1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.498 2007/03/21 15:15:12 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
193 PH/42 Another patch from the Sieve maintainer.
199 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
200 fixed by 4.65/MH/01 (is this a record?) are fixed:
202 (i) An empty string was always treated as zero by the numeric comparison
203 operators. This behaviour has been restored.
205 (ii) It is documented that the numeric comparison operators always treat
206 their arguments as decimal numbers. This was broken in that numbers
207 starting with 0 were being interpreted as octal.
209 While fixing these problems I realized that there was another issue that
210 hadn't been noticed. Values of message_size_limit (both the global option
211 and the transport option) were treated as octal if they started with 0.
212 The documentation was vague. These values are now always treated as
213 decimal, and I will make that clear in the documentation.
219 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
220 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
223 MH/01 Don't check that the operands of numeric comparison operators are
224 integers when their expansion is in "skipping" mode (fixes bug
225 introduced by 4.64-PH/07).
227 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
228 child addresses, Exim now panics and dies. Previously, because the count
229 is held in a short int, deliveries were likely to be lost. As such a
230 large number of recipients for a single message is ridiculous
231 (performance will be very, very poor), I have chosen to impose a limit
232 rather than extend the field.
238 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
239 leftover -K file (the existence of which was triggered by #402).
240 While we were at it, introduced process PID as part of the -K
241 filename. This should rule out race conditions when creating
244 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
245 processing considerably. Previous code took too long for large mails,
246 triggering a timeout which in turn triggers #401.
248 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
249 in the DK code in transports.c. sendfile() is not really portable,
250 hence the _LINUX specificness.
252 TF/01 In the add_headers option to the mail command in an Exim filter,
253 there was a bug that Exim would claim a syntax error in any
254 header after the first one which had an odd number of characters
257 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
258 callout verification, Exim cached a "reject" for the entire domain. This
259 is correct for most verifications, but it is not correct for a recipient
260 verification with use_sender or use_postmaster set, because in that case
261 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
262 case of MAIL FROM:<> rejection from other early rejections (e.g.
263 rejection of HELO). When verifying a recipient using a non-null MAIL
264 address, the cache is ignored if it shows MAIL FROM:<> rejection.
265 Whatever the result of the callout, the value of the domain cache is
266 left unchanged (for any other kind of callout, getting as far as trying
267 RCPT means that the domain itself is ok).
269 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
272 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
273 manifest itself as EPIPE rather than ECONNECT. When tidying away a
274 session, the daemon ignores ECONNECT errors and logs others; it now
275 ignores EPIPE as well.
277 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
278 (quoted-printable decoding).
280 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
281 later the small subsequent patch to fix an introduced bug.
283 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
285 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
287 PH/08 An error is now given if message_size_limit is specified negative.
289 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
290 to be given (somewhat) arbitrary names.
292 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
295 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
296 miscellaneous code fixes
298 PH/10 Added the log_reject_target ACL modifier to specify where to log
301 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
302 hostname. This is wrong, because it relates to the incoming message (and
303 probably the interface on which it is arriving) and not to the outgoing
304 callout (which could be using a different interface). This has been
305 changed to use the value of the helo_data option from the smtp transport
306 instead - this is what is used when a message is actually being sent. If
307 there is no remote transport (possible with a router that sets up host
308 addresses), $smtp_active_hostname is used.
310 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
311 tweaks were necessary in order to get it to work (see also 21 below):
312 (a) The code assumed that strncpy() returns a negative number on buffer
313 overflow, which isn't the case. Replaced with Exim's string_format()
315 (b) There were several signed/unsigned issues. I just did the minimum
316 hacking in of casts. There is scope for a larger refactoring.
317 (c) The code used strcasecmp() which is not a standard C function.
318 Replaced with Exim's strcmpic() function.
319 (d) The code set only $1; it now sets $auth1 as well.
320 (e) A simple test gave the error "authentication client didn't specify
321 service in request". It would seem that Dovecot has changed its
322 interface. Fortunately there's a specification; I followed it and
323 changed what the client sends and it appears to be working now.
325 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
328 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
329 address A is aliased to B and C, where B exists and C does not. Without
330 -v the output is "A verified" because verification stops after a
331 successful redirection if more than one address is generated. However,
332 with -v the child addresses are also verified. Exim was outputting "A
333 failed to verify" and then showing the successful verification for C,
334 with its parentage. It now outputs "B failed to verify", showing B's
335 parentage before showing the successful verification of C.
337 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
338 look up a TXT record in a specific list after matching in a combined
341 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
342 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
343 they consult the DNS. I had assumed they would set it the way they
344 wanted; and indeed my experiments on Linux seem to show that in some
345 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
346 To be on the safe side, however, I have now made the interface to
347 host_find_byname() similar to host_find_bydns(), with an argument
348 containing the DNS resolver options. The host_find_byname() function now
349 sets these options at its start, just as host_find_bydns() does. The smtp
350 transport options dns_qualify_single and dns_search_parents are passed to
351 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
352 of host_find_byname() use the default settings of RES_DEFNAMES
353 (qualify_single) but not RES_DNSRCH (search_parents).
355 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
356 spool_read_header() do less string testing, by means of a preliminary
357 switch on the second character of optional "-foo" lines. (This is
358 overdue, caused by the large number of possibilities that now exist.
359 Originally there were few.) While I was there, I also converted the
360 str(n)cmp tests so they don't re-test the leading "-" and the first
361 character, in the hope this might squeeze out yet more improvement.
363 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
364 flag allowing group syntax was set by the header_syntax check but not
365 turned off, possible causing trouble later; (2) The flag was not being
366 set at all for the header_verify test, causing "group"-style headers to
367 be rejected. I have now set it in this case, and also caused header_
368 verify to ignore an empty address taken from a group. While doing this, I
369 came across some other cases where the code for allowing group syntax
370 while scanning a header line wasn't quite right (mostly, not resetting
371 the flag correctly in the right place). These bugs could have caused
372 trouble for malformed header lines. I hope it is now all correct.
374 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
375 with the "reply" argument non-NULL. The code, however (which originally
376 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
377 but it didn't always do it. This confused somebody who was copying the
378 code for some other use. I have removed all the tests.
380 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
381 feature that was used to support insecure browsers during the U.S. crypto
382 embargo. It requires special client support, and Exim is probably the
383 only MTA that supported it -- and would never use it because real RSA is
384 always available. This code has been removed, because it had the bad
385 effect of slowing Exim down by computing (never used) parameters for the
386 RSA_EXPORT functionality.
388 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
389 authenticator to fail if there's a tab character in the incoming data
390 (there should never be unless someone is messing about, as it's supposed
391 to be base64-encoded). Also added, on Timo's advice, the "secured" option
392 if the connection is using TLS or if the remote IP is the same as the
393 local IP, and the "valid-client-cert option" if a client certificate has
396 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
397 authenticators. This can be used for authorization after authentication
398 succeeds. (In the case of plaintext, it servers for both authentication
401 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
402 if any retry times were supplied.
404 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
405 connection, where there is no client IP address to check. In this
406 situation, the verify now always succeeds.
408 PH/25 Applied John Jetmore's -Mset patch.
410 PH/26 Added -bem to be like -Mset, but loading a message from a file.
412 PH/27 In a string expansion for a processed (not raw) header when multiple
413 headers of the same name were present, leading whitespace was being
414 removed from all of them, but trailing whitespace was being removed only
415 from the last one. Now trailing whitespace is removed from each header
416 before concatenation. Completely empty headers in a concatenation (as
419 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
420 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
422 PH/29 [Removed. This was a change that I later backed out, and forgot to
423 correct the ChangeLog entry (that I had efficiently created) before
424 committing the later change.]
426 PH/30 Exim was sometimes attempting to deliver messages that had suffered
427 address errors (4xx response to RCPT) over the same connection as other
428 messages routed to the same hosts. Such deliveries are always "forced",
429 so retry times are not inspected. This resulted in far too many retries
430 for the affected addresses. The effect occurred only when there were more
431 hosts than the hosts_max_try setting in the smtp transport when it had
432 the 4xx errors. Those hosts that it had tried were not added to the list
433 of hosts for which the message was waiting, so if all were tried, there
434 was no problem. Two fixes have been applied:
436 (i) If there are any address or message errors in an SMTP delivery, none
437 of the hosts (tried or untried) are now added to the list of hosts
438 for which the message is waiting, so the message should not be a
439 candidate for sending over the same connection that was used for a
440 successful delivery of some other message. This seems entirely
441 reasonable: after all the message is NOT "waiting for some host".
442 This is so "obvious" that I'm not sure why it wasn't done
443 previously. Hope I haven't missed anything, but it can't do any
444 harm, as the worst effect is to miss an optimization.
446 (ii) If, despite (i), such a delivery is accidentally attempted, the
447 routing retry time is respected, so at least it doesn't keep
448 hammering the server.
450 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
451 in ${readsocket because some servers need this prod.
453 PH/32 Added some extra debug output when updating a wait-xxx database.
455 PH/33 The hint "could be header name not terminated by colon", which has been
456 given for certain expansion errors for a long time, was not being given
457 for the ${if def:h_colon_omitted{... case.
459 PH/34 The spec says: "With one important exception, whenever a domain list is
460 being scanned, $domain contains the subject domain." There was at least
461 one case where this was not true.
463 PH/35 The error "getsockname() failed: connection reset by peer" was being
464 written to the panic log as well as the main log, but it isn't really
465 panic-worthy as it just means the connection died rather early on. I have
466 removed the panic log writing for the ECONNRESET error when getsockname()
469 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
470 runs only) independently of the message's sender address. This meant
471 that, if the 4xx error was in fact related to the sender, a different
472 message to the same recipient with a different sender could confuse
473 things. In particualar, this can happen when sending to a greylisting
474 server, but other circumstances could also provoke similar problems.
475 I have changed the default so that the retry time for these errors is now
476 based a combination of the sender and recipient addresses. This change
477 can be overridden by setting address_retry_include_sender=false in the
480 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
481 remote server are returned as part of bounce messages. This was not
482 happening for LMTP over a pipe (the lmtp transport), but now it is the
483 same for both kinds of LMTP.
485 PH/38 Despite being documented as not happening, Exim was rewriting addresses
486 in header lines that were in fact CNAMEs. This is no longer the case.
488 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
489 and queue runs started by the daemon processed all messages. This has
490 been fixed so that -R and -S can now usefully be given with -q<time>.
492 PH/40 Import PCRE release 6.7 (fixes some bugs).
494 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
496 PH/42 Give an error if -q is specified more than once.
498 PH/43 Renamed the variables $interface_address and $interface_port as
499 $received_ip_address and $received_port, to make it clear that these
500 values apply to message reception, and not to the outgoing interface when
501 a message is delivered. (The old names remain recognized, of course.)
503 PH/44 There was no timeout on the connect() call when using a Unix domain
504 socket in the ${readsocket expansion. There now is.
506 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
507 be meaningful with "accept".
509 SC/01 Eximstats V1.43
510 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
512 SC/02 Eximstats V1.44
513 Use a glob alias rather than an array ref in the generated
514 parser. This improves both readability and performance.
516 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
517 Collect SpamAssassin and rejection statistics.
518 Don't display local sender or destination tables unless
519 there is data to show.
520 Added average volumes into the top table text output.
522 SC/04 Eximstats V1.46
523 Collect data on the number of addresses (recipients)
524 as well as the number of messages.
526 SC/05 Eximstats V1.47
527 Added 'Message too big' to the list of mail rejection
528 reasons (thanks to Marco Gaiarin).
530 SC/06 Eximstats V1.48
531 Mainlog lines which have GMT offsets and are too short to
532 have a flag are now skipped.
534 SC/07 Eximstats V1.49 (Alain Williams)
535 Added the -emptyok flag.
537 SC/08 Eximstats V1.50
538 Fixes for obtaining the IP address from reject messages.
540 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
541 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
542 whitesspace changes from 4.64-PH/27
544 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
547 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
548 are found, allow negative numbers in numeric criteria)
550 JJ/06 exipick.20061117.2, added new $message_body_missing variable
552 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
553 to match changes made in 4.64-PH/43
555 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
557 PH/47 Put in an explicit test for a DNS lookup of an address record where the
558 "domain" is actually an IP address, and force a failure. This locks out
559 those revolvers/nameservers that support "A-for-A" lookups, in
560 contravention of the specifications.
562 PH/48 When a host name was looked up from an IP address, and the subsequent
563 forward lookup of the name timed out, the host name was left in
564 $sender_host_name, contrary to the specification.
566 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
567 restricted to single-key lookups, Exim was not diagnosing an error if
568 * or *@ was used with a query-style lookup.
570 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
572 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
573 long ago, but noone interested enough thought of it. Let's just say that
574 the "1.1" means that there are some new functions that weren't there at
575 some point in the past.
577 PH/51 Error processing for expansion failure of helo_data from an smtp
578 transport during callout processing was broken.
580 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
581 tested/used via the -bh/-bhc/-bs options.
583 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
584 bug, fixed in subsequent PCRE releases).
586 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
587 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
589 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
595 SC/01 Use a glob alias rather than an array ref in eximstats generated
596 parser. This improves both readability and performance.
598 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
599 Don't display local sender or destination tables in eximstats unless
600 there is data to show.
601 Added average volumes into the eximstats top table text output.
603 SC/03 Collect data on the number of addresses (recipients) as well
604 as the number of messages in eximstats.
606 TF/01 Correct an error in the documentation for the redirect router. Exim
607 does (usually) call initgroups() when daemonizing.
609 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
610 with consistent privilege compared to when running as a daemon.
612 TF/03 Note in the spec that $authenticated_id is not set for local
613 submissions from trusted users.
615 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
616 Thanks to Dean Brooks <dean@iglou.com> for the patch.
618 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
619 by adding some example configuration directives to the default
620 configuration file. A little bit of work is required to uncomment the
621 directives and define how usernames and passwords are checked, but
622 there is now a framework to start from.
624 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
625 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
626 without this. I don't know how relevant this is to other LDAP libraries.
628 PH/02 Add the verb name to the "unknown ACL verb" error.
630 PH/03 Magnus Holmgren's patch for filter_prepend_home.
632 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
634 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
635 directory not expanded when it should be if an expanded home directory
636 was set for the address (which is overridden by the transport).
638 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
641 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
642 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
643 because it is too late at that time, and has no effect.
645 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
646 security issue with \' (bugzilla #107). I could not use the
647 PQescapeStringConn() function, because it needs a PGconn value as one of
650 PH/08 When testing addresses using -bt, indicate those final addresses that
651 are duplicates that would not cause an additional delivery. At least one
652 person was confused, thinking that -bt output corresponded to deliveries.
653 (Suppressing duplicates isn't a good idea as you lose the information
654 about possibly different redirections that led to the duplicates.)
656 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
657 systems where poll() doesn't work, in particular OS X.
659 PH/10 Added more information to debugging output for retry time not reached.
661 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
662 operations in malware.c.
664 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
667 PH/13 If write_rejectlog was set false when logging was sent to syslog with
668 syslog_duplication set false, log lines that would normally be written
669 both the the main log and to the reject log were not written to syslog at
672 PH/14 In the default configuration, change the use of "message" in ACL warn
673 statements to "add_header".
675 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
676 not followed by a command (e.g. "seen endif").
678 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
679 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
682 PH/17 Added extra conditions to the default value of delay_warning_condition
686 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
687 { match{$h_precedence:}{(?i)bulk|list|junk} } \
688 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
691 The Auto-Submitted: and various List- headers are standardised, whereas I
692 don't think Precedence: ever was.
694 PH/18 Refactored debugging code in route_finduser() to show more information,
695 in particular, the error code if getpwnam() issues one.
697 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
698 This is apparently needed in addition to the PH/07 change above to avoid
699 any possible encoding problems.
701 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
702 but not after initializing Perl.
704 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
705 output them only if debugging. By default they are written stderr,
706 apparently, which is not desirable.
708 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
711 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
714 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
716 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
717 authenticated or an ident call has been made. Suppress the default
718 values for $authenticated_id and $authenticated_sender (but permit -oMai
719 and -oMas) when testing with -bh.
721 PH/24 Re-jigged the order of the tests in the default configuration so that the
722 tests for valid domains and recipients precede the DNS black list and CSA
723 tests, on the grounds that those ones are more expensive.
725 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
726 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
727 This bug exists in every version of Exim that I still have, right back to
730 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
731 However, an attempt to turn on case-sensitivity in a regex key by
732 including (?-i) didn't work because the subject string was already
733 lowercased, and the effects were non-intuitive. It turns out that a
734 one-line patch can be used to allow (?-i) to work as expected.
740 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
741 other effects) broke the use of negated acl sub-conditions.
743 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
746 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
747 "Deny" causes Exim to reject the incoming connection with a 554 error.
748 Unfortunately, if there is a major crisis, such as a disk failure,
749 tcp-wrappers gives "deny", whereas what one would like would be some
750 kind of temporary error. A kludge has been added to help with this.
751 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
752 554 error is used if errno is still zero or contains ENOENT (which occurs
753 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
756 PH/03 Add -lutil to the default FreeBSD LIBS setting.
758 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
759 errors. Otherwise a message that provokes a temporary error (when other
760 messages do not) can cause a whole host to time out.
762 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
763 addresses were routed directly to files or pipes from a redirect router.
764 File deliveries just didn't batch; pipe deliveries might have suffered
767 PH/06 A failure to get a lock for a hints database would erroneously always say
768 "Failed to get write lock", even when it was really a read lock.
770 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
771 of 0600. This has been changed to use the value of the lockfile_mode
772 option (which defaults to 0600).
774 PH/08 Applied small patch from the Sieve maintainer.
776 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
777 folder from quota calculations, a direct delivery into this folder messed
778 up the contents of the maildirsize file. This was because the regex was
779 used only to exclude .Trash (or whatever) when the size of the mailbox
780 was calculated. There was no check that a delivery was happening into an
781 excluded directory. This bug has been fixed by ignoring all quota
782 processing for deliveries into excluded directories.
784 PH/10 Added the maildirfolder_create_regex option to appendfile.
790 PH/01 The code for finding all the local interface addresses on a FreeBSD
791 system running IPv6 was broken. This may well have applied to all BSD
792 systems, as well as to others that have similar system calls. The broken
793 code found IPv4 interfaces correctly, but gave incorrect values for the
794 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
795 that it would not match correctly against @[] and not recognize the IPv6
798 PH/02 The ipliteral router was not recognizing addresses of the form user@
799 [ipv6:....] because it didn't know about the "ipv6:" prefix.
801 PH/03 Added disable_ipv6.
803 PH/04 Changed $reply_address to use the raw form of the headers instead of the
804 decoded form, because it is most often used to construct To: headers
805 lines in autoreplies, and the decoded form may well be syntactically
806 invalid. However, $reply_address has leading white space removed, and all
807 newlines turned into spaces so that the autoreply transport does not
810 PH/05 If group was specified without a user on a router, and no group or user
811 was specified on a transport, the group from the router was ignored.
813 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
814 for visible compile-time settings that can be used to change these
815 numbers, for those that want even more. Backwards compatibility with old
816 spool files has been maintained. However, going back to a previous Exim
817 release will lost any variables that are in spool files.
819 PH/07 Two small changes when running in the test harness: increase delay when
820 passing a TCP/IP connection to a new process, in case the original
821 process has to generate a bounce, and remove special handling of
822 127.0.0.2 (sic), which is no longer necessary.
824 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
825 be the same on different OS.
827 PH/09 Moved a debug statement in filter processing to avoid a race problem when
830 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
831 whether --show-vars was specified or not
833 JJ/02 exipick: Added support for new ACL variable spool format introduced
836 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
837 syntactically invalid From: or Reply-to: line, and a filter used this to
838 generate an autoreply, and therefore failed to obtain an address for the
839 autoreply, Exim could try to deliver to a non-existent relative file
840 name, causing unrelated and misleading errors. What now happens is that
841 it logs this as a hard delivery error, but does not attempt to create a
844 PH/11 The exinext utility has a -C option for testing purposes, but although
845 the given file was scanned by exinext itself; it wasn't being passed on
846 when Exim was called.
848 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
849 an end-of-file indication when reading a command response.
851 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
852 compiled. In many other places in Exim, IPv6 addresses are always
853 recognized, so I have changed this. It also means that IPv4 domain
854 literals of the form [IPV4:n.n.n.n] are now always recognized.
856 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
857 used if the router is not running as root, for example, when verifying at
858 ACL time, or when using -bh. The debugging output from this situation was
859 non-existent - all you got was a failure to exec. I have made two
862 (a) Failures to set uid/gid, the current directory, or a process leader
863 in a subprocess such as that created by queryprogram now generate
864 suitable debugging ouput when -d is set.
866 (b) The queryprogram router detects when it is not running as root,
867 outputs suitable debugging information if -d is set, and then runs
868 the subprocess without attempting to change uid/gid.
870 PH/15 Minor change to Makefile for building test_host (undocumented testing
873 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
874 additional section of a DNS packet that returns MX or SRV records.
875 Instead, it always explicitly searches for A/AAAA records. This avoids
876 major problems that occur when a DNS server includes only records of one
877 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
878 fixed another bug: if SRV records were looked up and the corresponding
879 address records were *not* found in the additional section, the port
880 values from the SRV records were lost.
882 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
883 using the correct key (the original address) when searching the retry
884 rules in order to find which one to use for generating the retry hint.
886 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
887 adding the default one. Similarly, if it contains a Reply-To: header, the
888 errors_reply_to option, if set, is not used.
890 PH/19 When calculating a retry time, Exim used to measure the "time since
891 failure" by looking at the "first failed" field in the retry record. Now
892 it does not use this if it is later than than the arrival time of the
893 message. Instead it uses the arrival time. This makes for better
894 behaviour in cases where some deliveries succeed, thus re-setting the
895 "first failed" field. An example is a quota failure for a huge message
896 when small messages continue to be delivered. Without this change, the
897 "time since failure" will always be short, possible causing more frequent
898 delivery attempts for the huge message than are intended.
899 [Note: This change was subsequently modified - see PH/04 for 4.62.]
901 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
902 $1, $2, $3) because the numerical variables can be reset during some
903 expansion items (e.g. "match"), thereby losing the authentication data.
905 PH/21 Make -bV show the size of off_t variables so that the test suite can
906 decide whether to run tests for quotas > 2G.
908 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
909 mailbox_size, and mailbox_filecount in the appendfile transport. If a
910 filecount value is greater than 2G or if a quota value is greater than 2G
911 on a system where the size of off_t is not greater than 4, a panic error
914 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
915 never match. The debug and -bh output now contains an explicit error
916 message indicating a malformed IPv4 address or mask.
918 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
919 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
922 PH/25 Do not write to syslog when running in the test harness. The only
923 occasion when this arises is a failure to open the main or panic logs
924 (for which there is an explicit test).
926 PH/26 Added the /no_tell option to "control=freeze".
928 PH/27 If a host name lookup failed very early in a connection, for example, if
929 the IP address matched host_lookup and the reverse lookup yielded a name
930 that did not have a forward lookup, an error message of the form "no IP
931 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
932 could be logged. Now it outputs the IP address instead of "NULL".
934 PH/28 An enabling patch from MH: add new function child_open_exim2() which
935 allows the sender and the authenticated sender to be set when
936 submitting a message from within Exim. Since child_open_exim() is
937 documented for local_scan(), the new function should be too.
939 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
940 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
941 results in an empty string is now treated as unset.
943 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
945 PH/31 Added sender_verify_fail logging option.
947 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
948 needed by RFC 822 but not by RFC 2822 was commented out. I have now
949 tidied the source and removed it altogether.
951 PH/33 When a queue run was abandoned because the load average was too high, a
952 log line was always written; now it is written only if the queue_run log
953 selector is set. In addition, the log line for abandonment now contains
954 information about the queue run such as the pid. This is always present
955 in "start" and "stop" lines but was omitted from the "abandon" line.
957 PH/34 Omit spaces between a header name and the colon in the error message that
958 is given when verify = headers_syntax fails (if there are lots of them,
959 the message gets confusing).
961 PH/35 Change the default for dns_check_names_pattern to allow slashes within
962 names, as there are now some PTR records that contain slashes. This check
963 is only to protect against broken name servers that fall over on strange
964 characters, so the fact that it applies to all lookups doesn't matter.
966 PH/36 Now that the new test suite is complete, we can remove some of the
967 special code in Exim that was needed for the old test suite. For example,
968 sorting DNS records because real resolvers return them in an arbitrary
969 order. The new test suite's fake resolver always returns records in the
972 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
973 bounces) except when queue_only is set, to avoid logging races between
974 the different processes.
976 PH/38 Panic-die if .include specifies a non-absolute path.
978 PH/39 A tweak to the "H" retry rule from its user.
980 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
981 a label. They prevented compilation on older perls.
983 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
984 a warning to be raised on newish perls.
986 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
987 on queue. Changes to match documented behaviour of showing count of
988 messages matching specified criteria.
990 PH/40 Changed the default ident timeout from 30s to 5s.
992 PH/41 Added support for the use of login_cap features, on those BSD systems
993 that have them, for controlling the resources used by pipe deliveries.
995 PH/42 The content-scanning code uses fopen() to create files in which to put
996 message data. Previously it was not paying any attention to the mode of
997 the files. Exim runs with umask(0) because the rest of the code creates
998 files with open(), and sets the required mode explicitly. Thus, these
999 files were ending up world-writeable. This was not a big issue, because,
1000 being within the spool directory, they were not world-accessible. I have
1001 created a function called modefopen, which takes an additional mode
1002 argument. It sets umask(777), creates the file, chmods it to the required
1003 mode, then resets the umask. All the relevant calls to fopen() in the
1004 content scanning code have been changed to use this function.
1006 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1007 to 24 hours. This avoids potential overflow problems when processing G
1008 and H retry rules. I suspect nobody ever tinkers with this value.
1010 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1012 PH/45 When the plaintext authenticator is running as a client, the server's
1013 challenges are checked to ensure they are valid base64 strings. By
1014 default, the authentication attempt is cancelled if an invalid string is
1015 received. Setting client_ignore_invalid_base64 true ignores these errors.
1016 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1017 they are received. Thus, the responses can be made to depend on the
1018 challenges. If an invalid string is ignored, an empty string is placed in
1021 PH/46 Messages that are created by the autoreply transport now contains a
1022 References: header, in accordance with RFCs 2822 and 3834.
1024 PH/47 Added authenticated_sender_force to the smtp transport.
1026 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1028 PH/49 Installed latest patch from the Sieve maintainer.
1030 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1031 was also set, the appendfile transport was unnecessarily scanning a
1032 directory of message files (e.g. for maildir delivery) to find the count
1033 of files (along with the size), even though it did not need this
1034 information. It now does the scan only if it needs to find either the
1035 size of the count of files.
1037 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1039 PH/52 Two bugs concerned with error handling when the smtp transport is
1042 (i) Exim was not creating retry information for temporary errors given
1043 for individual recipients after the DATA command when the smtp transport
1044 was used in LMTP mode. This meant that they could be retried too
1045 frequently, and not timed out correctly.
1047 (ii) Exim was setting the flag that allows error details to be returned
1048 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1049 recipients that were returned after the DATA command.
1051 PH/53 This is related to PH/52, but is more general: for any failing address,
1052 when detailed error information was permitted to be returned to the
1053 sender, but the error was temporary, then after the final timeout, only
1054 "retry timeout exceeded" was returned. Now it returns the full error as
1055 well as "retry timeout exceeded".
1057 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1058 do this, and (what is worse) MTAs that accept it.
1060 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1061 will now be deprecated.
1063 PH/56 New os.c-cygwin from the Cygwin maintainer.
1065 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1066 formats (previously only available in exim formats via -bpr, -bpru,
1067 and -bpra. Now also available in native and exiqgrep formats)
1069 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1070 with very large, slow to parse queues
1072 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1074 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1076 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1077 responses to authentication challenges, though it was showing the
1078 challenges; (ii) I've removed the CR characters from the debug output for
1081 PH/58 Allow for the insertion of a newline as well as a space when a string
1082 is turned into more than one encoded-word during RFC 2047 encoding. The
1083 Sieve code now uses this.
1085 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1086 data_4xx, lost_connection, tls_required.
1088 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1089 message was being sent as an SMTP response.
1091 PH/61 Add -l and -k options to exicyclog.
1093 PH/62 When verifying, if an address was redirected to one new address, so that
1094 verification continued, and the new address failed or deferred after
1095 having set something in $address_data, the value of $address_data was not
1096 passed back to the ACL. This was different to the case when no
1097 redirection occurred. The value is now passed back in both cases.
1099 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1100 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1101 use login_cap.h, so on its own it isn't the distinguishing feature. The
1102 new name refers directly to the setclassresources() function.
1104 PH/65 Added configuration files for NetBSD3.
1106 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1108 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1109 is preferred over IPv4.
1111 PH/68 The bounce_return_message and bounce_return_body options were not being
1112 honoured for bounces generated during the reception of non-SMTP messages.
1113 In particular, this applied to messages rejected by the ACL. This bug has
1114 been fixed. However, if bounce_return_message is true and bounce_return_
1115 body is false, the headers that are returned for a non-SMTP message
1116 include only those that have been read before the error was detected.
1117 (In the case of an ACL rejection, they have all been read.)
1119 PH/69 The HTML version of the specification is now built in a directory called
1120 spec_html instead of spec.html, because the latter looks like a path with
1121 a MIME-type, and this confuses some software.
1123 PH/70 Catch two compiler warnings in sieve.c.
1125 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1126 function verify_get_ident() calls ip_connect() to connect a socket, but
1127 if the "connect()" function timed out, ip_connect() used to close the
1128 socket. However, verify_get_ident() also closes the socket later, and in
1129 between Exim writes to the log, which may get opened at this point. When
1130 the socket was closed in ip_connect(), the log could get the same file
1131 descriptor number as the socket. This naturally causes chaos. The fix is
1132 not to close the socket in ip_connect(); the socket should be closed by
1133 the function that creates it. There was only one place in the code where
1134 this was missing, in the iplookup router, which I don't think anybody now
1135 uses, but I've fixed it anyway.
1137 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1138 well as to direct DNS lookups. Otherwise the handling of names in host
1139 lists is inconsistent and therefore confusing.
1145 PH/01 Two changes to the default runtime configuration:
1147 (1) Move the checks for relay_from_hosts and authenticated clients from
1148 after to before the (commented out) DNS black list checks.
1150 (2) Add control=submission to the relay_from_hosts and authenticated
1151 clients checks, on the grounds that messages accepted by these
1152 statements are most likely to be submissions.
1154 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1156 (1) Generate an error if the third argument for the ${prvs expansion is
1159 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1162 (3) Reset the variables that are obtained from the first argument of
1163 ${prvscheck and used in the second argument before leaving the code,
1164 because their memory is reclaimed, so using them afterwards may do
1167 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1168 one (it's much easier than Tom thought :-).
1170 (5) Because of (4), we can now allow for the use of $prvscheck_result
1171 inside the third argument.
1173 PH/03 For some reason, the default setting of PATH when running a command from
1174 a pipe transport was just "/usr/bin". I have changed it to
1177 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1178 anything to be listed in the output from -bV.
1180 PH/05 When a filter generated an autoreply, the entire To: header line was
1181 quoted in the delivery log line, like this:
1183 => >A.N.Other <ano@some.domain> <original@ddress> ...
1185 This has been changed so that it extracts the operative address. There
1186 may be more than one such address. If so, they are comma-separated, like
1189 => >ano@some.domain,ona@other.domain <original@ddress> ...
1191 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1192 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1193 not being looked up in the rDNS to get a host name, Exim was showing the
1194 IP address twice in Received: lines, even though the IP addresses were
1195 identical. For example:
1197 Received: from [1.2.3.4] (helo=[1.2.3.4])
1199 However, if the real host name was known, it was omitting the HELO data
1200 if it matched the actual IP address. This has been tidied up so that it
1201 doesn't show the same IP address twice.
1203 PH/07 When both +timestamp and +memory debugging was on, the value given by
1204 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1205 called by the debug printing, thereby overwriting the timestamp buffer.
1206 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1208 PH/08 When the original message was included in an autoreply transport, it
1209 always said "this is a copy of the message, including all the headers",
1210 even if body_only or headers_only was set. It now gives an appropriate
1213 PH/09 Applied a patch from the Sieve maintainer which:
1215 o fixes some comments
1216 o adds the (disabled) notify extension core
1217 o adds some debug output for the result of if/elsif tests
1218 o points to the current vacation draft in the documentation
1219 and documents the missing references header update
1223 o fixes a bug in processing the envelope test (when testing
1224 multiple envelope elements, the last element determinted the
1227 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1228 Electronic Mail") by including:
1230 Auto-submitted: auto-generated
1232 in the messages that it generates (bounce messages and others, such as
1233 warnings). In the case of bounce messages for non-SMTP mesages, there was
1234 also a typo: it was using "Auto_submitted" (underscore instead of
1235 hyphen). Since every message generated by Exim is necessarily in response
1236 to another message, thes have all been changed to:
1238 Auto-Submitted: auto-replied
1240 in accordance with these statements in the RFC:
1242 The auto-replied keyword:
1244 - SHOULD be used on messages sent in direct response to another
1245 message by an automatic process,
1247 - MUST NOT be used on manually-generated messages,
1249 - MAY be used on Delivery Status Notifications (DSNs) and Message
1250 Disposition Notifications (MDNs),
1252 - MUST NOT be used on messages generated by automatic or periodic
1253 processes, except for messages which are automatic responses to
1256 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1257 to the default Received: header definition.
1259 PH/12 Added log selector acl_warn_skipped (default on).
1261 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1262 variables because (a) they are in the wrong storage pool and (b) even if
1263 they were copied, it wouldn't work properly because of the caching.
1265 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1266 checking when decoding. Apparently there are clients that generate
1267 overlong encoded strings. Why am I not surprised?
1269 PH/15 If the first argument of "${if match_address" was not empty, but did not
1270 contain an "@" character, Exim crashed. Now it writes a panic log message
1271 and treats the condition as false.
1273 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1275 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1276 extension "envelope-auth". The code is finished and in agreement with
1277 other implementations, but there is no documentation so far and in fact,
1278 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1279 not changing the active code.
1281 Print executed "if" and "elsif" statements when debugging is used. This
1282 helps a great deal to understand what a filter does.
1284 Document more things not specified clearly in RFC3028. I had all this
1285 sorted out, when out of a sudden new issues came to my mind. Oops."
1287 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1290 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1291 canonical form (as documented). However, after a host name lookup from
1292 the IP address, check_host() was doing a simple string comparison with
1293 addresses acquired from the DNS when checking that the found name did
1294 have the original IP as one of its addresses. Since any found IPv6
1295 addresses are likely to be in abbreviated form, the comparison could
1296 fail. Luckily, there already exists a function for doing the comparison
1297 by converting both addresses to binary, so now that is used instead of
1298 the text comparison.
1300 PH/20 There was another similar case to PH/19, when a complete host name was
1301 given in a host list; looking up its IP address could give an abbreviated
1302 form, whereas the current host's name might or might not be abbreviated.
1303 The same fix has been applied.
1309 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1310 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1313 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1314 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1316 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1318 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1319 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1320 Also, for "Auto-Submitted", treat anything other than "no" as
1321 non-personal, in accordance with RFC 3834. (Previously it treated
1322 anything starting "auto-" as non-personal.)
1324 TF/01 The control=submission/name=... option had a problem with syntax
1325 errors if the name included a slash character. The /name= option
1326 now slurps the rest of the string, so it can include any characters
1327 but it must come last in the list of options (after /sender_retain
1330 PH/05 Some modifications to the interface to the fake nameserver for the new
1338 TK/01 Added the "success_on_redirect" address verification option. See
1339 NewStuff for rationale and an example.
1341 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1343 PH/02 Patch to exigrep to allow it to work on syslog lines.
1345 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1346 fread() to skip over the body file's header line, because in Cygwin the
1347 header line is locked and is inaccessible.
1349 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1350 co-exist for some time) to make it clear that it is the Exim ID that is
1351 referenced, not the Message-ID: header line.
1353 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1354 string_format() function, because snprintf() does not exist on all
1357 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1358 ${stat: expansion item.
1360 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1361 protocol synchronization error", to keep the pedants happy.
1363 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1364 well as for IRIX systems, when gcc is being used. See the host.c source
1367 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1369 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1372 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1373 transport and to the smtp transport in LMTP mode.
1375 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1377 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1378 run a filter in a subprocess. This could lead to confusion in subsequent
1379 lookups in the parent process. There should also be a search_tidyup() at
1380 the end of the subprocess.
1382 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1383 only if the host matched helo_try_verify_hosts, which caused the
1384 verification to occur when the EHLO/HELO command was issued. The ACL just
1385 tested the remembered result. Now, if a previous verification attempt has
1386 not happened, "verify = helo" does it there and then.
1388 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1390 TK/03 Fix log output including CR from clamd.
1392 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1393 exist provoked a memory error which could cause a segfault.
1395 PH/15 Installed PCRE 6.2
1397 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1399 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1400 of the problem. Specifically, suggested +O2 rather than +O1 for the
1403 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1405 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1406 record, the port was not correctly logged when the outgoing_port log
1407 selector was set (it logged the transort's default port).
1409 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1410 fallback_hosts, and "hosts" in the smtp transport.
1412 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1413 host errors such as "Connection refused".
1415 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1416 authentication with radiusclient 0.4.9:
1418 - Error returned from rc_read_config was caught wrongly
1419 - Username/password not passed on to radius server due to wrong length.
1421 The presumption is that some radiusclient API changes for 4.51/PH/17
1422 were not taken care of correctly. The code is still untested by me (my
1423 Linux distribution still has 0.3.2 of radiusclient), but it was
1424 contributed by a Radius user.
1426 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1427 expanding the "port" option of the smtp transport.
1429 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1430 while reading a MIME header. Thanks to Tom Hughes for a patch.
1432 PH/24 Include config.h inside local_scan.h so that configuration settings are
1435 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1436 that in an ACL for RCPT (for example), you can examine exactly what was
1439 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1440 commands, but it was not correctly comparing the address with the actual
1441 client host address. Thus, it would show the EHLO address in Received:
1442 header lines when this was not necessary.
1444 PH/27 Added the % operator to ${eval:}.
1446 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1447 it should be ignoring failures (because with -C, for example, it has lost
1448 privilege). It wasn't ignoring creation failures other than "already
1451 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1454 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1455 because an input line was too long, either on its own, or by virtue of
1456 too many continuations, the temporary file was not being removed, and the
1457 return code was incorrect.
1459 PH/31 Missing "BOOL" in function definition in filtertest.c.
1461 PH/32 Applied Sieve patches from the maintainer.
1463 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1465 PH/33 Added "verify = not_blind".
1467 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1468 Local/Makefile (with some defaults set). These are used in built scripts
1469 such as exicyclog, but they have never been used in the exim_install
1470 script (though there are many overriding facilities there). I have
1471 arranged that the exim_install script now takes note of these two
1474 PH/35 Installed configuration files for Dragonfly.
1476 PH/36 When a locally submitted message by a trusted user did not contain a
1477 From: header, and the sender address was obtained from -f or from an SMTP
1478 MAIL command, and the trusted user did not use -F to supply a sender
1479 name, $originator_name was incorrectly used when constructing a From:
1480 header. Furthermore, $originator_name was used for submission mode
1481 messages from external hosts without From: headers in a similar way,
1482 which is clearly wrong.
1484 PH/37 Added control=suppress_local_fixups.
1486 PH/38 When log_selector = +received_sender was set, and the addition of the
1487 sender made the log line's construction buffer exactly full, or one byte
1488 less than full, an overflow happened when the terminating "\n" was
1491 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1492 when the result of a list match is failure because a DNS lookup failed.
1494 PH/40 RM_COMMAND is now used in the building process.
1496 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1497 the "build-* directories that it finds.
1499 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1500 address in a domain literal was a prefix of an interface address.
1502 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1503 when verifying a sender address, unless rewrite_headers is false.
1505 PH/44 Wrote a long comment about why errors_to addresses are verified as
1506 recipients, not senders.
1508 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1509 the ratelimit ACL was added.
1511 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1513 PH/46 Added extra information about PostgreSQL errors to the error string.
1515 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1516 avoiding the need to install special zones in a real server. This is
1517 backwards compatible; if it can't find the fake resolver, it drops back.
1518 Thus, both old and new test suites can be run.
1520 TF/02 Added util/ratelimit.pl
1522 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1523 clock is set back in time.
1525 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1526 Candler <B.Candler@pobox.com>.
1528 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1529 OK for addresses that are the result of redirections.
1531 PH/48 A number of further additions for the benefit of the new test suite,
1532 including a fake gethostbyname() that interfaces to the fake DNS resolver
1535 TF/06 The fix for widen_domains has also been applied to qualify_single and
1536 search_parents which are the other dnslookup options that can cause
1539 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1542 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1544 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1545 service" if it encountered an MX record with an empty target hostname.
1546 The message is now "an MX or SRV record indicated no SMTP service".
1548 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1549 if the DNS of the sending site is misconfigured. This is quite a
1550 common situation. This change restores the behaviour of treating a
1551 helo verification defer as a failure.
1553 PH/51 If self=fail was set on a router, the bounce message did not include the
1554 actual error message.
1560 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1562 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1563 command itself ended in error, the underlying message about the transport
1564 filter timeout was being overwritten with the pipe command error. Now the
1565 underlying error message should be appended to the second error message.
1567 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1569 PH/02 Reduce the amount of output that "make" produces by default. Full output
1570 can still be requested.
1572 PH/03 The warning log line about a condition test deferring for a "warn" verb
1573 was being output only once per connection, rather than after each
1574 occurrence (because it was using the same function as for successful
1575 "warn" verbs). This seems wrong, so I have changed it.
1577 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1578 it should not have, which might have caused a crash in the right
1579 circumstances, but probably never did.
1581 PH/04 Installed a modified version of Tony Finch's patch to make submission
1582 mode fix the return path as well as the Sender: header line, and to
1583 add a /name= option so that you can make the user's friendly name appear
1586 TF/03 Added the control = fakedefer ACL modifier.
1588 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1589 Mark Lowes for thorough testing.
1591 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1593 TK/03 Merged latest SRS patch from Miles Wilton.
1595 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1596 with the definition in sysexits.h (which is #included earlier).
1597 Fortunately, Exim does not actually use EX_OK. The code used to try to
1598 preserve the sysexits.h value, by assumimg that macro definitions were
1599 scanned for macro replacements. I have been disabused of this notion,
1600 so now the code just undefines EX_OK before #including unistd.h.
1602 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1603 in the smtp transport. When a block could not be written in a single
1604 write() function, the timeout was being re-applied to each part-write.
1605 This seems wrong - if the receiver was accepting one byte at a time it
1606 would take for ever. The timeout is now adjusted when this happens. It
1607 doesn't have to be particularly precise.
1609 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1610 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1612 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1613 without a domain if the check to <postmaster@domain> fails.
1615 SC/01 Eximstats: added -xls and the ability to specify output files
1616 (patch written by Frank Heydlauf).
1618 SC/02 Eximstats: use FileHandles for outputing results.
1620 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1622 SC/04 Eximstats: fixed display of large numbers with -nvr option
1624 SC/05 Eximstats: fixed merging of reports with empty tables.
1626 SC/06 Eximstats: added the -include_original_destination flag
1628 SC/07 Eximstats: removed tabs and trailing whitespace.
1630 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1632 TK/06 MBOX spool code: Add real "From " MBOX separator line
1633 so the .eml file is really in mbox format (even though
1634 most programs do not really care). Patch from Alex Miller.
1636 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1637 The latter is generated from $received_to and is only set if the
1638 message has one envelope recipient. SA can use these headers,
1639 obviously out-of-the-box. Patch from Alex Miller.
1641 PH/08 The ${def test on a variable was returning false if the variable's
1642 value was "0", contrary to what the specification has always said!
1643 The result should be true unless the variable is empty.
1645 PH/09 The syntax error of a character other than { following "${if
1646 def:variable_name" (after optional whitespace) was not being diagnosed.
1647 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1648 accidental colon was present, for example, could give incorrect results.
1650 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1651 result is used (not including appendfile, where other changes are about
1654 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1655 This involved changing a lot of size variables from int to off_t. It
1656 should work with maildirs and everything.
1658 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1659 spamd dying while we are connected to it.
1661 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1664 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1665 function for BDB 4.3.
1667 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1669 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1670 the sources for PCRE have been re-arranged and more files are now
1673 PH/15 The code I had for printing potentially long long variables in PH/11
1674 above was not the best (it lost precision). The length of off_t variables
1675 is now inspected at build time, and an appropriate printing format (%ld
1676 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1677 to be "long long int" or "long int". This is needed for the internal
1678 formatting function string_vformat().
1680 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1681 the configuration file to be ":syslog", then the script "guesses" where
1682 the logs files are, rather than using the compiled in default. In our
1683 case the guess is not the same as the compiled default, so the script
1684 suddenly stopped working when I started to use syslog. The patch checks
1685 to see if log_file_path is "". If so, it attempts to read it from exim
1686 with no configuration file to get the compiled in version, before it
1687 falls back to the previous guessing code."
1689 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1690 implementing BATV in an Exim configuration. See NewStuff for the gory
1693 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1694 Makefile that are specific to HP-UX.
1696 PH/18 If the "use_postmaster" option was set for a recipient callout together
1697 with the "random" option, the postmaster address was used as the MAIL
1698 FROM address for the random test, but not for the subsequent recipient
1699 test. It is now used for both.
1701 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1702 patch removes a few documentation additions to RFC 3028, because the
1703 latest draft now contains them. It adds the new en;ascii-case comparator
1704 and a new error check for 8bit text in MIME parts. Comparator and
1705 require names are now matched exactly. I enabled the subaddress
1706 extension, but it is not well tested yet (read: it works for me)."
1708 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1709 rework some of the code of TK/09 above to avoid the hardwired use of
1710 "%lld" and "long long". Replaced the call to snprintf() with a call to
1713 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1714 records point to non-existent hosts", "retry timeout exceeded", and
1715 "retry time not reached for any host after a long failure period".
1717 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1718 experimental DomainKeys support:
1720 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1721 (2) On an error such as an illegally used "control", the wrong name for
1722 the control was given.
1724 These problems did NOT occur unless DomainKeys support was compiled.
1726 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1728 PH/24 Added ${if match_ip condition.
1730 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1731 databases so that it will be absolutely obvious if a crash occurs in the
1732 DB library. This is a regular occurrence (often caused by mis-matched
1735 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1736 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1737 picked up on a user's system that detects such things. There doesn't seem
1738 to be a gcc warning option for this - only an attribute that has to be
1739 put on the function's prototype. It seems that in Fedora Core 4 they have
1740 set this on a number of new functions. No doubt there will be more in due
1743 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1744 specify a transport. However, if an address that was verified by such a
1745 router was the subject of a callout, Exim crashed because it tried to
1746 read the rcpt_include_affixes from the non-existent transport. Now it
1747 just assumes that the setting of that option is false. This bug was
1748 introduced by 4.51/PH/31.
1750 PH/28 Changed -d+all to exclude +memory, because that information is very
1751 rarely of interest, but it makes the output a lot bigger. People tend to
1752 do -d+all out of habit.
1754 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1755 code in os-type was giving problems when libc.so lives in lib64, like on
1758 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1759 aren't the modern standard, and it seems that some systems' include files
1760 don't always have them. Exim was already checking for some of the newer
1761 ones like T_AAAA, and defining it itself. I've added checks for all the
1762 record types that Exim uses.
1764 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1765 not automatically generating a new one, as it is supposed to. This
1766 prevented TLS from working. If the file did exist, but contained invalid
1767 data, a new version was generated, as expected. It was only the case of a
1768 non-existent file that was broken.
1770 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1771 with a change in libdomainkeys > 0.64.
1773 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1774 from DNS. If the selector record carries the flag, it now has
1775 precedence over the domain-wide flag.
1777 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1779 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1780 the use of a port name that isn't defined in /etc/services) occurred, the
1781 message was deferred as in a normal delivery, and thus remained on the
1782 spool, instead of being failed because of the mua_wrapper setting. This
1783 is now fixed, and I tidied up some of the mua_wrapper messages at the
1786 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1787 the messages in a hash of arrays rather than using individual hashes.
1788 This is a bit cleaner and results in dramatic memory savings, albeit
1789 at a slight CPU cost.
1791 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1792 as requested by Marc Sherman.
1794 SC/10 Eximstats: added histograms for user specified patterns as requested
1797 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1799 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1800 fopen() in the content-scanning modules that did not already have it.
1806 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1807 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1809 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1811 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1813 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1814 to test Sieve filters that use "vacation".
1816 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1817 that changes the way the GnuTLS parameters are stored in the cache file.
1818 The new format can be generated externally. For backward compatibility,
1819 if the data in the cache doesn't make sense, Exim assumes it has read an
1820 old-format file, and it generates new data and writes a new file. This
1821 means that you can't go back to an older release without removing the
1824 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1825 work if there are any delivery delays because "one_time" forces the
1826 parent to be marked "delivered", so its unseen clone is never tried
1827 again. For this reason, Exim now forbids the simultaneous setting of
1830 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1831 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1832 entry if you want to know the details. The fix, however, neglected to
1833 consider the case when local delivery batching is involved. The test for
1834 "previously delivered" was not happening when checking to see if an
1835 address could be batched with a previous (undelivered) one; under
1836 certain circumstances this could lead to multiple deliveries to the same
1839 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1840 in its include files, and this causes problems building Exim.
1842 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1843 header_syntax) but Exim was just ignoring anything given after a slash.
1844 In particular, this caused confusion with an attempt to use "verify =
1845 reverse_host_lookup/defer_ok". An error is now given when options are
1846 supplied for verify items that do not have them. (Maybe reverse_host_
1847 lookup should have a defer_ok option, but that's a different point.)
1849 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1850 defined by RFC 821) to 2048, because there were problems with some AUTH
1851 commands, and RFC 1869 says the size should be increased for extended
1852 SMTP commands that take arguments.
1854 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1857 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1858 "unknown" error; now it says that the functionality isn't in the binary.
1860 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1861 an address' error message when a string expansion fails (syntax or
1862 whatever). Otherwise the password may appear in the log. Following change
1863 PH/42 below, there is no longer a chance of it appearing in a bounce
1866 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1868 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1869 of its addresses was ever tried. (Bugzilla bug #2.)
1871 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1872 the result incorrectly in the debug output. (It correctly added a newline
1873 to what was transported.)
1875 TF/01 Added $received_time.
1877 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1878 commented out examples of how to interface to a virus scanner and to
1879 SpamAssassin. Also added commented examples of av_scanner and
1880 spamd_address settings.
1882 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1883 and controls are allowed in which ACLs. There were a couple of minor
1884 errors. Some of the entries in the conditions table (which is a table of
1885 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1886 negation of where the condition IS allowed.
1888 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1890 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1891 header file does not have a version number, so I've had to invent a new
1892 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1893 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1894 radiusclient), but it was contributed by a Radius user.
1896 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1897 files or directories, for OpenSSL.
1899 PH/19 When an Exim process that is running as root has to create an Exim log
1900 file, it does so in a subprocess that runs as exim:exim so as to get the
1901 ownership right at creation (otherwise, other Exim processes might see
1902 the file with the wrong ownership). There was no test for failure of this
1903 fork() call, which would lead to the process getting stuck as it waited
1904 for a non-existent subprocess. Forks do occasionally fail when resources
1905 run out. I reviewed all the other calls to fork(); they all seem to check
1908 PH/20 When checking for unexpected SMTP input at connect time (before writing
1909 the banner), Exim was not dealing correctly with a non-positive return
1910 from the read() function. If the client had disconnected by this time,
1911 the result was a log entry for a synchronization error with an empty
1912 string after "input=" when read() returned zero. If read() returned -1
1913 (an event I could not check), uninitialized data bytes were printed.
1914 There were reports of junk text (parts of files, etc) appearing after
1917 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1919 PH/22 Added support for macro redefinition, and (re)definition in between
1920 driver and ACL definitions.
1922 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1923 forgetting to use the resulting value; it was using the unexpanded value.
1925 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1926 hadn't been configured. The fix is from Juergen Kreileder, who
1927 understands it better than I do:
1929 "Here's what I see happening with three configured cyrus_sasl
1930 authenticators configured (plain, login, cram-md5):
1932 On startup auth_cyrus_sasl_init() gets called for each of these.
1933 This means three calls to sasl_listmech() without a specified mech_list.
1934 => SASL tests which mechs of all available mechs actually work
1935 => three warnings about OTP not working
1936 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1938 With the patch, sasl_listmech() also gets called three times. But now
1939 SASL's mech_list option is set to the server_mech specified in the the
1940 authenticator. Or in other words, the answer from sasl_listmech()
1941 gets limited to just the mech you're testing for (which is different
1943 => the return list contains just 'plain' or 'login', 'cram-md5' or
1944 nothing depending on the value of ob->server_mech.
1946 I've just tested the patch: Authentication still works fine,
1947 unavailable mechs specified in the exim configuration are still
1948 caught, and the auth.log warnings about OTP are gone."
1950 PH/25 When debugging is enabled, the contents of the command line are added
1951 to the debugging output, even when log_selector=+arguments is not
1954 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1955 answer is "GNU", and only if the return is "GNU/something" is the answer
1958 PH/27 $acl_verify_message is now set immediately after the failure of a
1959 verification in an ACL, and so is available in subsequent modifiers. In
1960 particular, the message can be preserved by coding like this:
1962 warn !verify = sender
1963 set acl_m0 = $acl_verify_message
1965 Previously, $acl_verify_message was set only while expanding "message"
1966 and "log_message" when a very denied access.
1968 PH/28 Modified OS/os.c-Linux with
1970 -#ifndef OS_LOAD_AVERAGE
1971 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1973 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1974 nomenclature these days.)
1976 PH/29 Installed patch from the Sieve maintainer that adds the options
1977 sieve_useraddress and sieve_subaddress to the redirect router.
1979 PH/30 In these circumstances:
1980 . Two addresses routed to the same list of hosts;
1981 . First host does not offer TLS;
1982 . First host accepts first address;
1983 . First host gives temporary error to second address;
1984 . Second host offers TLS and a TLS session is established;
1985 . Second host accepts second address.
1986 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1987 and peerdn, if requested) that were in fact used only for the second
1990 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1991 attention to any local part prefix or suffix that was matched by the
1992 router that accepted the address. It now behaves in the same way as it
1993 does for delivery: the affixes are removed from the local part unless
1994 rcpt_include_affixes is set on the transport.
1996 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1997 timeout during the DATA phase of an incoming message.
1999 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2000 applied a patch sanctioned by the Sieve maintainer.
2002 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2003 the uid or gid is negative. A case of a negative gid caused this to be
2004 noticed. The fix allows for either to be negative.
2006 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2007 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2010 PH/36 The change PH/12 above was broken. Fixed it.
2012 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2013 the grounds that routing the same address twice would always produce the
2014 same answer. This might have been true once, but it is certainly no
2015 longer true now. Routing a child address may depend on the previous
2016 routing that produced that child. Some complicated redirection strategies
2017 went wrong when messages had multiple recipients, and made Exim's
2018 behaviour dependent on the order in which the addresses were given.
2020 I have moved the duplicate checking until after the routing is complete.
2021 Exim scans the addresses that are assigned to local and remote
2022 transports, and removes any duplicates. This means that more work will be
2023 done, as duplicates will always all be routed, but duplicates are
2024 presumably rare, so I don't expect this is of any significance.
2026 For deliveries to pipes, files, and autoreplies, the duplicate checking
2027 still happens during the routing process, since they are not going to be
2030 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2031 It corrects a timeout issue with spamd. This is Ian's comment: "The
2032 background is that sometimes spamd either never reads data from a
2033 connection it has accepted, or it never writes response data. The exiscan
2034 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2035 blindly assumes that writes won't block so it may never time out."
2037 PH/39 Allow G after quota size as well as K and M.
2039 PH/40 The value set for $authenticated_id in an authenticator may not contain
2040 binary zeroes or newlines because the value is written to log lines and
2041 to spool files. There was no check on this. Now the value is run through
2042 the string_printing() function so that such characters are converted to
2043 printable escape sequences.
2045 PH/41 $message_linecount is a new variable that contains the total number of
2046 lines in the message. Compare $body_linecount, which is the count for the
2049 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2050 bounce and delay warning messages, except in certain special cases, which
2053 (a) An SMTP error message from a remote host;
2054 (b) A message specified in a :fail: redirection;
2055 (c) A message specified in a "fail" command in a system filter;
2056 (d) A message specified in a FAIL return from the queryprogram router;
2057 (e) A message specified by the cannot_route_message router option.
2059 In these cases only, Exim does include the error details in bounce and
2060 warning messages. There are also a few cases where bland messages such
2061 as "unrouteable address" or "local delivery error" are given.
2063 PH/43 $value is now also set for the "else" part of a ${run expansion.
2065 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2066 being worked on, but at least Exim now implements the latest version to
2069 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2070 process to complete was treated as a delivery failure, a timeout while
2071 writing the message to the pipe was logged, but erroneously treated as a
2072 successful delivery. Such timeouts include transport filter timeouts. For
2073 consistency with the overall process timeout, these timeouts are now
2074 treated as errors, giving rise to delivery failures by default. However,
2075 there is now a new Boolean option for the pipe transport called
2076 timeout_defer, which, if set TRUE, converts the failures into defers for
2077 both kinds of timeout. A transport filter timeout is now identified in
2080 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2081 systems where "make" and "gmake" are different, calling "gmake" at top
2082 level broke things. I've arranged for the value of $(MAKE) to be passed
2083 from the Makefile to this script so that it can call the same version of
2087 A note about Exim versions 4.44 and 4.50
2088 ----------------------------------------
2090 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2091 changes of various kinds. As a consequence, a big documentation update was
2092 needed. This delayed the release for rather longer than seemed good, especially
2093 in the light of a couple of (minor) security issues. Therefore, the changes
2094 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2095 release. So 4.44 and 4.50 are in effect two different branches that both start
2098 I have left the 4.50 change log unchanged; it contains all the changes since
2099 4.43. The change log for 4.44 is below; many of its items are identical to
2100 those for 4.50. This seems to be the most sensible way to preserve the
2101 historical information.
2107 1. Minor wording change to the doc/README.SIEVE file.
2109 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2110 computation of the current number of files was incorrect.
2112 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2113 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2116 4. Give more explanation in the error message when the command for a transport
2117 filter fails to execute.
2119 5. There are several places where Exim runs a non-Exim command in a
2120 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2121 was being done only for the command run by the queryprogram router. It is
2122 now done for all such subprocesses. The other cases are: ${run, transport
2123 filters, and the commands run by the lmtp and pipe transports.
2125 6. Added CONFIGURE_GROUP build-time option.
2127 7. Some older OS have a limit of 256 on the maximum number of file
2128 descriptors. Exim was using setrlimit() to set 1000 as a large value
2129 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2130 systems. I've change it so that if it can't get 1000, it tries for 256.
2132 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2133 was an oversight, and furthermore, ever since the addition of extra
2134 controls (e.g. 4.43/32), the checks on when to allow different forms of
2135 "control" were broken. There should now be diagnostics for all cases when a
2136 control that does not make sense is encountered.
2138 9. Added the /retain_sender option to "control=submission".
2140 10. $recipients is now available in the predata ACL (oversight).
2142 11. Tidy the search cache before the fork to do a delivery from a message
2143 received from the command line. Otherwise the child will trigger a lookup
2144 failure and thereby defer the delivery if it tries to use (for example) a
2145 cached ldap connection that the parent has called unbind on.
2147 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2148 of $address_data from the recipient verification was clobbered by the
2149 sender verification.
2151 13. The value of address_data from a sender verification is now available in
2152 $sender_address_data in subsequent conditions in the ACL statement.
2154 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2156 15. Added a new option "connect=<time>" to callout options, to set a different
2159 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2160 was its contents. (It was OK if the option was not defined at all.)
2162 17. A "Completed" log line is now written for messages that are removed from
2163 the spool by the -Mrm option.
2165 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2166 information about exactly what failed.
2168 19. Added -dd to debug only the daemon process.
2170 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2171 handles timeouts, both on the server side and network timeouts. Renamed the
2172 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2174 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2175 It is now set to "smtps".
2177 22. $host_address is now set to the target address during the checking of
2178 ignore_target_hosts.
2180 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2181 being passed; this would have caused $sender_host_name to have been used if
2182 matching the list had actually called for a host name (not very likely,
2183 since this list is usually IP addresses). A host name is now passed as
2186 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2187 code that specifies a non-restarting handler (typically sigaction() in
2188 modern systems) in an attempt to fix a rare and obscure crash bug.
2190 25. Narrowed the window for a race in the daemon that could cause it to ignore
2191 SIGCHLD signals. This is not a major problem, because they are used only to
2192 wake it up if nothing else does.
2194 26. A malformed maildirsize file could cause Exim to calculate negative values
2195 for the mailbox size or file count. Odd effects could occur as a result.
2196 The maildirsize information is now recalculated if the size or filecount
2199 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2200 support for a long time. Removed HAVE_SYS_VFS_H.
2202 28. Installed the latest version of exipick from John Jetmore.
2204 29. In an address list, if the pattern was not a regular expression, an empty
2205 subject address (from a bounce message) matched only if the pattern was an
2206 empty string. Non-empty patterns were not even tested. This was the wrong
2207 because it is perfectly reasonable to use an empty address as part of a
2208 database query. An empty address is now tested by patterns that are
2209 lookups. However, all the other forms of pattern expect the subject to
2210 contain a local part and a domain, and therefore, for them, an empty
2211 address still always fails if the pattern is not itself empty.
2213 30. Exim went into a mad DNS loop when attempting to do a callout where the
2214 host was specified on an smtp transport, and looking it up yielded more
2215 than one IP address.
2217 31. Re-factored the code for checking spool and log partition space into a
2218 function that finds that data and another that does the check. The former
2219 is then used to implement four new variables: $spool_space, $log_space,
2220 $spool_inodes, and $log_inodes.
2222 32. The RFC2047 encoding function was originally intended for short strings
2223 such as real names; it was not keeping to the 75-character limit for
2224 encoded words that the RFC imposes. It now respects the limit, and
2225 generates multiple encoded words if necessary. To be on the safe side, I
2226 have increased the buffer size for the ${rfc2047: expansion operator from
2229 33. It is now permitted to omit both strings after an "if" condition; if the
2230 condition is true, the result is "true". As before, when the second string
2231 is omitted, a false condition yields an empty string. This makes it less
2232 cumbersome to write custom ACL and router conditions.
2234 34. Failure to deliver a bounce message always caused it to be frozen, even if
2235 there was an errors_to setting on the router. The errors_to setting is now
2238 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2239 canonical form (fully expanded) before being placed in
2240 $sender_host_address.
2242 36. The table in the code that translates DNS record types into text (T_A to
2243 "A" for instance) was missing entries for NS and CNAME. It is just possible
2244 that this could have caused confusion if both these types were looked up
2245 for the same domain, because the text type is used as part of Exim's
2246 per-process caching. But the chance of anyone hitting this buglet seems
2249 37. The dnsdb lookup has been extended in a number of ways.
2251 (1) There is a new type, "zns", which walks up the domain tree until it
2252 finds some nameserver records. It should be used with care.
2254 (2) There is a new type, "mxh", which is like "mx" except that it returns
2255 just the host names, not the priorities.
2257 (3) It is now possible to give a list of domains (or IP addresses) to be
2258 looked up. The behaviour when one of the lookups defers can be
2259 controlled by a keyword.
2261 (4) It is now possible to specify the separator character for use when
2262 multiple records are returned.
2264 38. The dnslists ACL condition has been extended: it is now possible to supply
2265 a list of IP addresses and/or domains to be looked up in a particular DNS
2268 39. Added log_selector=+queue_time_overall.
2270 40. When running the queue in the test harness, wait just a tad after forking a
2271 delivery process, to get repeatability of debugging output.
2273 41. Include certificate and key file names in error message when GnuTLS fails
2274 to set them up, because the GnuTLS error message doesn't include the name
2275 of the failing file when there is a problem reading it.
2277 42. Allow both -bf and -bF in the same test run.
2279 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2281 44. The "Exiscan patch" is now merged into the mainline Exim source.
2283 45. Sometimes the final signoff response after QUIT could fail to get
2284 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2285 < 0 before doing a fflush(). This bug looks as though it goes back to the
2286 introduction of TLS in release 3.20, but "sometimes" must have been rare
2287 because the tests only now provoked it.
2289 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2290 (this can affect the format of dates).
2292 47. exim_tidydb, when checking for the continued existence of a message for
2293 which it has found a message-specific retry record, was not finding
2294 messages that were in split spool directories. Consequently, it was
2295 deleting retry records that should have stayed in existence.
2297 48. Steve fixed some bugs in eximstats.
2299 49. The SPA authentication driver was not abandoning authentication and moving
2300 on to the next authenticator when an expansion was forced to fail,
2301 contradicting the general specification for all authenticators. Instead it
2302 was generating a temporary error. It now behaves as specified.
2304 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2305 (the order specifies the preference for clients). The order is now AES256,
2306 AES128, 3DES, ARCFOUR128.
2308 51. Small patch to Sieve code - explicitly set From: when generating an
2311 52. Exim crashed if a remote delivery caused a very long error message to be
2312 recorded - for instance if somebody sent an entire SpamAssassin report back
2313 as a large number of 550 error lines. This bug was coincidentally fixed by
2314 increasing the size of one of Exim's internal buffers (big_buffer) that
2315 happened as part of the Exiscan merge. However, to be on the safe side, I
2316 have made the code more robust (and fixed the comments that describe what
2319 53. Now that there can be additional text after "Completed" in log lines (if
2320 the queue_time_overall log selector is set), a one-byte patch to exigrep
2321 was needed to allow it to recognize "Completed" as not the last thing in
2324 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2325 patch that reportedly fixes this has been added. I am not expert enough to
2326 create a test for it. This is what the patch creator wrote:
2328 "I found a little strange behaviour of ldap code when working with
2329 Windows 2003 AD Domain, where users was placed in more than one
2330 Organization Units. When I tried to give exim partial DN, the exit code
2331 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2332 But simultaneously result of request was absolutely normal ldap result,
2333 so I produce this patch..."
2335 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2336 REFERENCE, so I have modified the code to exclude the patch when that macro
2339 55. Some experimental protocols are using DNS PTR records for new purposes. The
2340 keys for these records are domain names, not reversed IP addresses. The
2341 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2342 leaves it alone. Component reversal etc. now happens only for IP addresses.
2345 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2347 57. Double the size of the debug message buffer (to 2048) so that more of very
2348 long debug lines gets shown.
2350 58. The exicyclog utility now does better if the number of log files to keep
2351 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2353 59. Two changes related to the smtp_active_hostname option:
2355 (1) $smtp_active_hostname is now available as a variable.
2356 (2) The default for smtp_banner uses $smtp_active_hostname instead
2357 of $primary_hostname.
2359 60. The host_aton() function is supposed to be passed a string that is known
2360 to be a valid IP address. However, in the case of IPv6 addresses, it was
2361 not checking this. This is a hostage to fortune. Exim now panics and dies
2362 if the condition is not met. A case was found where this could be provoked
2363 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2364 components; fortuitously, this particular loophole had already been fixed
2365 by change 4.50/55 above.
2367 If there are any other similar loopholes, the new check in host_aton()
2368 itself should stop them being exploited. The report I received stated that
2369 data on the command line could provoke the exploit when Exim was running as
2370 exim, but did not say which command line option was involved. All I could
2371 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2372 running as the user.
2375 61. There was a buffer overflow vulnerability in the SPA authentication code
2376 (which came originally from the Samba project). I have added a test to the
2377 spa_base64_to_bits() function which I hope fixes it.
2380 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2381 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2383 63. The daemon start-up calls getloadavg() while still root for those OS that
2384 need the first call to be done as root, but it missed one case: when
2385 deliver_queue_load_max is set with deliver_drop_privilege. This is
2386 necessary for the benefit of the queue runner, because there is no re-exec
2387 when deliver_drop_privilege is set.
2389 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2390 This has been fixed.
2392 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2393 was in use, was not putting the data itself into the right store pool;
2394 consequently, it could be overwritten for a subsequent message in the same
2395 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2398 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2400 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2401 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2402 the function were treating the return as a boolean value, which happened to
2403 work because 0=false and not-0=true, but is not correct code.
2405 68. The host_aton() function was not handling scoped IPv6 addresses (those
2406 with, for example, "%eth0" on the end) correctly.
2408 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2409 negated items (that is, ~something) in unsigned ints. Some compilers
2410 apparently mutter when there is no cast.
2412 70. If an address verification called from an ACL failed, and did not produce a
2413 user-specific message (i.e. there was only a "system" message), nothing was
2414 put in $acl_verify_message. In this situation, it now puts the system
2417 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2418 session; change 4.31/43 added the unwanted input to the log line - except
2419 that it did not do this in the start of session case. It now does.
2421 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2422 This is wrong and can cause the other end to generate a synchronization
2423 error if it is another Exim or anything else that does the synchronization
2424 check. A QUIT command is no longer sent after a timeout.
2426 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2427 during host lookups.
2429 74. The defer_ok option of callout verification was not working if it was used
2430 when verifying addresses in header lines, that is, for this case:
2432 verify = header_sender/callout=defer_ok
2434 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2435 those file descriptors could be used for SMTP connections. If anything
2436 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2437 could be sent to the SMTP client, causing chaos. The daemon now opens
2438 stdin, stdout, and stderr to /dev/null when it puts itself into the
2441 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2442 log by default. The user can override this with suitable Perl magic.
2444 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2445 the log message when discard triggers, was not working for the DATA ACL or
2446 for the non-SMTP ACL.
2448 78. Error message wording change in sieve.c.
2450 79. If smtp_accept_max_per_host was set, the number of connections could be
2451 restricted to fewer than expected, because the daemon was trying to set up
2452 a new connection before checking whether the processes handling previous
2453 connections had finished. The check for completed processes is now done
2454 earlier. On busy systems, this bug wouldn't be noticed because something
2455 else would have woken the daemon, and it would have reaped the completed
2458 80. If a message was submitted locally by a user whose login name contained one
2459 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2460 It caused a spool format error. I have fixed the spool reading code. A
2461 related problem was that the "from" clause in the Received: line became
2462 illegal because of the space(s). It is now covered by ${quote_local_part.
2464 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2467 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2469 83. Patch from Sieve maintainer to fix unterminated string problem in
2470 "vacation" handling.
2472 84. Some minor changes to the Linux configuration files to help with other
2473 OS variants using glibc.
2475 85. One more patch for Sieve to update vacation handling to latest spec.
2478 ----------------------------------------------------
2479 See the note above about the 4.44 and 4.50 releases.
2480 ----------------------------------------------------
2486 1. Change 4.43/35 introduced a bug that caused file counts to be
2487 incorrectly computed when quota_filecount was set in an appendfile
2490 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2491 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2494 3. Give more explanation in the error message when the command for a transport
2495 filter fails to execute.
2497 4. There are several places where Exim runs a non-Exim command in a
2498 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2499 was being done only for the command run by the queryprogram router. It is
2500 now done for all such subprocesses. The other cases are: ${run, transport
2501 filters, and the commands run by the lmtp and pipe transports.
2503 5. Some older OS have a limit of 256 on the maximum number of file
2504 descriptors. Exim was using setrlimit() to set 1000 as a large value
2505 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2506 systems. I've change it so that if it can't get 1000, it tries for 256.
2508 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2509 was an oversight, and furthermore, ever since the addition of extra
2510 controls (e.g. 4.43/32), the checks on when to allow different forms of
2511 "control" were broken. There should now be diagnostics for all cases when a
2512 control that does not make sense is encountered.
2514 7. $recipients is now available in the predata ACL (oversight).
2516 8. Tidy the search cache before the fork to do a delivery from a message
2517 received from the command line. Otherwise the child will trigger a lookup
2518 failure and thereby defer the delivery if it tries to use (for example) a
2519 cached ldap connection that the parent has called unbind on.
2521 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2522 of $address_data from the recipient verification was clobbered by the
2523 sender verification.
2525 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2526 was its contents. (It was OK if the option was not defined at all.)
2528 11. A "Completed" log line is now written for messages that are removed from
2529 the spool by the -Mrm option.
2531 12. $host_address is now set to the target address during the checking of
2532 ignore_target_hosts.
2534 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2535 being passed; this would have caused $sender_host_name to have been used if
2536 matching the list had actually called for a host name (not very likely,
2537 since this list is usually IP addresses). A host name is now passed as
2540 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2541 code that specifies a non-restarting handler (typically sigaction() in
2542 modern systems) in an attempt to fix a rare and obscure crash bug.
2544 15. Narrowed the window for a race in the daemon that could cause it to ignore
2545 SIGCHLD signals. This is not a major problem, because they are used only to
2546 wake it up if nothing else does.
2548 16. A malformed maildirsize file could cause Exim to calculate negative values
2549 for the mailbox size or file count. Odd effects could occur as a result.
2550 The maildirsize information is now recalculated if the size or filecount
2553 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2554 support for a long time. Removed HAVE_SYS_VFS_H.
2556 18. Updated exipick to current release from John Jetmore.
2558 19. Allow an empty sender to be matched against a lookup in an address list.
2559 Previously the only cases considered were a regular expression, or an
2562 20. Exim went into a mad DNS lookup loop when doing a callout where the
2563 host was specified on the transport, if the DNS lookup yielded more than
2566 21. The RFC2047 encoding function was originally intended for short strings
2567 such as real names; it was not keeping to the 75-character limit for
2568 encoded words that the RFC imposes. It now respects the limit, and
2569 generates multiple encoded words if necessary. To be on the safe side, I
2570 have increased the buffer size for the ${rfc2047: expansion operator from
2573 22. Failure to deliver a bounce message always caused it to be frozen, even if
2574 there was an errors_to setting on the router. The errors_to setting is now
2577 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2578 canonical form (fully expanded) before being placed in
2579 $sender_host_address.
2581 24. Updated eximstats to version 1.33
2583 25. Include certificate and key file names in error message when GnuTLS fails
2584 to set them up, because the GnuTLS error message doesn't include the name
2585 of the failing file when there is a problem reading it.
2587 26. Expand error message when OpenSSL has problems setting up cert/key files.
2590 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2591 (this can affect the format of dates).
2593 28. exim_tidydb, when checking for the continued existence of a message for
2594 which it has found a message-specific retry record, was not finding
2595 messages that were in split spool directories. Consequently, it was
2596 deleting retry records that should have stayed in existence.
2598 29. eximstats updated to version 1.35
2599 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2600 1.35 - bugfix such that pie charts by volume are generated correctly
2602 30. The SPA authentication driver was not abandoning authentication and moving
2603 on to the next authenticator when an expansion was forced to fail,
2604 contradicting the general specification for all authenticators. Instead it
2605 was generating a temporary error. It now behaves as specified.
2607 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2608 (the order specifies the preference for clients). The order is now AES256,
2609 AES128, 3DES, ARCFOUR128.
2611 31. Small patch to Sieve code - explicitly set From: when generating an
2614 32. Exim crashed if a remote delivery caused a very long error message to be
2615 recorded - for instance if somebody sent an entire SpamAssassin report back
2616 as a large number of 550 error lines. This bug was coincidentally fixed by
2617 increasing the size of one of Exim's internal buffers (big_buffer) that
2618 happened as part of the Exiscan merge. However, to be on the safe side, I
2619 have made the code more robust (and fixed the comments that describe what
2622 33. Some experimental protocols are using DNS PTR records for new purposes. The
2623 keys for these records are domain names, not reversed IP addresses. The
2624 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2625 leaves it alone. Component reversal etc. now happens only for IP addresses.
2628 34. The host_aton() function is supposed to be passed a string that is known
2629 to be a valid IP address. However, in the case of IPv6 addresses, it was
2630 not checking this. This is a hostage to fortune. Exim now panics and dies
2631 if the condition is not met. A case was found where this could be provoked
2632 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2633 components; fortuitously, this particular loophole had already been fixed
2634 by change 4.50/55 or 4.44/33 above.
2636 If there are any other similar loopholes, the new check in host_aton()
2637 itself should stop them being exploited. The report I received stated that
2638 data on the command line could provoke the exploit when Exim was running as
2639 exim, but did not say which command line option was involved. All I could
2640 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2641 running as the user.
2644 35. There was a buffer overflow vulnerability in the SPA authentication code
2645 (which came originally from the Samba project). I have added a test to the
2646 spa_base64_to_bits() function which I hope fixes it.
2649 36. The daemon start-up calls getloadavg() while still root for those OS that
2650 need the first call to be done as root, but it missed one case: when
2651 deliver_queue_load_max is set with deliver_drop_privilege. This is
2652 necessary for the benefit of the queue runner, because there is no re-exec
2653 when deliver_drop_privilege is set.
2655 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2656 was in use, was not putting the data itself into the right store pool;
2657 consequently, it could be overwritten for a subsequent message in the same
2658 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2661 38. Sometimes the final signoff response after QUIT could fail to get
2662 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2663 < 0 before doing a fflush(). This bug looks as though it goes back to the
2664 introduction of TLS in release 3.20, but "sometimes" must have been rare
2665 because the tests only now provoked it.
2671 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2672 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2673 Now it returns an integer. A number of calls were still expecting a T/F
2674 return. Fortuitously, in all cases, the tests worked in OK situations,
2675 which is the norm. However, things would have gone wrong on any write
2676 failures on the smtp file descriptor. This function is used when sending
2677 messages over SMTP and also when doing verify callouts.
2679 2. When Exim is called to do synchronous delivery of a locally submitted
2680 message (the -odf or -odi options), it no longer closes stderr before doing
2683 3. Implemented the mua_wrapper option.
2685 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2687 5. Implemented the functions header_remove(), header_testname(),
2688 header_add_at_position(), and receive_remove_recipient(), and exported them
2691 6. If an ACL "warn" statement specified the addition of headers, Exim already
2692 inserted X-ACL-Warn: at the start if there was no header name. However, it
2693 was not making this test for the second and subsequent header lines if
2694 there were newlines in the string. This meant that an invalid header could
2695 be inserted if Exim was badly configured.
2697 7. Allow an ACL "warn" statement to add header lines at the start or after all
2698 the Received: headers, as well as at the end.
2700 8. Added the rcpt_4xx retry error code.
2702 9. Added postmaster_mailfrom=xxx to callout verification option.
2704 10. Added mailfrom=xxxx to the callout verification option, for verify=
2707 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2708 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2710 12. Inserted some casts to stop certain compilers warning when using pointer
2711 differences as field lengths or precisions in printf-type calls (mostly
2712 affecting debugging statements).
2714 13. Added optional readline() support for -be (dynamically loaded).
2716 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2717 same clock tick as a message's arrival, so that its received time was the
2718 same as the "first fail" time on the retry record, and that message
2719 remained on the queue past the ultimate address timeout, every queue runner
2720 would try a delivery (because it was past the ultimate address timeout) but
2721 after another failure, the ultimate address timeout, which should have then
2722 bounced the address, did not kick in. This was a "< instead of <=" error;
2723 in most cases the first failure would have been in the next clock tick
2724 after the received time, and all would be well.
2726 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2727 being recognized when the domain list was tested by the match_domain
2728 condition in an expansion string.
2730 16. Added the ${str2b64: operator.
2732 17. Exim was always calling setrlimit() to set a large limit for the number of
2733 processes, without checking whether the existing limit was already
2734 adequate. (It did check for the limit on file descriptors.) Furthermore,
2735 errors from getrlimit() and setrlimit() were being ignored. Now they are
2736 logged to the main and panic logs, but Exim does carry on, to try to do its
2737 job under whatever limits there are.
2739 18. Imported PCRE 5.0.
2741 19. Trivial typo in log message " temporarily refused connection" (the leading
2744 20. If the log selector return_path_on_delivery was set and an address was
2745 redirected to /dev/null, the delivery process crashed because it assumed
2746 that a return path would always be set for a "successful" delivery. In this
2747 case, the whole delivery is bypassed as an optimization, and therefore no
2750 21. Internal re-arrangement: the function for sending a challenge and reading
2751 a response while authentication was assuming a zero-terminated challenge
2752 string. It's now changed to take a pointer and a length, to allow for
2753 binary data in such strings.
2755 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2757 23. Exim was not respecting finduser_retries when seeking the login of the
2758 uid under which it was called; it was always trying 10 times. (The default
2759 setting of finduser_retries is zero.) Also, it was sleeping after the final
2760 failure, which is pointless.
2762 24. Implemented tls_on_connect_ports.
2764 25. Implemented acl_smtp_predata.
2766 26. If the domain in control=submission is set empty, Exim assumes that the
2767 authenticated id is a complete email address when it generates From: or
2768 Sender: header lines.
2770 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2771 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2772 chown and chgrp in /bin and hostname in /usr/bin.
2774 28. Exim was keeping the "process log" file open after each use, just as it
2775 does for the main log. This opens the possibility of it remaining open for
2776 long periods when the USR1 signal hits a daemon. Occasional processlog
2777 errors were reported, that could have been caused by this. Anyway, it seems
2778 much more sensible not to leave this file open at all, so that is what now
2781 29. The long-running daemon process does not normally write to the log once it
2782 has entered its main loop, and it closes the log before doing so. This is
2783 so that log files can straightforwardly be renamed and moved. However,
2784 there are a couple of unusual error situations where the daemon does write
2785 log entries, and I had neglected to close the log afterwards.
2787 30. The text of an SMTP error response that was received during a remote
2788 delivery was being truncated at 512 bytes. This is too short for some of
2789 the long messages that one sometimes sees. I've increased the limit to
2792 31. It is now possible to make retry rules that apply only when a message has a
2793 specific sender, in particular, an empty sender.
2795 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2796 it possible to be selective about when SMTP synchronization is enforced.
2798 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2800 32. Implemented hosts_connection_nolog.
2802 33. Added an ACL for QUIT.
2804 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2807 35. Added mailbox_size and mailbox_filecount to appendfile.
2809 36. Added control = no_multiline_responses to ACLs.
2811 37. There was a bug in the logic of the code that waits for the clock to tick
2812 in the case where the clock went backwards by a substantial amount such
2813 that the microsecond fraction of "now" was more than the microsecond
2814 fraction of "then" (but the whole seconds number was less).
2816 38. Added support for the libradius Radius client library this is found on
2817 FreeBSD (previously only the radiusclient library was supported).
2823 1. When certain lookups returned multiple values in the form name=value, the
2824 quoting of the values was not always being done properly. Specifically:
2825 (a) If the value started with a double quote, but contained no whitespace,
2827 (b) If the value contained whitespace other than a space character (i.e.
2828 tabs or newlines or carriage returns) it was not quoted.
2829 This fix has been applied to the mysql and pgsql lookups by writing a
2830 separate quoting function and calling it from the lookup code. The fix
2831 should probably also be applied to nisplus, ibase and oracle lookups, but
2832 since I cannot test any of those, I have not disturbed their existing code.
2834 2. A hit in the callout cache for a specific address caused a log line with no
2835 reason for rejecting RCPT. Now it says "Previous (cached) callout
2836 verification failure".
2838 3. There was an off-by-one bug in the queryprogram router. An over-long
2839 return line was truncated at 256 instead of 255 characters, thereby
2840 overflowing its buffer with the terminating zero. As well as fixing this, I
2841 have increased the buffer size to 1024 (and made a note to document this).
2843 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2844 when Exim is waiting for an SMTP response from a remote server, Exim
2845 restarts its select() call on the socket, thereby resetting its timeout.
2846 This is not a problem when such interrupts are rare. Somebody set up a cron
2847 job to run exiwhat every 2 minutes, which is less than the normal select()
2848 timeout (5 or 10 minutes). This meant that the select() timeout never
2849 kicked in because it was always reset. I have fixed this by comparing the
2850 time when an interrupt arrives with the time at the start of the first call
2851 to select(). If more time than the timeout has elapsed, the interrupt is
2852 treated as a timeout.
2854 5. Some internal re-factoring in preparation for the addition of Sieve
2855 extensions (by MH). In particular, the "personal" test is moved to a
2856 separate function, and given an option for scanning Cc: and Bcc: (which is
2857 not set for Exim filters).
2859 6. When Exim created an email address using the login of the caller as the
2860 local part (e.g. when creating a From: or Sender: header line), it was not
2861 quoting the local part when it contained special characters such as @.
2863 7. Installed new OpenBSD configuration files.
2865 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2866 try to make them clearer.
2868 9. Callout options, other than the timeout value, were being ignored when
2869 verifying sender addresses in header lines. For example, when using
2871 verify = header_sender/callout=no_cache
2873 the cache was (incorrectly) being used.
2875 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2876 only the Cygwin environment.
2878 11. When return_path_on_delivery was set as a log selector, if different remote
2879 addresses in the same message used different return paths and parallel
2880 remote delivery occurred, the wrong values would sometimes be logged.
2881 (Whenever a remote delivery process finished, the return path value from
2882 the most recently started remote delivery process was logged.)
2884 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2885 lines when AUTH and/or TLS are in use. This is the "received protocol"
2886 field. Exim used to use "asmtp" for authenticated SMTP, without any
2887 indication (in the protocol name) for TLS use. Now it follows the RFC and
2888 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2889 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2890 names appear in log lines as well as in Received: header lines.
2892 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2893 extensions, and comparison tests, and to fix some bugs.
2895 14. Changes to the "personal" filter test:
2897 (1) The test was buggy in that it was just doing the equivalent of
2898 "contains" tests on header lines. For example, if a user's address was
2899 anne@some.where, the "personal" test would incorrectly be true for
2901 To: susanne@some.where
2903 This test is now done by extracting each address from the header in turn,
2904 and checking the entire address. Other tests that are part of "personal"
2905 are now done using regular expressions (for example, to check local parts
2906 of addresses in From: header lines).
2908 (2) The list of non-personal local parts in From: addresses has been
2909 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2910 taken from the Sieve specification recommendations.
2912 (3) If the message contains any header line starting with "List-" it is
2913 treated as non-personal.
2915 (4) The test for "circular" in the Subject: header line has been removed
2916 because it now seems ill-conceived.
2918 15. Minor typos in src/EDITME comments corrected.
2920 16. Installed latest exipick from John Jetmore.
2922 17. If headers_add on a router specified a text string that was too long for
2923 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2924 of string_sprintf() is now avoided.
2926 18. $message_body_size was not set (it was always zero) when running the DATA
2927 ACL and the local_scan() function.
2929 19. For the "mail" command in an Exim filter, no default was being set for
2930 the once_repeat time, causing a random time value to be used if "once" was
2931 specified. (If the value happened to be <= 0, no repeat happened.) The
2932 default is now 0s, meaning "never repeat". The "vacation" command was OK
2933 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2934 (I found it when inspecting the code).
2936 20. There is now an overall timeout for performing a callout verification. It
2937 defaults to 4 times the callout timeout, which applies to individual SMTP
2938 commands during the callout. The overall timeout applies when there is more
2939 than one host that can be tried. The timeout is checked before trying the
2940 next host. This prevents very long delays if there are a large number of
2941 hosts and all are timing out (e.g. when the network connections are timing
2942 out). The value of the overall timeout can be changed by specifying an
2943 additional sub-option for "callout", called "maxwait". For example:
2945 verify = sender/callout=5s,maxwait=20s
2947 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2948 to the end before writing, but this should make it even safer).
2950 22. Exim was forgetting that it had advertised PIPELINING for the second and
2951 subsequent messages on an SMTP connection. It was also not resetting its
2952 memory on STARTTLS and an internal HELO.
2954 23. When Exim logs an SMTP synchronization error within a session, it now
2955 records whether PIPELINING has been advertised or not.
2957 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2958 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2959 rather than long int.
2961 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2963 26. Added the never_mail option to autoreply.
2969 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2970 crash if the getsockname() call failed; this can happen if a connection is
2971 closed very soon after it is established. The problem was simply in the
2972 order in which certain operations were done, causing Exim to try to write
2973 to the SMTP stream before it had set up the file descriptor. The bug has
2974 been fixed by making things happen in the correct order.
2980 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2981 before the connection was closed, thus losing the rejection response.
2983 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2984 some early Solaris releases, but causes trouble in current releases where
2985 socklen_t is defined.
2987 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2990 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2993 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2994 that releasing the top of it at the end releases what was used for sub-
2995 expansions (unless the block got too big). However, discard this block if
2996 the first thing is a variable or header, so that we can use its block when
2997 it is dynamic (useful for very large $message_headers, for example).
2999 6. Lookups now cache *every* query, not just the most recent. A new, separate
3000 store pool is used for this. It can be recovered when all lookup caches are
3001 flushed. Lookups now release memory at the end of their result strings.
3002 This has involved some general refactoring of the lookup sources.
3004 7. Some code has been added to the store_xxx() functions to reduce the amount
3005 of flapping under certain conditions.
3007 8. log_incoming_interface used to affect only the <= reception log lines. Now
3008 it causes the local interface and port to be added to several more SMTP log
3009 lines, for example "SMTP connection from", and rejection lines.
3011 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3013 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3015 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3016 could be overwritten at the end of the current message (or the start of a
3017 new message if it was set in a HELO ACL). The value is now preserved for
3018 the duration of the SMTP connection.
3020 12. If a transport had a headers_rewrite setting, and a matching header line
3021 contained an unqualified address, that address was qualified, even if it
3022 did not match any rewriting rules. The underlying bug was that the values
3023 of the flags that permit the existence of unqualified sender and recipient
3024 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3025 non-local messages, and by -bnq for local messages) were not being
3026 preserved with the message after it was received.
3028 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3029 "next input=" as part of the text comprising the host identity instead of
3030 the correct text. The code was using the same buffer for two different
3031 strings. However, depending on which order the printing function evaluated
3032 its arguments, the bug did not always show up. Under Linux, for example, my
3033 test suite worked just fine.
3035 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3036 doesn't work with some older versions of Perl. It has been changed to "my",
3037 which in any case is probably the better facility to use.
3039 15. A really picky compiler found some instances of statements for creating
3040 error messages that either had too many or two few arguments for the format
3043 16. The size of the buffer for calls to the DNS resolver has been increased
3044 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3045 for addresses that have a lot of PTR records. This alleviates a problem; it
3046 does not fully solve it.
3048 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3049 buffer now truncates the list and logs the incident, which is the same
3050 action as happens when Exim is looking up a host name and its aliases.
3051 Previously in this situation something unpredictable would happen;
3052 sometimes it was "internal error: store_reset failed".
3054 18. If a server dropped the connection unexpectedly when an Exim client was
3055 using GnuTLS and trying to read a response, the client delivery process
3056 crashed while trying to generate an error log message.
3058 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3059 string, for example:
3061 warn message = H1: something\nH2: something
3063 the text was added as a single header line from Exim's point of view
3064 though it ended up OK in the delivered message. However, searching for the
3065 second and subsequent header lines using $h_h2: did not work. This has been
3066 fixed. Similarly, if a system filter added multiple headers in this way,
3067 the routers could not see them.
3069 20. Expanded the error message when iplsearch is called with an invalid key to
3070 suggest using net-iplsearch in a host list.
3072 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3073 ACLs are no longer actually imposed (and a message to that effect is
3076 22. If a "gecos" field in a passwd entry contained escaped characters, in
3077 particular, if it contained a \" sequence, Exim got it wrong when building
3078 a From: or a Sender: header from that name. A second bug also caused
3079 incorrect handling when an unquoted " was present following a character
3080 that needed quoting.
3082 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3083 was not being matched caselessly.
3085 24. Arranged for all hyphens in the exim.8 source to be escaped with
3088 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3089 properly. Recipient callout cache records were still being keyed to include
3090 the sender, even when use_sender was set false. This led to far more
3091 callouts that were necessary. The sender is no longer included in the key
3092 when use_sender is false.
3094 26. Added "control = submission" modifier to ACLs.
3096 27. Added the ${base62d: operator to decode base 62 numbers.
3098 28. dnsdb lookups can now access SRV records.
3100 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3101 the configuration file.
3103 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3104 (-v) mode. This makes the output for a verbose queue run more intelligible.
3106 31. Added a use_postmaster feature to recipient callouts.
3108 32. Added the $body_zerocount variable, containing the number of binary zero
3109 bytes in the message body.
3111 33. The time of last modification of the "new" subdirectory is now used as the
3112 "mailbox time last read" when there is a quota error for a maildir
3115 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3117 35. Added +ignore_unknown as a special item in host lists.
3119 36. Code for decoding IPv6 addresses in host lists is now included, even if
3120 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3121 address was recognized as an IP address, but was then not correctly decoded
3122 into binary, causing unexpected and incorrect effects when compared with
3129 1. Very minor rewording of debugging text in manualroute to say "list of
3130 hosts" instead of "hostlist".
3132 2. If verify=header_syntax was set, and a header line with an unqualified
3133 address (no domain) and a large number of spaces between the end of the
3134 name and the colon was received, the reception process suffered a buffer
3135 overflow, and (when I tested it) crashed. This was caused by some obsolete
3136 code that should have been removed. The fix is to remove it!
3138 3. When running in the test harness, delay a bit after writing a bounce
3139 message to get a bit more predictability in the log output.
3141 4. Added a call to search_tidyup() just before forking a reception process. In
3142 theory, someone could use a lookup in the expansion of smtp_accept_max_
3143 per_host which, without the tidyup, could leave open a database connection.
3145 5. Added the variables $recipient_data and $sender_data which get set from a
3146 lookup success in an ACL "recipients" or "senders" condition, or a router
3147 "senders" option, similar to $domain_data and $local_part_data.
3149 6. Moved the writing of debug_print from before to after the "senders" test
3152 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3153 problems for message scanning, either using a data ACL, or using
3154 local_scan() because the Received: header was not generated till after they
3155 were called (in order to set the time as the time of reception completion).
3156 I have revised the way this works. The header is now generated after the
3157 body is received, but before the ACL or local_scan() are called. After they
3158 are run, the timestamp in the header is updated.
3164 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3165 before starting a queue runner without re-exec. This happened only when
3166 deliver_drop_privilege was set or when the Exim user was set to root. The
3167 effect of the bug was that timeouts during subsequent deliveries caused
3168 crashes instead of being properly handled. The handler is now left at its
3169 default (and expected) setting.
3171 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3172 message, again when deliver_drop_privilege is set or Exim is run as root.
3173 The bug described in (1) was not present in this case, but the tidying up
3174 of the other signals was missing. I have made the two cases consistent.
3176 3. The ignore_target_hosts setting on a manualroute router was being ignored
3177 for hosts that were looked up using the /MX notation.
3179 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3182 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3183 operated on the sender address. After changing the $sender_address to <>
3184 for the sender address verify, Exim was re-instated it as the original
3185 (before rewriting) address, but remembering that it had rewritten it, so it
3186 wasn't rewriting it again. This bug also had the effect of breaking the
3187 sender address verification caching when the sender address was rewritten.
3189 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3190 This has been changed so that if the ip literal address matches
3191 ignore_target_hosts, the router declines.
3193 7. Added expansion conditions match_domain, match_address, and match_local_
3194 part (NOT match_host).
3196 8. The placeholder for the Received: header didn't have a length field set.
3198 9. Added code to Exim itself and to exim_lock to test for a specific race
3199 condition that could lead to file corruption when using MBX delivery. The
3200 issue is with the lockfile that is created in /tmp. If this file is removed
3201 after a process has opened it but before that process has acquired a lock,
3202 there is the potential for a second process to recreate the file and also
3203 acquire a lock. This could lead to two Exim processes writing to the file
3204 at the same time. The added code performs the same test as UW imapd; it
3205 checks after acquiring the lock that its file descriptor still refers to
3206 the same named file.
3208 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3209 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3212 11. Added the smtp_active_hostname option. If used, this will typically be made
3213 to depend on the incoming interface address. Because $interface_address is
3214 not set up until the daemon has forked a reception process, error responses
3215 that can happen earlier (such as "too many connections") no longer contain
3218 12. If an expansion in a condition on a "warn" statement fails because a lookup
3219 defers, the "warn" statement is abandoned, and the next ACL statement is
3220 processed. Previously this caused the whole ACL to be aborted.
3222 13. Added the iplsearch lookup type.
3224 14. Added ident_timeout as a log selector.
3226 15. Added tls_certificate_verified as a log selector.
3228 16. Added a global option tls_require_ciphers (compare the smtp transport
3229 option of the same name). This controls incoming TLS connections.
3231 17. I finally figured out how to make tls_require_ciphers do a similar thing
3232 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3233 before starting the TLS session.
3235 18. Tabs are now shown as \t in -bP output.
3237 19. If the log selector return_path_on_delivery was set, Exim crashed when
3238 bouncing a message because it had too many Received: header lines.
3240 20. If two routers both had headers_remove settings, and the first one included
3241 a superfluous trailing colon, the final name in the first list and the
3242 first name in the second list were incorrectly joined into one item (with a
3243 colon in the middle).
3249 1. Added -C and -D options to the exinext utility, mainly to make it easier
3250 to include in the automated testing, but these could be helpful when
3251 multiple configurations are in use.
3253 2. The exinext utility was not formatting the output nicely when there was
3254 an alternate port involved in the retry record key, nor when there was a
3255 message id as well (for retries that were specific to a specific message
3256 and a specific host). It was also confused by IPv6 addresses, because of
3257 the additional colons they contain. I have fixed the IPv4 problem, and
3258 patched it up to do a reasonable job for IPv6.
3260 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3261 delivery, the log line now contains "pipelined" if PIPELINING was used.
3263 4. An SMTP transport process used to panic and die if the bind() call to set
3264 an explicit outgoing interface failed. This has been changed; it is now
3265 treated in the same way as a connect() failure.
3267 5. A reference to $sender_host_name in the part of a conditional expansion
3268 that was being skipped was still causing a DNS lookup. This no longer
3271 6. The def: expansion condition was not recognizing references to header lines
3272 that used bh_ and bheader_.
3274 7. Added the _cache feature to named lists.
3276 8. The code for checking quota_filecount in the appendfile transport was
3277 allowing one more file than it should have been.
3279 9. For compatibility with Sendmail, the command line option
3287 and sets the incoming protocol and host name (for trusted callers). The
3288 host name and its colon can be omitted when only the protocol is to be set.
3289 Note the Exim already has two private options, -pd and -ps, that refer to
3290 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3291 "s", but I don't think that's a major issue.
3293 10. A number of refactoring changes to the code, none of which should affect
3296 (a) The number of logging options was getting close to filling up the
3297 32-bit word that was used as a bit map. I have split them into two classes:
3298 those that are passed in the argument to log_write(), and those that are
3299 only ever tested independently outside of that function. These are now in
3300 separate 32-bit words, so there is plenty of room for expansion again.
3301 There is no change in the user interface or the logging behaviour.
3303 (b) When building, for example, log lines, the code previously used a
3304 macro that called string_cat() twice, in order to add two strings. This is
3305 not really sufficiently general. Furthermore, there was one instance where
3306 it was actually wrong because one of the argument was used twice, and in
3307 one call a function was used. (As it happened, calling the function twice
3308 did not affect the overall behaviour.) The macro has been replaced by a
3309 function that can join an arbitrary number of extra strings onto a growing
3312 (c) The code for expansion conditions now uses a table and a binary chop
3313 instead of a serial search (which was left over from when there were very
3314 few conditions). Also, it now recognizes conditions like "pam" even when
3315 the relevant support is not compiled in: a suitably worded error message is
3316 given if an attempt is made to use such a condition.
3318 11. Added ${time_interval:xxxxx}.
3320 12. A bug was causing one of the ddress fields not to be passed back correctly
3321 from remote delivery subprocesses. The field in question was not being
3322 subsequently used, so this caused to problems in practice.
3324 13. Added new log selectors queue_time and deliver_time.
3326 14. Might have fixed a bug in maildirsizefile handling that threw up
3327 "unexpected character" debug warnings, and recalculated the data
3328 unnecessarily. In any case, I expanded the warning message to give more
3331 15. Added the message "Restricted characters in address" to the statements in
3332 the default ACL that block characters like @ and % in local parts.
3334 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3335 Three changes have been made:
3337 (a) There was a serious bug; a negative response to MAIL caused the whole
3338 recipient domain to be cached as invalid, thereby blocking all messages
3339 to all local parts at the same domain, from all senders. This bug has
3340 been fixed. The domain is no longer cached after a negative response to
3341 MAIL if the sender used is not empty.
3343 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3346 (c) A new callout option, "use_sender" has been added for people who want
3347 the modified behaviour.
3353 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3356 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3357 indeed breaks things for older releases.
3359 3. Added additional logging to the case where there is a problem reading data
3360 from a filter that is running in a subprocess using a pipe, in order to
3361 try to track down a specific problem.
3363 4. Testing facility fudge: when running in the test harness and attempting
3364 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3365 getting "No route to host". Convert this to a timeout.
3367 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3370 6. Some OS don't have socklen_t but use size_t instead. This affects the
3371 fifth argument of getsockopt() amongst other things. This is now
3372 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3373 can be set for individual OS. I have set it for SunOS5, OSF1, and
3374 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3375 some earlier ones do not.
3377 7. Change 4.30/15 was not doing the test caselessly.
3379 8. The standard form for an IPv6 address literal was being rejected by address
3380 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3381 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3382 this, as well as the form without the "IPv6" on the front (but only when
3383 address literals are enabled, of course).
3385 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3387 10. Exim crashed if a message with an empty sender address specified by -f
3388 encountered a router with an errors_to setting. This could be provoked only
3389 by a command such as
3393 where an empty string was supplied; "<>" did not hit this bug.
3395 11. Installed PCRE release 4.5.
3397 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3398 remained set. It is now erased.
3400 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3401 times from message ids (which are base 36 rather than the normal 62).
3403 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3404 were being counted as actual protocol errors, and logged if the log
3405 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3406 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3407 and DATA following a set of rejected RCPTs do not count as protocol errors.
3408 In other words, Exim assumes they were pipelined, though this may not
3409 actually be the case. Of course, in all cases the client gets an
3410 appropriate error code.
3412 15. If a lookup fails in an ACL condition, a message about the failure may
3413 be available; it is used if testing the ACL cannot continue, because most
3414 such messages specify what the cause of the deferral is. However, some
3415 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3416 that caused an old message to be retained and used if a later statement
3417 caused a defer, replacing the real cause of the deferral.
3419 16. If an IP address had so many PTR records that the DNS lookup buffer
3420 was not large enough to hold them, Exim could crash while trying to process
3421 the truncated data. It now detects and logs this case.
3423 17. Further to 4.21/58, another change has been made: if (and only if) the
3424 first line of a message (the first header line) ends with CRLF, a bare LF
3425 in a subsequent header line has a space inserted after it, so as not to
3426 terminate the header.
3428 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3429 unnecessarily, used atoi() instead of strtol(), and didn't check the
3430 termination when getting file sizes from file names by regex.
3432 19. Completely re-implemented the support for maildirsize files, in the light
3433 of a number of problems with the previous contributed implementation
3434 (4.30/29). In particular:
3436 . If the quota is zero, the maildirsize file is maintained, but no quota is
3439 . If the maildir directory does not exist, it is created before any attempt
3440 to write a maildirsize file.
3442 . The quota value in the file is just a cache; if the quota is changed in
3443 the transport, the new value overrides.
3445 . A regular expression is available for excluding directories from the
3448 20. The autoreply transport checks the characters in options that define the
3449 message's headers; it allows continued headers, but it was checking with
3450 isspace() after an embedded newline instead of explicitly looking for a
3453 21. If all the "regular" hosts to which an address was routed had passed their
3454 expiry times, and had not reached their retry times, the address was
3455 bounced, even if fallback hosts were defined. Now Exim should go on to try
3458 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3459 equivalent code in the SMTP transport. Some hosts send humungous responses
3460 to HELO/EHLO, more than 1024 it seems.
3462 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3463 gives compiler warnings in some environments. I've now done it "properly",
3466 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3467 (because of problems with the built-in one) was declared to return uschar *
3468 instead of char *, causing compiler failure.
3470 25. Fixed a file descriptor leak when processing alias/forward files.
3472 26. Fixed a minor format string issue in dbfn.c.
3474 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3476 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3477 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3480 29. When a sender address is verified, it is cached, to save repeating the test
3481 when there is more than one recipient in a message. However, when the
3482 verification involves a callout, it is possible for different callout
3483 options to be set for different recipients. It is too complicated to keep
3484 track of this in the cache, so now Exim always runs a verification when a
3485 callout is required, relying on the callout cache for the optimization.
3486 The overhead is duplication of the address routing, but this should not be
3489 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3490 to be verified with callout=postmaster, and the main callout worked but the
3491 postmaster check failed, the verification correctly failed. However, if a
3492 subsequent RCPT command asked for sender verification *without* the
3493 postmaster check, incorrect caching caused this verification also to fail,
3496 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3497 it was not caching the DNS options (qualify_single, search_parents) that
3498 were used when the lookup failed. A subsequent lookup with different
3499 options therefore always gave the same answer, though there were cases
3500 where it should not have. (Example: a "domains = !$mx_any" option on a
3501 dnslookup router: the "domains" option is always processed without any
3502 widening, but the router might have qualify_single set.) Now Exim uses the
3503 cached value only when the same options are set.
3505 32. Added John Jetmore's "exipick" utility to the distribution.
3507 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3508 than a timeout (e.g. a certificate is required, and is not provided), an
3509 Exim server now closes the connection immediately. Previously it waited for
3510 the client to close - but if the client is SSL, it seems that they each
3511 wait for each other, leading to a delay before one of them times out.
3513 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3514 maintained 0.8.x compatibility because I don't think many are using it, and
3515 it is clearly obsolete.
3517 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3520 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3521 client certificate was expired. A simple patch fixes this, though I don't
3522 understand the full logic of why the verify callback is called multiple
3525 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3526 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3527 which causes problems with some clients (such as the Certicom SSL Plus
3528 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3529 disables the coutermeasure allowing Eudora to connect."
3531 38. Exim was not checking that a write() to a log file succeeded. This could
3532 lead to Bad Things if a log got too big, in particular if it hit a file
3533 size limit. Exim now panics and dies if it cannot write to a log file, just
3534 as it does if it cannot open a log file.
3536 39. Modified OS/Makefile-Linux so that it now contains
3538 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3540 The two -D definitions ensure that Exim is compiled with large file
3541 support, which makes it possible to handle log files that are bigger than
3544 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3545 instance) a domain was checked against a named list that involved a lookup,
3546 causing $domain_data to be set, then another domain was checked against the
3547 same list, then the first domain was re-checked, the value of $domain_data
3548 after the final check could be wrong. In particular, if the second check
3549 failed, it could be set empty. This bug probably also applied to
3552 41. The strip_trailing_dot option was not being applied to the address given
3553 with the -f command-line option.
3555 42. The code for reading a message's header from the spool was incrementing
3556 $received_count, but never initializing it. This meant that the value was
3557 incorrect (doubled) while delivering a message in the same process in which
3558 it was received. In the most common configuration of Exim, this never
3559 happens - a fresh exec is done - but it can happen when
3560 deliver_drop_privilege is set.
3562 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3563 it now includes up to 150 characters of the unexpected data in the log
3566 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3567 and building data strings. The size of both of these buffers was 10 000
3568 bytes - far larger than anybody would *ever* want, thought I. Needless to
3569 say, somebody hit the limit. I have increased the maximum line length to
3570 20 000 and the maximum data length of concatenated lines to 100 000. I have
3571 also fixed two bugs, because there was no checking on these buffers. Tsk,
3572 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3573 buffer is too small.
3575 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3576 lsearch lookups. Now it does.
3578 46. When parsing a route_list item in a manualroute router, a fixed-length
3579 buffer was used for the list of hosts. I made this 1024 bytes long,
3580 thinking that nobody would ever have a list of hosts that long. Wrong.
3581 Somebody had a whole pile of complicated expansion conditions, and the
3582 string was silently truncated, leading to an expansion error. It turns out
3583 that it is easier to change to an unlimited length (owing to other changes
3584 that have happened since this code was originally written) than to build
3585 structure for giving a limitation error. The length of the item that
3586 expands into the list of hosts is now unlimited.
3588 47. The lsearch lookup could not handle data where the length of text line was
3589 more than 4095 characters. Such lines were truncated, leading to shortened
3590 data being returned. It should now handle lines of any length.
3592 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3593 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3596 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3597 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3598 no longer happens in comment lines. A list of replacements is now placed
3599 at the head of all of the source files, except those whose only change is
3600 to replace PERL_COMMAND in the very first #! line.
3602 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3603 messages on the queue, with a bottom-up merge sort, using code contributed
3604 by Michael Haardt. This should make operations like -bp somewhat faster on
3605 large queues. It won't affect queue runners, except when queue_run_in_order
3608 51. Installed eximstats 1.31 in the distribution.
3610 52. Added support for SRV lookups to the dnslookup router.
3612 53. If an ACL referred to $message_body or $message_body_end, the value was not
3613 reset for any messages that followed in the same SMTP session.
3615 54. The store-handling optimization for building very long strings was not
3616 differentiating between the different store pools. I don't think this
3617 actually made any difference in practice, but I've tidied it.
3619 55. While running the routers to verify a sender address, $sender_address
3620 was still set to the sender address. This is wrong, because when routing to
3621 send a bounce to the sender, it would be empty. Therefore, I have changed
3622 it so that, while verifying a sender address, $sender_address is set to <>.
3623 (There is no change to what happens when verifying a recipient address.)
3625 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3626 target A or AAAA records (if not already returned) without resetting the
3627 qualify_single or search_parents options of the DNS resolver. These are
3628 inappropriate in this case because the targets of MX and SRV records must
3629 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3630 target that, when qualified, matched something in the local domain. These
3631 two options are now turned off when doing these lookups.
3633 57. It seems that at least some releases of Reiserfs (which does not have the
3634 concept of a fixed number of inodes) returns zero and not -1 for the
3635 number of available inodes. This interacted badly with check_spool_inodes,
3636 which assumed that -1 was the "no such thing" setting. What I have done is
3637 to check that the total number of inodes is greater than zero before doing
3638 the test of how many are available.
3640 58. When a "warn" ACL statement has a log_message modifier, the message is
3641 remembered, and not repeated. This is to avoid a lot of repetition when a
3642 message has many recipients that cause the same warning to be written.
3643 Howewer, Exim was preserving the list of already written lines for an
3644 entire SMTP session, which doesn't seem right. The memory is now reset if a
3645 new message is started.
3647 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3648 debugging output unless log_rewrite was also set.
3650 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3651 of (int)(handle) when we know that handle contains (void *)(-1).
3653 61. The Exim daemon panic-logs an error return when it closes the incoming
3654 connection. However "connection reset by peer" seems to be common, and
3655 isn't really an error worthy of noting specially, so that particular error
3658 62. When Exim is trying to find all the local interfaces, it used to panic and
3659 die if the ioctl to get the interface flags failed. However, it seems that
3660 on at least one OS (Solaris 9) it is possible to have an interface that is
3661 included in the list of interfaces, but for which you get a failure error
3662 for this call. This happens when the interface is not "plumbed" into a
3663 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3664 failure of the "get flags" call assumes that the interface is down.
3666 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3667 makes life easier for people who are doing arithmetic on fields extracted
3668 from dates, where you often get leading zeros that should not be
3669 interpreted as octal.
3671 64. Added qualify_domain to the redirect router, to override the global
3674 65. If a pathologically long header line contained very many addresses (the
3675 report of this problem mentioned 10 000) and each of them was rewritten,
3676 Exim could use up a very large amount of memory. (It kept on making new
3677 copies of the header line as it rewrote, and never released the old ones.)
3678 At the expense of a bit more processing, the header rewriting function has
3679 been changed so that it no longer eats memory in this way.
3681 66. The generation of the Received: header has been moved from the time that a
3682 message starts to be received, to the time that it finishes. The timestamp
3683 in the Received: header should now be very close to that of the <= log
3684 line. There are two side-effects of this change:
3686 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3687 logged header lines no longer include the local Received: line, because
3688 it has not yet been created. The same applies to a copy of the message
3689 that is returned to a non-SMTP sender when a message is rejected.
3691 (b) When a filter file is tested using -bf, no additional Received: header
3692 is added to the test message. After some thought, I decided that this
3695 This change does not affect the value of $received_for. It is still set
3696 after address rewriting, but before local_scan() is called.
3698 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3700 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3701 gave an unhelpful panic error message, and a defer error. I have managed to
3702 change this behaviour so that it now rejects any supplied certificate,
3703 which seems right, as the list of acceptable certificates is empty.
3705 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3706 gave an unhelpful defer error. I have not managed to make this reject any
3707 supplied certificates, but the error message it gives is "no certificate
3708 supplied", which is not helpful.
3710 70. exigrep's output now also includes lines that are not associated with any
3711 message, but which match the given pattern. Implemented by a patch from
3712 Martin Sluka, which also tidied up the Perl a bit.
3714 71. Recipient callout verification, like sender verification, was using <> in
3715 the MAIL FROM command. This isn't really the right thing, since the actual
3716 sender may affect whether the remote host accepts the recipient or not. I
3717 have changed it to use the actual sender in the callout; this means that
3718 the cache record is now keyed on a recipient/sender pair, not just the
3719 recipient address. There doesn't seem to be a real danger of callout loops,
3720 since a callout by the remote host to check the sender would use <>.
3721 [SEE ABOVE: changed after hitting problems.]
3723 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3724 temporary errors. However, in the case of such a code being given after
3725 the end of a data transmission (i.e. after ".") Exim was failing to write
3726 a retry record for the message. (Yes, there was some broken host that was
3727 actually sending 8xx at this point.)
3729 73. An unknown lookup type in a host list could cause Exim to panic-die when
3730 the list was checked. (An example that provoked this was putting <; in the
3731 middle of a list instead of at the start.) If this happened during a DATA
3732 ACL check, a -D file could be left lying around. This kind of configuration
3733 error no longer causes Exim to die; instead it causes a defer errror. The
3734 incident is still logged to the main and panic logs.
3736 74. Buglet left over from Exim 3 conversion. The message "too many messages
3737 in one connection" was written to the rejectlog but not the mainlog, except
3738 when address rewriting (yes!) was being logged.
3740 75. Added write_rejectlog option.
3742 76. When a system filter was run not as root (that is, when system_filter_user
3743 was set), the values of the $n variables were not being returned to the
3744 main process; thus, they were not subsequently available in the $sn
3747 77. Added +return_path_on_delivery log selector.
3749 78. A connection timeout was being treated differently from recipients deferred
3750 when testing hosts_max_try with a message that was older than the host's
3751 retry timeout. (The host should not be counted, thus allowing all hosts to
3752 be tried at least once before bouncing.) This may have been the cause of an
3753 occasionally reported bug whereby a message would remain on the queue
3754 longer than the retry timeout, but would be bounced if a delivery was
3755 forced. I say "may" because I never totally pinned down the problem;
3756 setting up timeout/retry tests is difficult. See also the next item.
3758 79. The ultimate address timeout was not being applied to errors that involved
3759 a combination of host plus message (for example, a timeout on a MAIL
3760 command). When an address resolved to a number of possible hosts, and they
3761 were not all tried for each delivery (e.g. because of hosts_max_try), a
3762 message could remain on the queue longer than the retry timeout.
3764 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3767 81. Fixed an obscure SMTP outgoing bug which required at least the following
3768 conditions: (a) there was another message waiting for the same server;
3769 (b) the server returned 5xx to all RCPT commands in the first message so
3770 that the message was not completed; (c) the server dropped the connection
3771 or gave a negative response to the RSET that Exim sends to abort the
3772 transaction. The observed case was a dropped connection after DATA that had
3773 been sent in pipelining mode. That is, the server had advertised PIPELINING
3774 but was not implementing it correctly. The effect of the bug was incorrect
3775 behaviour, such as trying another host, and this could lead to a crash.
3781 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3782 and daemon.c were passed as pointers to ints; they should have been
3783 pointers to socklen_t variables (which are typically unsigned ints).
3785 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3788 3. Fixed a really odd bug that affected only the testing scheme; patching a
3789 certain fixed string in the binary changed the value of another string that
3790 happened to be identical to the end of the original first string.
3792 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3793 name", it returns that address as the IP address. On some operating
3794 systems (e.g. Solaris), it also passes back the IP address string as the
3795 "host name". However, on others (e.g. Linux), it passes back an empty
3796 string. Exim wasn't checking for this, and was changing the host name to an
3797 empty string, assuming it had been canonicized.
3799 5. Although rare, it is permitted to have more than one PTR record for a given
3800 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3801 all the names associated with an address, because they do in Solaris.
3802 However, it seems that they do not in Linux for data that comes from the
3803 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3804 I found this out when I moved to a new Linux workstation and tried to run
3805 the Exim test suite.
3807 To get round this problem I have changed the code so that it now does its
3808 own call to the DNS to look up PTR records when searching for a host name.
3809 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3810 addresses that are only in /etc/hosts are still found.
3812 This behaviour is, however, controlled by an option called host_lookup_
3813 order, which defaults to "bydns:byaddr". If people want to use the other
3814 order, or indeed, just use one or the other means of lookup, they can
3815 specify it in this variable.
3817 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3818 some operating systems, this comes back from gethostbyaddr() as an empty
3819 string, and this is what Exim used to test for. However, it seems that in
3820 other systems, "." is yielded. Exim now tests for this case too.
3822 7. The values of check_spool_space and check_log_space are now held internally
3823 as a number of kilobytes instead of an absolute number of bytes. If a
3824 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3825 kilobyte. This means that much larger values can be stored.
3827 8. Exim monitor: an attempt to get the action menu when not actually pointing
3828 at a message produces an empty menu entitled "No message selected". This
3829 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3830 no entries in it ("Shell widget menu has zero width and/or height"). So I
3831 have added a single, blank menu entry in this case.
3833 9. Added ${quote_local_part.
3835 10. MIME decoding is now applied to the contents of Subject: header lines when
3838 11. Now that a reference to $sender_host_address automatically causes a reverse
3839 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3840 host lookup before query-style lookups in lists that might use this
3841 variable. This has therefore been abolished, and the "net-" prefix is no
3842 longer necessary for query-style lookups.
3844 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3845 have been a typo for LFLAGS, so it has been changed.
3847 13. The install script calls Exim with "-C /dev/null" in order to find the
3848 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3849 to be output. Howeve, since Exim outputs its version number before the
3850 error, it didn't break the script. It just looked ugly. I fixed this by
3851 always allowing "-C /dev/null" if the caller is root.
3853 14. Ignore overlarge ACL variable number when reading spool file - insurance
3854 against a later release with more variables having written the file.
3856 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3857 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3858 this, as well as the form without the "IPv6" on the front.
3860 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3861 OS/Makefile-Darwin file.
3863 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3864 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3866 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3867 variable. However, some OS use ints for inodes. I've added cast to long int
3868 to get rid of the compiler warning.
3870 19. I had forgotten to lock out "/../" in configuration file names when
3871 ALT_CONFIG_PREFIX was set.
3873 20. Routers used for verification do not need to specify transports. However,
3874 if such a router generated a host list, and callout was configured, Exim
3875 crashed, because it could not find a port number from the (non-existent)
3876 transport. It now assumes port 25 in this circumstance.
3878 21. Added the -t option to exigrep.
3880 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3881 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3882 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3883 them is set, LOOKUP_LSEARCH is forced.
3885 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3887 24. Added sender and host information to the "rejected by local_scan()" log
3888 line; previously there was no indication of these.
3890 25. Added .include_if_exists.
3892 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3893 Linux. It turns out that not all file systems support this. Apparently some
3894 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3895 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3896 not supported on the file descriptor, is now ignored when Exim is trying to
3897 sync a directory. This applies only to Linux.
3899 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3901 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3902 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3905 29. Added support for maildirsize files from supplied patch (modified a bit).
3907 30. The use of :fail: followed by an empty string could lead Exim to respond to
3908 sender verification failures with (e.g.):
3910 550 Verification failed for <xxx>
3911 550 Sender verify failed
3913 where the first response line was missing the '-' that indicates it is not
3914 the final line of the response.
3916 31. The loop for finding the name of the user that called Exim had a hardwired
3917 limit of 10; it now uses the value of finduser_retries, which is used for
3918 all other user lookups.
3920 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3923 33. Exim was neglecting to zero errno before one call of strtol() when
3924 expanding a string and expecting an integer value. On some systems this
3925 resulted in spurious "integer overflow" errors. Also, it was casting the
3926 result into an int without checking.
3928 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3929 did not work. The code looks as if it has *never* worked, though it appears
3930 to have been documented since at least releast 1.62. I have made it work.
3932 35. The "timeout_DNS" error in retry rules, also documented since at least
3933 1.62, also never worked. As it isn't clear exactly what this means, and
3934 clearly it isn't a major issue, I have abolished the feature by treating it
3935 as "timeout", and writing a warning to the main and panic logs.
3937 36. The display of retry rules for -brt wasn't always showing the error code
3940 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3941 timeout_connect_A, timeout_connect_MX.
3943 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3944 to the empty sender.
3946 39. The daemon was not analysing the content of -oX till after it had closed
3947 stderr and disconnected from the controlling terminal. This meant that any
3948 syntax errors were only noted on the panic log, and the return code from
3949 the command was 0. By re-arranging the code a little, I've made the
3950 decoding happen first, so such errors now appear on stderr, and the return
3951 code is 1. However, the actual setting up of the sockets still happens in
3952 the disconnected process, so errors there are still only recorded on the
3955 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3956 connections (as happens on some IP stacks) was logged at start up time as
3957 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3958 it from "IPv6 and IPv4", which means that two separate sockets are being
3961 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3962 says whether AF_INET or AF_INET6 was passed as an argument.
3964 42. Exiwhat output was messed up when time zones were included in log
3967 43. Exiwhat now gives more information about the daemon's listening ports,
3968 and whether -tls-on-connect was used.
3970 44. The "port" option of the smtp transport is now expanded.
3972 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3973 silently ignored. Now an error message is written to the main and panic
3976 46. There's a new ACL modifier called "logwrite" which writes to a log file
3977 as soon as it is encountered.
3979 47. Added $local_user_uid and $local_user_gid at routing time.
3981 48. Exim crashed when trying to verify a sender address that was being
3984 49. Exim was recognizing only a space character after ".include". It now also
3985 recognizes a tab character.
3987 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3988 extracting the relevant information from the specification. The man page no
3989 longer contains scrambled data for the -d option, and I've added a section
3990 at the front about calling Exim under different names.
3992 51. Added "extra_headers" argument to the "mail" command in filter files.
3994 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3997 53. Installed eximstats 1.29.
3999 54. Added transport_filter_timeout as a generic transport option.
4001 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4002 Cc: header lines. This was required by RFC 822, but it not required by RFC
4005 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4006 incoming messages that did not have them. Now it does so only if the
4007 message originates locally, that is, if there is no associated remote host
4008 address. When Resent- header lines are present, this applies to the Resent-
4009 lines rather than the non-Resent- lines.
4011 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4012 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4014 58. Messages for configuration errors now include the name of the main
4015 configuration file - useful now that there may be more than one file in a
4016 list (.included file names were always shown).
4018 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4019 for those rare installations that do not start the daemon as root or run it
4020 setuid root. I've cut out the call to initgroups() if the daemon is not
4023 60. The Exim user and group can now be bound into the binary as text strings
4024 that are looked up at the start of Exim's processing.
4026 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4028 62. Added $mailstore_basename variable.
4030 63. Installed patch to sieve.c from Michael Haardt.
4032 64. When Exim failed to open the panic log after failing to open the main log,
4033 the original message it was trying to log was written to stderr and debug
4034 output, but if they were not available (the usual case in production), it
4035 was lost. Now it is written to syslog before the two lines that record the
4036 failures to open the logs.
4038 65. Users' Exim filters run in subprocesses under the user's uid. It is
4039 possible for a "deliver" command or an alias in a "personal" command to
4040 provoke an address rewrite. If logging of address rewriting is configured,
4041 this fails because the process is not running as root or exim. There may be
4042 a better way of dealing with this, but for the moment (because 4.30 needs
4043 to be released), I have disabled address rewrite logging when running a
4044 filter in a non-root, non-exim process.
4050 1. The buildconfig auxiliary program wasn't quoting the value set for
4051 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4052 not defined. This bug was masked in 4.22 by the effect that was fixed in
4055 2. Some messages that were rejected after a message id was allocated were
4056 shown as "incomplete" by exigrep. It no longer does this for messages that
4057 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4059 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4060 have allow_domain_literals set, the ID did not get logged in the <= line.
4061 Domain literals are now always recognized in Message-ID: header lines.
4063 4. The first argument for a ${extract expansion item is the key name or field
4064 number. Leading and trailing spaces in this item were not being ignored,
4065 causing some misleading effects.
4067 5. When deliver_drop_privilege was set, single queue runner processes started
4068 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4069 same command in the process it spins off) were not dropping privilege.
4071 6. When the daemon running as "exim" started a queue runner, it always
4072 re-executed Exim in the spun-off process. This is a waste of effort when
4073 deliver_drop_privilege is set. The new process now just calls the
4074 queue-runner function directly.
4080 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4083 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4084 ignored. Though the use of -oP was forcing the writing of a pid file, it
4085 was always written to the default place.
4087 3. If the message "no IP address found for host xxxx" is generated during
4088 incoming verification, it is now followed by identification of the incoming
4089 connection (so you can more easily find what provoked it).
4091 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4093 5. Added some features to "harden" Exim a bit more against certain attacks:
4095 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4096 be put in Local/Makefile. This is like the never_users runtime option,
4097 but it cannot be overridden. The default setting is "root".
4099 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4100 prefix string with which any file named in a -C command line option
4103 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4104 is retained for -C and -D only if the caller of Exim is root. Without
4105 it, the exim user may also use -C and -D and retain privilege.
4107 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4108 command line option is disabled.
4110 6. Macro names set by the -D option must start with an upper case letter, just
4111 like macro names defined in the configuration file.
4113 7. Added "dereference=" facility to LDAP.
4115 8. Two instances of the typo "uknown" in the source files are fixed.
4117 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4118 the Configure-Makefile script screwed up while processing it.
4120 10. Incorporated PCRE 4.4.
4122 11. The SMTP synchronization check was not operating right at the start of an
4123 SMTP session. For example, it could not catch a HELO sent before the client
4124 waited for the greeting. There is now a check for outstanding input at the
4125 point when the greeting is written. Because of the duplex, asynchronous
4126 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4127 way, but not yet received, when the check is performed.
4129 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4130 on TCP/IP sockets, because this apparently causes some broken clients to
4133 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4134 unchanged) from the Cygwin maintainer.
4136 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4137 was supported instead of testing for mbx. Effectively a typo.
4139 15. The spa authenticator server code was not checking that the input it
4140 received was valid base64.
4142 16. The debug output line for the "set" modifier in ACLs was not showing the
4143 name of the variable that was being set.
4145 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4147 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4148 lookup on its own. Something else was needed to trigger the lookup. For
4149 example, a match in host_lookup or the need for a host name in a host list.
4150 Now, if $sender_host_name is referenced and the host name has not yet been
4151 looked up, a lookup is performed. If the lookup fails, the variable remains
4152 empty, and $host_lookup_failed is set to "1".
4154 19. Added "eqi" as a case-independent comparison operator.
4156 20. The saslauthd authentication condition could segfault if neither service
4157 nor realm was specified.
4159 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4160 error message that was logged was misleading, and incoming SMTP
4161 connections were dropped. The message is now more accurate, and temporary
4162 errors are given to SMTP connections.
4164 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4165 (except RSET) with a 421 error, until QUIT is received. However, it was
4166 failing to send a response to QUIT.
4168 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4169 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4170 ACL is now run after the helo_try_verify_hosts code.
4172 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4173 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4174 case-independent, so other case variants are also recognized). Apparently
4175 some systems use these upper case variants.
4177 25. If more than two messages were waiting for the same host, and a transport
4178 filter was specified for the transport, Exim sent two messages over the
4179 same TCP/IP connection, and then failed with "socket operation on non-
4180 socket" when it tried to send the third.
4182 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4184 27. The extern definition of crypt16() in expand.c was not being excluded when
4185 the OS had its own crypt16() function.
4187 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4188 as a preferred synonym for return_size_limit, both as an option and as an
4191 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4193 30. Changed the default configuration ACL to relax the local part checking rule
4194 for addresses that are not in any local domains. For these addresses,
4195 slashes and pipe symbols are allowed within local parts, but the sequence
4196 /../ is explicitly forbidden.
4198 31. SPA server authentication was not clearing the challenge buffer before
4201 32. log_message in a "warn" ACL statement was writing to the reject log as
4202 well as to the main log, which contradicts the documentation and doesn't
4203 seem right (because no rejection is happening). So I have stopped it.
4205 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4206 However, I am unable to do any testing of this.
4208 34. Fixed an infelicity in the appendfile transport. When checking directories
4209 for a mailbox, to see if any needed to be created, it was accidentally
4210 using path names with one or more superfluous leading slashes; tracing
4211 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4213 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4214 contents are added to the log line that is written for every discarded
4215 recipient. (Previously a log_message setting was ignored.)
4217 36. The ${quote: operator now quotes the string if it is empty.
4219 37. The install script runs exim in order to find its version number. If for
4220 some reason other than non-existence or emptiness, which it checks, it
4221 could not run './exim', it was installing it with an empty version number,
4222 i.e. as "exim-". This error state is now caught, and the installation is
4225 38. An argument was missing from the function that creates an error message
4226 when Exim fails to connect to the socket for saslauthd authentication.
4227 This could cause Exim to crash, or give a corrupted message.
4229 39. Added isip, isip4, and isip6 to ${if conditions.
4231 40. The ACL variables $acl_xx are now saved with the message, and can be
4232 accessed later in routers, transports, and filters.
4234 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4235 strings in the file are not string-expanded.
4237 42. If a MAIL command specified a SIZE value that was too large to fit into an
4238 int variable, the check against message_size_limit failed. Such values are
4239 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4240 day this will have to be increased, but I don't think I want to be around
4241 when emails are that large.
4248 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4249 iconv() is not standard in FreeBSD.
4251 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4252 IPv6 enabled. The observed symptom was a segmentation fault on return from
4253 the function os_common_find_running_interfaces() in src/os.c.
4255 3. In the check_special_case() function in daemon.c I had used "errno" as an
4256 argument name, which causes warnings on some systems. This was basically a
4257 typo, since it was named "eno" in the comments!
4259 4. The code that waits for the clock to tick (at a resolution of some fraction
4260 of a second) so as to ensure message-id uniqueness was always waiting for
4261 at least one whole tick, when it could have waited for less. [This is
4262 almost certainly not relevant at current processor speeds, where it is
4263 unlikely to ever wait at all. But we try to future-proof.]
4265 5. The function that sleeps for a time interval that includes fractions of a
4266 second contained a race. It did not block SIGALRM between setting the
4267 timer, and suspending (a couple of lines later). If the interval was short
4268 and the sigsuspend() was delayed until after it had expired, the suspension
4269 never ended. On busy systems this could lead to processes getting stuck for
4272 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4273 process, before it forks any delivery processes. The open lookup caching
4274 mechanism meant that the open file or database connection was passed into
4275 the delivery process. The problem was that delivery processes always tidy
4276 up cached lookup data. This could cause a problem for the next delivery
4277 process started by the queue runner, because the external queue runner
4278 process does not know about the closure. So the next delivery process
4279 still has data in the lookup cache. In the case of a file lookup, there was
4280 no problem because closing a file descriptor in a subprocess doesn't affect
4281 the parent. However, if the lookup was caching a connection to a database,
4282 the connection was closed, and the second delivery process was likely to
4283 see errors such as "PGSQL: query failed: server closed the connection
4284 unexpectedly". The problem has been fixed by closing all cached lookups
4285 in a queue runner before running a delivery process.
4287 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4288 seem to have the "const" qualifier which it has on other OS. I've
4291 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4292 *of the same type* (client or server) with the same public name that an
4293 error should be diagnosed.
4295 9. When Exim looked up a host name for an IP address, but failed to find the
4296 original IP address when looking up the host name (a safety check), it
4297 output the message "<ip address> does not match any IP for NULL", which was
4298 confusing, to say the least. The bug was that the host name should have
4299 appeared instead of "NULL".
4301 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4302 user that is built into the binary, and the -C or -D options is used, root
4303 privilege is dropped before the configuration file is read. In addition,
4304 logging is switched to stderr instead of the normal log files. If the
4305 configuration then re-defines the Exim user, the unprivileged environment
4306 is probably not what is expected, so Exim logs a panic warning message (but
4309 However, if deliver_drop_privilege is set, the unprivileged state may well
4310 be exactly what is intended, so the warning has been cut out in that case,
4311 and Exim is allowed to try to write to its normal log files.
4317 1. smtp_return_error_details was not giving details for temporary sender
4318 or receiver verification errors.
4320 2. Diagnose a configuration error if two authenticators have the same public
4323 3. Exim used not to create the message log file for a message until the first
4324 delivery attempt. This could be confusing when incoming messages were held
4325 for policy or load reasons. The message log file is now created at the time
4326 the message is received, and an initial "Received" line is written to it.
4328 4. The automatically generated man page for command line options had a minor
4329 bug that caused no ill effects; however, a more serious problem was that
4330 the procedure for building the man page automatically didn't always
4331 operate. Consequently, release 4.20 contains an out-of-date version. This
4332 shouldn't happen again.
4334 5. When building Exim with embedded Perl support, the script that builds the
4335 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4336 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4338 6. The freeze_tell option was not being used for messages that were frozen on
4339 arrival, either by an ACL or by local_scan().
4341 7. Added the smtp_incomplete_transaction log selector.
4343 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4344 was accepting AUTH without a new EHLO.
4346 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4347 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4350 10. Logging of TCP/IP connections (when configured) now happens in the main
4351 daemon process instead of the child process, so that the TCP/IP connection
4352 count is more accurate (but it can never be perfect).
4354 11. The use of "drop" in a nested ACL was not being handled correctly in the
4355 outer ACL. Now, if condition failure induced by the nested "drop" causes
4356 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4357 or "require"), the connection is dropped.
4359 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4360 that yield "discard" can now be used with an "accept" or a "discard" verb,
4361 but an error is generated for any others (because I can't see a useful way
4362 to define what should happen).
4364 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4365 are now processed in the same way as lines in the Exim configuration file.
4366 In particular, continuation lines are supported.
4368 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4370 15. Added -ti meaning -t -i.
4372 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4373 domains, and warn by logging if others are found.
4375 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4376 returned from GIFCONF when Exim is trying to find the list of interfaces on
4377 a host. The code in os.c has been modified to copy each ifreq to an aligned
4378 structure in all cases.
4380 Also, in some cases, the returned ifreq's were being copied to a 'struct
4381 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4382 means the last couple of bytes of an IPv6 address could be chopped if the
4383 ifreq contained only a normal sockaddr (14 bytes storage).
4385 18. Named domain lists were not supported in the hosts_treat_as_local option.
4386 An entry such as +xxxx was not recognized, and was treated as a literal
4389 19. Ensure that header lines added by a DATA ACL are included in the reject log
4390 if the ACL subsequently rejects the message.
4392 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4393 MD5 (which is deprecated).
4395 21. When testing a filter file using -bf, Exim was writing a message when it
4396 took the sender from a "From " line in the message, but it was not doing so
4397 when it took $return_path from a Return-Path: header line. It now does.
4399 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4400 with a valid header line field name (a series of printing characters
4401 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4403 23. Changed "disc" in the source to "disk" to conform to the documentation and
4404 the book and for uniformity.
4406 24. Ignore Sendmail's -Ooption=value command line item.
4408 25. When execve() failed while trying to run a command in a pipe transport,
4409 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4410 could be confused with a return value of 69 from the command itself. This
4411 has been changed to 127, the value the shell returns if it is asked to run
4412 a non-existent command. The wording for the related log line suggests a
4413 non-existent command as the problem.
4415 26. If received_header_text expands to an empty string, do not add a Received:
4416 header line to the message. (Well, it adds a token one on the spool, but
4417 marks it "old" so that it doesn't get used or transmitted.)
4419 27. Installed eximstats 1.28 (addition of -nt option).
4421 28. There was no check for failure on the call to getsockname() in the daemon
4422 code. This can fail if there is a shortage of resources on the system, with
4423 ENOMEM, for example. A temporary error is now given on failure.
4425 29. Contrary to the C standard, it seems that in some environments, the
4426 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4427 program. Exim now does this explicitly; it affects the formatting of
4428 timestamps using strftime().
4430 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4431 complaints. I've now initialized all the variables, to avoid this.
4433 32. Header lines added by a system filter were not being "seen" during
4434 transport-time rewrites.
4436 33. The info_callback() function passed to OpenSSL is set up with type void
4437 (*)(SSL *, int, int), as described somewhere. However, when calling the
4438 function (actually a macro) that sets it up, the type void(*)() is
4439 expected. I've put in a cast to prevent warnings from picky compilers.
4441 34. If a DNS black list lookup found a CNAME record, but there were no A
4442 records associated with the domain it pointed at, Exim crashed.
4444 35. If a DNS black list lookup returned more than one A record, Exim ignored
4445 all but the first. It now scans all returned addresses if a particular IP
4446 value is being sought. In this situation, the contents of the
4447 $dnslist_value variable are a list of all the addresses, separated by a
4450 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4451 to accept a host name and all its aliases if the forward lookup for any of
4452 them yielded the IP address of the incoming connection. Now it accepts only
4453 those names whose forward lookup yields the correct IP address. Any other
4454 names are discarded. This closes a loophole whereby a rogue DNS
4455 administrator could create reverse DNS records to break through a
4456 wildcarded host restriction in an ACL.
4458 37. If a user filter or a system filter that ran in a subprocess used any of
4459 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4460 the wrong values were passed to the pipe command ($thisaddress had the
4461 value of $0, $0 had the value of $1, etc). This bug was introduced by
4462 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4464 38. Improved the line breaking for long SMTP error messages from ACLs.
4465 Previously, if there was no break point between 40 and 75 characters, Exim
4466 left the rest of the message alone. Two changes have been made: (a) I've
4467 reduced the minimum length to 35 characters; (b) if it can't find a break
4468 point between 35 and 75 characters, it looks ahead and uses the first one
4469 that it finds. This may give the occasional overlong line, but at least the
4470 remaining text gets split now.
4472 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4473 file descriptors might be low, and that setting 1000 would always raise it.
4474 It turns out that in some environments, the limit is already over 1000 and
4475 that lowering it causes trouble. So now Exim takes care not to decrease it.
4477 40. When delivering a message, the value of $return_path is set to $sender_
4478 address at the start of routing (routers may change the value). By an
4479 oversight, this default was not being set up when an address was tested by
4480 -bt or -bv, which affected the outcome if any router or filter referred to
4483 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4484 log only when "message" or "log_message" are set. However, if one of the
4485 conditions was an address verification, or a call to a nested ACL, the
4486 messages generated by the underlying test were being passed through. This
4487 no longer happens. The underlying message is available in $acl_verify_
4488 message for both "message" and "log_message" expansions, so it can be
4489 passed through if needed.
4491 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4492 new expansion $bh_ to give the encoded byte string without charset
4493 translation. Translation happens only if iconv() is available; HAVE_ICONV
4494 indicates this at build time. HEADERS_CHARSET gives the charset to
4495 translate to; headers_charset can change it in the configuration, and
4496 "headers charset" can change it in an individual filter file.
4498 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4499 that creates RFC 2047 encoded "words" labels them as that charset instead
4500 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4501 expansion operator; (ii) when Exim creates a From: line for a local
4502 message; (iii) when a header line is rewritten to include a "phrase" part.
4504 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4505 buggy, causing it to skip the first lines of messages whose message ID
4506 ended in 'D'. This would not have bitten before Exim release 4.14, because
4507 message IDs were unlikely to end in 'D' before then. The effect was to have
4508 incorrect size information for certain domains.
4510 45. #include "config.h" was missing at the start of the crypt16.c module. This
4511 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4514 46. If there was a timeout during a "random" callout check, Exim treated it as
4515 a failure of the random address, and carried on sending RSET and the real
4516 address. If the delay was just some slowness somewhere, the response to the
4517 original RCPT would be taken as a response to RSET and so on, causing
4518 mayhem of various kinds.
4520 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4521 when I implemented it. It didn't allow for the fact that some option values
4522 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4523 the right test for positive values.
4525 48. Domain names in DNS records are case-independent. Exim always looks them up
4526 in lower case. Some resolvers return domain names in exactly the case they
4527 appear in the zone file, that is, they may contain uppercase letters. Not
4528 all resolvers do this - some return always lower case. Exim was treating a
4529 change of case by a resolver as a change of domain, similar to a widening
4530 of a domain abbreviation. This triggered its re-routing code and so it was
4531 trying to route what was effectively the same domain again. This normally
4532 caused routing to fail (because the router wouldn't handle the domain
4533 twice). Now Exim checks for this case specially, and just changes the
4534 casing of the domain that it ultimately uses when it transmits the message
4537 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4540 50. If a filter generated a file delivery with a non-absolute name (possible if
4541 no home directory exists for the router), the forbid_file option was not
4544 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4545 the existing equality matching.
4547 52. Exim was using ints instead of ino_t variables in some places where it was
4548 dealing with inode numbers.
4550 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4551 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4552 called TMPDIR, and if it finds it is different, it changes its value.
4554 54. The smtp_printf() function is now made available to local_scan() so
4555 additional output lines can be written before returning. There is also an
4556 smtp_fflush() function to enable the detection of a dropped connection.
4557 The variables smtp_input and smtp_batched_input are exported to
4560 55. Changed the default runtime configuration: the message "Unknown user"
4561 has been removed from the ACL, and instead placed on the localuser router,
4562 using the cannot_route_message feature. This means that any verification
4563 failures that generate their own messages won't get overridden. Similarly,
4564 the "Unrouteable address" message that was in the ACL for unverifiable
4565 relay addresses has also been removed.
4567 56. Added hosts_avoid_esmtp to the smtp transport.
4569 57. The exicyclog script was not checking for the esoteric option
4570 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4571 will work only if exicyclog is run under the appropriate euid.
4573 58. Following a discussion on the list, the rules by which Exim recognises line
4574 endings on incoming messages have been changed. The -dropcr and drop_cr
4575 options are now no-ops, retained only for backwards compatibility. The
4576 following line terminators are recognized: LF CRLF CR. However, special
4577 processing applies to CR:
4579 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4580 nor a local message in the state where . is a terminator.
4582 (ii) If a bare CR is encountered in a header line, an extra space is added
4583 after the line terminator so as not to end the header. The reasoning
4584 behind this is that bare CRs in header lines are most likely either
4585 to be mistakes, or people trying to play silly games.
4587 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4588 was being incorrectly given as 18 bytes larger than it should have been.
4589 This is a VOB (very old bug).
4591 60. This may never have affected anything current, but just in case it has:
4592 When the local host is found other than at the start of a list of hosts,
4593 the local host, those with the same MX, and any that follow, are discarded.
4594 When the list in question was part of a longer list of hosts, the following
4595 hosts (not currently being processed) were also being discarded. This no
4596 longer happens. I'm not sure if this situation could ever has previously
4599 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4602 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4603 header. This is something that is recommended in a new Internet Draft, and
4604 is something that is documented as being done by Sendmail. There are two
4605 possible values. For messages generated by the autoreply transport, Exim
4608 Auto-Submitted: auto-replied
4610 whereas for all other generated messages (e.g. bounces) it adds
4612 Auto-Submitted: auto-generated
4614 63. The "personal" condition in filters now includes a test for the
4615 Auto-Submitted: header. If it contains the string "auto-" the message it
4616 not considered personal.
4618 64. Added rcpt_include_affixes as a generic transport option.
4620 65. Added queue_only_override (default true).
4622 66. Added the syslog_duplication option.
4624 67. If what should have been the first header line of a message consisted of
4625 a space followed by a colon, Exim was mis-interpreting it as a header line.
4626 It isn't of course - it is syntactically invalid and should therefore be
4627 treated as the start of the message body. The misbehaviour could have
4628 caused a number of strange effects, including loss of data in subsequent
4629 header lines, and spool format errors.
4631 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4632 client host had authenticated. This control can now be exercised by an ACL
4633 for more flexibility.
4635 69. By default, callouts do not happen when testing with -bh. There is now a
4636 variant, -bhc, which does actually run the callout code, including
4637 consulting and updating the callout cache.
4639 70. Added support for saslauthd authentication, courtesy of Alexander
4642 71. If statvfs() failed on the spool or log directories while checking their
4643 size for availability, Exim confusingly gave the error "space shortage".
4644 Furthermore, in debugging mode it crashed with a floating point exception.
4645 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4646 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4647 serious problem, Exim now writes to the main and panic logs when this
4648 happens, with details of the failure. It then refuses to accept the
4649 incoming message, giving the message "spool directory problem" or "log
4650 directory problem" with a 421 code for SMTP messages.
4652 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4653 0, 1, and 2 exist, because some OS complain for execs without them (see
4654 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4655 descriptors. However, the code omitted to check that the open succeeded,
4656 causing mysterious errors if for some reason the permissions on /dev/null
4657 got screwed. Now Exim writes a message to the main and panic logs, and
4658 bombs out if it can't open /dev/null.
4660 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4661 interact so that it is all more flexible. It is supposed to remain
4662 backwards compatible. Also added extra_local_interfaces.
4664 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4665 to bomb out with an assertion failure - to the client this appears as a
4666 connection drop. This problem occurs in the part of the code that was taken
4667 from the Samba project. Fortunately, the assertion is in a very simple
4668 function, so I have fixed this by reproducing the function inline in the
4669 one place where it is called, and arranging for authentication to fail
4670 instead of killing the process with assert().
4672 75. The SPA client code was not working when the server requested OEM rather
4673 than Unicode encoding.
4675 76. Added code to make require_files with a specific uid setting more usable in
4676 the case where statting the file as root fails - usually a non-root-mounted
4677 NFS file system. When this happens and the failure is EACCES, Exim now
4678 forks a subprocess and does the per-uid checking as the relevant uid.
4680 77. Added process_log_path.
4682 78. If log_file_path was not explicitly set, a setting of check_log_space or
4683 check_log_inodes was ignored.
4685 79. If a space check for the spool or log partitions fails, the incident is now
4686 logged. Of course, in the latter case the data may get lost...
4688 80. Added the %p formatting code to string_format() so that it can be used to
4689 print addresses in debug_print(). Adjusted all the address printing in the
4690 debugging in store.c to use %p rather than %d.
4692 81. There was a concern that a line of code in smtp_in.c could overflow a
4693 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4694 initially expressed, the concern was not well-founded, because trailing
4695 spaces are removed early. However, if the trailing spaces were followed by
4696 a NULL, they did not get removed, so the overflow was possible. Two fixes
4699 (a) I re-wrote the offending code in a cleaner fashion.
4700 (b) If an incoming SMTP command contains a NULL character, it is rejected
4703 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4704 runs initgroups(), so that if the Exim user is in any additional groups,
4705 they will be used during message reception.
4711 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob