now a
4634 variant, -bhc, which does actually run the callout code, including
4635 consulting and updating the callout cache.
4637 70. Added support for saslauthd authentication, courtesy of Alexander
4640 71. If statvfs() failed on the spool or log directories while checking their
4641 size for availability, Exim confusingly gave the error "space shortage".
4642 Furthermore, in debugging mode it crashed with a floating point exception.
4643 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4644 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4645 serious problem, Exim now writes to the main and panic logs when this
4646 happens, with details of the failure. It then refuses to accept the
4647 incoming message, giving the message "spool directory problem" or "log
4648 directory problem" with a 421 code for SMTP messages.
4650 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4651 0, 1, and 2 exist, because some OS complain for execs without them (see
4652 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4653 descriptors. However, the code omitted to check that the open succeeded,
4654 causing mysterious errors if for some reason the permissions on /dev/null
4655 got screwed. Now Exim writes a message to the main and panic logs, and
4656 bombs out if it can't open /dev/null.
4658 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4659 interact so that it is all more flexible. It is supposed to remain
4660 backwards compatible. Also added extra_local_interfaces.
4662 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4663 to bomb out with an assertion failure - to the client this appears as a
4664 connection drop. This problem occurs in the part of the code that was taken
4665 from the Samba project. Fortunately, the assertion is in a very simple
4666 function, so I have fixed this by reproducing the function inline in the
4667 one place where it is called, and arranging for authentication to fail
4668 instead of killing the process with assert().
4670 75. The SPA client code was not working when the server requested OEM rather
4671 than Unicode encoding.
4673 76. Added code to make require_files with a specific uid setting more usable in
4674 the case where statting the file as root fails - usually a non-root-mounted
4675 NFS file system. When this happens and the failure is EACCES, Exim now
4676 forks a subprocess and does the per-uid checking as the relevant uid.
4678 77. Added process_log_path.
4680 78. If log_file_path was not explicitly set, a setting of check_log_space or
4681 check_log_inodes was ignored.
4683 79. If a space check for the spool or log partitions fails, the incident is now
4684 logged. Of course, in the latter case the data may get lost...
4686 80. Added the %p formatting code to string_format() so that it can be used to
4687 print addresses in debug_print(). Adjusted all the address printing in the
4688 debugging in store.c to use %p rather than %d.
4690 81. There was a concern that a line of code in smtp_in.c could overflow a
4691 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4692 initially expressed, the concern was not well-founded, because trailing
4693 spaces are removed early. However, if the trailing spaces were followed by
4694 a NULL, they did not get removed, so the overflow was possible. Two fixes
4697 (a) I re-wrote the offending code in a cleaner fashion.
4698 (b) If an incoming SMTP command contains a NULL character, it is rejected
4701 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4702 runs initgroups(), so that if the Exim user is in any additional groups,
4703 they will be used during message reception.
4709 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob