1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.23 2004/11/11 16:03:47 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
10 1. Minor wording change to the doc/README.SIEVE file.
12 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
13 computation of the current number of files was incorrect.
15 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
16 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
19 4. Give more explanation in the error message when the command for a transport
20 filter fails to execute.
22 5. There are several places where Exim runs a non-Exim command in a
23 subprocess. The SIGUSR1 signal should be disabled for these processes. This
24 was being done only for the command run by the queryprogram router. It is
25 now done for all such subprocesses. The other cases are: ${run, transport
26 filters, and the commands run by the lmtp and pipe transports.
28 6. Added CONFIGURE_GROUP build-time option.
30 7. Some older OS have a limit of 256 on the maximum number of file
31 descriptors. Exim was using setrlimit() to set 1000 as a large value
32 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
33 systems. I've change it so that if it can't get 1000, it tries for 256.
35 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
36 was an oversight, and furthermore, ever since the addition of extra
37 controls (e.g. 4.43/32), the checks on when to allow different forms of
38 "control" were broken. There should now be diagnostics for all cases when a
39 control that does not make sense is encountered.
41 9. Added the /retain_sender option to "control=submission".
43 10. $recipients is now available in the predata ACL (oversight).
45 11. Tidy the search cache before the fork to do a delivery from a message
46 received from the command line. Otherwise the child will trigger a lookup
47 failure and thereby defer the delivery if it tries to use (for example) a
48 cached ldap connection that the parent has called unbind on.
50 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
51 of $address_data from the recipient verification was clobbered by the
54 13. The value of address_data from a sender verification is now available in
55 $sender_address_data in subsequent conditions in the ACL statement.
57 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
59 15. Added a new option "connect=<time>" to callout options, to set a different
62 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
63 was its contents. (It was OK if the option was not defined at all.)
65 17. A "Completed" log line is now written for messages that are removed from
66 the spool by the -Mrm option.
68 18. New variables $sender_verify_failure and $recipient_verify_failure contain
69 information about exactly what failed.
71 19. Added -dd to debug only the daemon process.
73 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
74 handles timeouts, both on the server side and network timeouts. Renamed the
75 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
77 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
78 It is now set to "smtps".
80 22. $host_address is now set to the target address during the checking of
83 23. When checking ignore_target_hosts for an ipliteral router, no host name was
84 being passed; this would have caused $sender_host_name to have been used if
85 matching the list had actually called for a host name (not very likely,
86 since this list is usually IP addresses). A host name is now passed as
89 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
90 code that specifies a non-restarting handler (typically sigaction() in
91 modern systems) in an attempt to fix a rare and obscure crash bug.
93 25. Narrowed the window for a race in the daemon that could cause it to ignore
94 SIGCHLD signals. This is not a major problem, because they are used only to
95 wake it up if nothing else does.
101 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
102 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
103 Now it returns an integer. A number of calls were still expecting a T/F
104 return. Fortuitously, in all cases, the tests worked in OK situations,
105 which is the norm. However, things would have gone wrong on any write
106 failures on the smtp file descriptor. This function is used when sending
107 messages over SMTP and also when doing verify callouts.
109 2. When Exim is called to do synchronous delivery of a locally submitted
110 message (the -odf or -odi options), it no longer closes stderr before doing
113 3. Implemented the mua_wrapper option.
115 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
117 5. Implemented the functions header_remove(), header_testname(),
118 header_add_at_position(), and receive_remove_recipient(), and exported them
121 6. If an ACL "warn" statement specified the addition of headers, Exim already
122 inserted X-ACL-Warn: at the start if there was no header name. However, it
123 was not making this test for the second and subsequent header lines if
124 there were newlines in the string. This meant that an invalid header could
125 be inserted if Exim was badly configured.
127 7. Allow an ACL "warn" statement to add header lines at the start or after all
128 the Received: headers, as well as at the end.
130 8. Added the rcpt_4xx retry error code.
132 9. Added postmaster_mailfrom=xxx to callout verification option.
134 10. Added mailfrom=xxxx to the callout verification option, for verify=
137 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
138 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
140 12. Inserted some casts to stop certain compilers warning when using pointer
141 differences as field lengths or precisions in printf-type calls (mostly
142 affecting debugging statements).
144 13. Added optional readline() support for -be (dynamically loaded).
146 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
147 same clock tick as a message's arrival, so that its received time was the
148 same as the "first fail" time on the retry record, and that message
149 remained on the queue past the ultimate address timeout, every queue runner
150 would try a delivery (because it was past the ultimate address timeout) but
151 after another failure, the ultimate address timeout, which should have then
152 bounced the address, did not kick in. This was a "< instead of <=" error;
153 in most cases the first failure would have been in the next clock tick
154 after the received time, and all would be well.
156 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
157 being recognized when the domain list was tested by the match_domain
158 condition in an expansion string.
160 16. Added the ${str2b64: operator.
162 17. Exim was always calling setrlimit() to set a large limit for the number of
163 processes, without checking whether the existing limit was already
164 adequate. (It did check for the limit on file descriptors.) Furthermore,
165 errors from getrlimit() and setrlimit() were being ignored. Now they are
166 logged to the main and panic logs, but Exim does carry on, to try to do its
167 job under whatever limits there are.
169 18. Imported PCRE 5.0.
171 19. Trivial typo in log message " temporarily refused connection" (the leading
174 20. If the log selector return_path_on_delivery was set and an address was
175 redirected to /dev/null, the delivery process crashed because it assumed
176 that a return path would always be set for a "successful" delivery. In this
177 case, the whole delivery is bypassed as an optimization, and therefore no
180 21. Internal re-arrangement: the function for sending a challenge and reading
181 a response while authentication was assuming a zero-terminated challenge
182 string. It's now changed to take a pointer and a length, to allow for
183 binary data in such strings.
185 22. Added the cyrus_sasl authenticator (code supplied by MBM).
187 23. Exim was not respecting finduser_retries when seeking the login of the
188 uid under which it was called; it was always trying 10 times. (The default
189 setting of finduser_retries is zero.) Also, it was sleeping after the final
190 failure, which is pointless.
192 24. Implemented tls_on_connect_ports.
194 25. Implemented acl_smtp_predata.
196 26. If the domain in control=submission is set empty, Exim assumes that the
197 authenticated id is a complete email address when it generates From: or
198 Sender: header lines.
200 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
201 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
202 chown and chgrp in /bin and hostname in /usr/bin.
204 28. Exim was keeping the "process log" file open after each use, just as it
205 does for the main log. This opens the possibility of it remaining open for
206 long periods when the USR1 signal hits a daemon. Occasional processlog
207 errors were reported, that could have been caused by this. Anyway, it seems
208 much more sensible not to leave this file open at all, so that is what now
211 29. The long-running daemon process does not normally write to the log once it
212 has entered its main loop, and it closes the log before doing so. This is
213 so that log files can straightforwardly be renamed and moved. However,
214 there are a couple of unusual error situations where the daemon does write
215 log entries, and I had neglected to close the log afterwards.
217 30. The text of an SMTP error response that was received during a remote
218 delivery was being truncated at 512 bytes. This is too short for some of
219 the long messages that one sometimes sees. I've increased the limit to
222 31. It is now possible to make retry rules that apply only when a message has a
223 specific sender, in particular, an empty sender.
225 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
226 it possible to be selective about when SMTP synchronization is enforced.
228 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
230 32. Implemented hosts_connection_nolog.
232 33. Added an ACL for QUIT.
234 34. Setting "delay_warning=" to disable warnings was not working; it gave a
237 35. Added mailbox_size and mailbox_filecount to appendfile.
239 36. Added control = no_multiline_responses to ACLs.
241 37. There was a bug in the logic of the code that waits for the clock to tick
242 in the case where the clock went backwards by a substantial amount such
243 that the microsecond fraction of "now" was more than the microsecond
244 fraction of "then" (but the whole seconds number was less).
246 38. Added support for the libradius Radius client library this is found on
247 FreeBSD (previously only the radiusclient library was supported).
253 1. When certain lookups returned multiple values in the form name=value, the
254 quoting of the values was not always being done properly. Specifically:
255 (a) If the value started with a double quote, but contained no whitespace,
257 (b) If the value contained whitespace other than a space character (i.e.
258 tabs or newlines or carriage returns) it was not quoted.
259 This fix has been applied to the mysql and pgsql lookups by writing a
260 separate quoting function and calling it from the lookup code. The fix
261 should probably also be applied to nisplus, ibase and oracle lookups, but
262 since I cannot test any of those, I have not disturbed their existing code.
264 2. A hit in the callout cache for a specific address caused a log line with no
265 reason for rejecting RCPT. Now it says "Previous (cached) callout
266 verification failure".
268 3. There was an off-by-one bug in the queryprogram router. An over-long
269 return line was truncated at 256 instead of 255 characters, thereby
270 overflowing its buffer with the terminating zero. As well as fixing this, I
271 have increased the buffer size to 1024 (and made a note to document this).
273 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
274 when Exim is waiting for an SMTP response from a remote server, Exim
275 restarts its select() call on the socket, thereby resetting its timeout.
276 This is not a problem when such interrupts are rare. Somebody set up a cron
277 job to run exiwhat every 2 minutes, which is less than the normal select()
278 timeout (5 or 10 minutes). This meant that the select() timeout never
279 kicked in because it was always reset. I have fixed this by comparing the
280 time when an interrupt arrives with the time at the start of the first call
281 to select(). If more time than the timeout has elapsed, the interrupt is
282 treated as a timeout.
284 5. Some internal re-factoring in preparation for the addition of Sieve
285 extensions (by MH). In particular, the "personal" test is moved to a
286 separate function, and given an option for scanning Cc: and Bcc: (which is
287 not set for Exim filters).
289 6. When Exim created an email address using the login of the caller as the
290 local part (e.g. when creating a From: or Sender: header line), it was not
291 quoting the local part when it contained special characters such as @.
293 7. Installed new OpenBSD configuration files.
295 8. Reworded some messages for syntax errors in "and" and "or" conditions to
296 try to make them clearer.
298 9. Callout options, other than the timeout value, were being ignored when
299 verifying sender addresses in header lines. For example, when using
301 verify = header_sender/callout=no_cache
303 the cache was (incorrectly) being used.
305 10. Added a missing instance of ${EXE} to the exim_install script; this affects
306 only the Cygwin environment.
308 11. When return_path_on_delivery was set as a log selector, if different remote
309 addresses in the same message used different return paths and parallel
310 remote delivery occurred, the wrong values would sometimes be logged.
311 (Whenever a remote delivery process finished, the return path value from
312 the most recently started remote delivery process was logged.)
314 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
315 lines when AUTH and/or TLS are in use. This is the "received protocol"
316 field. Exim used to use "asmtp" for authenticated SMTP, without any
317 indication (in the protocol name) for TLS use. Now it follows the RFC and
318 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
319 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
320 names appear in log lines as well as in Received: header lines.
322 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
323 extensions, and comparison tests, and to fix some bugs.
325 14. Changes to the "personal" filter test:
327 (1) The test was buggy in that it was just doing the equivalent of
328 "contains" tests on header lines. For example, if a user's address was
329 anne@some.where, the "personal" test would incorrectly be true for
331 To: susanne@some.where
333 This test is now done by extracting each address from the header in turn,
334 and checking the entire address. Other tests that are part of "personal"
335 are now done using regular expressions (for example, to check local parts
336 of addresses in From: header lines).
338 (2) The list of non-personal local parts in From: addresses has been
339 extended to include "listserv", "majordomo", "*-request", and "owner-*",
340 taken from the Sieve specification recommendations.
342 (3) If the message contains any header line starting with "List-" it is
343 treated as non-personal.
345 (4) The test for "circular" in the Subject: header line has been removed
346 because it now seems ill-conceived.
348 15. Minor typos in src/EDITME comments corrected.
350 16. Installed latest exipick from John Jetmore.
352 17. If headers_add on a router specified a text string that was too long for
353 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
354 of string_sprintf() is now avoided.
356 18. $message_body_size was not set (it was always zero) when running the DATA
357 ACL and the local_scan() function.
359 19. For the "mail" command in an Exim filter, no default was being set for
360 the once_repeat time, causing a random time value to be used if "once" was
361 specified. (If the value happened to be <= 0, no repeat happened.) The
362 default is now 0s, meaning "never repeat". The "vacation" command was OK
363 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
364 (I found it when inspecting the code).
366 20. There is now an overall timeout for performing a callout verification. It
367 defaults to 4 times the callout timeout, which applies to individual SMTP
368 commands during the callout. The overall timeout applies when there is more
369 than one host that can be tried. The timeout is checked before trying the
370 next host. This prevents very long delays if there are a large number of
371 hosts and all are timing out (e.g. when the network connections are timing
372 out). The value of the overall timeout can be changed by specifying an
373 additional sub-option for "callout", called "maxwait". For example:
375 verify = sender/callout=5s,maxwait=20s
377 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
378 to the end before writing, but this should make it even safer).
380 22. Exim was forgetting that it had advertised PIPELINING for the second and
381 subsequent messages on an SMTP connection. It was also not resetting its
382 memory on STARTTLS and an internal HELO.
384 23. When Exim logs an SMTP synchronization error within a session, it now
385 records whether PIPELINING has been advertised or not.
387 24. Added 3 instances of "(long int)" casts to time_t variables that were being
388 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
389 rather than long int.
391 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
393 26. Added the never_mail option to autoreply.
399 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
400 crash if the getsockname() call failed; this can happen if a connection is
401 closed very soon after it is established. The problem was simply in the
402 order in which certain operations were done, causing Exim to try to write
403 to the SMTP stream before it had set up the file descriptor. The bug has
404 been fixed by making things happen in the correct order.
410 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
411 before the connection was closed, thus losing the rejection response.
413 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
414 some early Solaris releases, but causes trouble in current releases where
415 socklen_t is defined.
417 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
420 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
423 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
424 that releasing the top of it at the end releases what was used for sub-
425 expansions (unless the block got too big). However, discard this block if
426 the first thing is a variable or header, so that we can use its block when
427 it is dynamic (useful for very large $message_headers, for example).
429 6. Lookups now cache *every* query, not just the most recent. A new, separate
430 store pool is used for this. It can be recovered when all lookup caches are
431 flushed. Lookups now release memory at the end of their result strings.
432 This has involved some general refactoring of the lookup sources.
434 7. Some code has been added to the store_xxx() functions to reduce the amount
435 of flapping under certain conditions.
437 8. log_incoming_interface used to affect only the <= reception log lines. Now
438 it causes the local interface and port to be added to several more SMTP log
439 lines, for example "SMTP connection from", and rejection lines.
441 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
443 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
445 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
446 could be overwritten at the end of the current message (or the start of a
447 new message if it was set in a HELO ACL). The value is now preserved for
448 the duration of the SMTP connection.
450 12. If a transport had a headers_rewrite setting, and a matching header line
451 contained an unqualified address, that address was qualified, even if it
452 did not match any rewriting rules. The underlying bug was that the values
453 of the flags that permit the existence of unqualified sender and recipient
454 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
455 non-local messages, and by -bnq for local messages) were not being
456 preserved with the message after it was received.
458 13. When Exim was logging an SMTP synchronization error, it could sometimes log
459 "next input=" as part of the text comprising the host identity instead of
460 the correct text. The code was using the same buffer for two different
461 strings. However, depending on which order the printing function evaluated
462 its arguments, the bug did not always show up. Under Linux, for example, my
463 test suite worked just fine.
465 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
466 doesn't work with some older versions of Perl. It has been changed to "my",
467 which in any case is probably the better facility to use.
469 15. A really picky compiler found some instances of statements for creating
470 error messages that either had too many or two few arguments for the format
473 16. The size of the buffer for calls to the DNS resolver has been increased
474 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
475 for addresses that have a lot of PTR records. This alleviates a problem; it
476 does not fully solve it.
478 17. A dnsdb lookup for PTR records that receives more data than will fit in the
479 buffer now truncates the list and logs the incident, which is the same
480 action as happens when Exim is looking up a host name and its aliases.
481 Previously in this situation something unpredictable would happen;
482 sometimes it was "internal error: store_reset failed".
484 18. If a server dropped the connection unexpectedly when an Exim client was
485 using GnuTLS and trying to read a response, the client delivery process
486 crashed while trying to generate an error log message.
488 19. If a "warn" verb in an ACL added multiple headers to a message in a single
491 warn message = H1: something\nH2: something
493 the text was added as a single header line from Exim's point of view
494 though it ended up OK in the delivered message. However, searching for the
495 second and subsequent header lines using $h_h2: did not work. This has been
496 fixed. Similarly, if a system filter added multiple headers in this way,
497 the routers could not see them.
499 20. Expanded the error message when iplsearch is called with an invalid key to
500 suggest using net-iplsearch in a host list.
502 21. When running tests using -bh, any delays imposed by "delay" modifiers in
503 ACLs are no longer actually imposed (and a message to that effect is
506 22. If a "gecos" field in a passwd entry contained escaped characters, in
507 particular, if it contained a \" sequence, Exim got it wrong when building
508 a From: or a Sender: header from that name. A second bug also caused
509 incorrect handling when an unquoted " was present following a character
512 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
513 was not being matched caselessly.
515 24. Arranged for all hyphens in the exim.8 source to be escaped with
518 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
519 properly. Recipient callout cache records were still being keyed to include
520 the sender, even when use_sender was set false. This led to far more
521 callouts that were necessary. The sender is no longer included in the key
522 when use_sender is false.
524 26. Added "control = submission" modifier to ACLs.
526 27. Added the ${base62d: operator to decode base 62 numbers.
528 28. dnsdb lookups can now access SRV records.
530 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
531 the configuration file.
533 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
534 (-v) mode. This makes the output for a verbose queue run more intelligible.
536 31. Added a use_postmaster feature to recipient callouts.
538 32. Added the $body_zerocount variable, containing the number of binary zero
539 bytes in the message body.
541 33. The time of last modification of the "new" subdirectory is now used as the
542 "mailbox time last read" when there is a quota error for a maildir
545 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
547 35. Added +ignore_unknown as a special item in host lists.
549 36. Code for decoding IPv6 addresses in host lists is now included, even if
550 IPv6 support is not being compiled. This fixes a bug in which an IPv6
551 address was recognized as an IP address, but was then not correctly decoded
552 into binary, causing unexpected and incorrect effects when compared with
559 1. Very minor rewording of debugging text in manualroute to say "list of
560 hosts" instead of "hostlist".
562 2. If verify=header_syntax was set, and a header line with an unqualified
563 address (no domain) and a large number of spaces between the end of the
564 name and the colon was received, the reception process suffered a buffer
565 overflow, and (when I tested it) crashed. This was caused by some obsolete
566 code that should have been removed. The fix is to remove it!
568 3. When running in the test harness, delay a bit after writing a bounce
569 message to get a bit more predictability in the log output.
571 4. Added a call to search_tidyup() just before forking a reception process. In
572 theory, someone could use a lookup in the expansion of smtp_accept_max_
573 per_host which, without the tidyup, could leave open a database connection.
575 5. Added the variables $recipient_data and $sender_data which get set from a
576 lookup success in an ACL "recipients" or "senders" condition, or a router
577 "senders" option, similar to $domain_data and $local_part_data.
579 6. Moved the writing of debug_print from before to after the "senders" test
582 7. Change 4.31/66 (moving the time when the Received: is generated) caused
583 problems for message scanning, either using a data ACL, or using
584 local_scan() because the Received: header was not generated till after they
585 were called (in order to set the time as the time of reception completion).
586 I have revised the way this works. The header is now generated after the
587 body is received, but before the ACL or local_scan() are called. After they
588 are run, the timestamp in the header is updated.
594 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
595 before starting a queue runner without re-exec. This happened only when
596 deliver_drop_privilege was set or when the Exim user was set to root. The
597 effect of the bug was that timeouts during subsequent deliveries caused
598 crashes instead of being properly handled. The handler is now left at its
599 default (and expected) setting.
601 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
602 message, again when deliver_drop_privilege is set or Exim is run as root.
603 The bug described in (1) was not present in this case, but the tidying up
604 of the other signals was missing. I have made the two cases consistent.
606 3. The ignore_target_hosts setting on a manualroute router was being ignored
607 for hosts that were looked up using the /MX notation.
609 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
612 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
613 operated on the sender address. After changing the $sender_address to <>
614 for the sender address verify, Exim was re-instated it as the original
615 (before rewriting) address, but remembering that it had rewritten it, so it
616 wasn't rewriting it again. This bug also had the effect of breaking the
617 sender address verification caching when the sender address was rewritten.
619 6. The ignore_target_hosts option was being ignored by the ipliteral router.
620 This has been changed so that if the ip literal address matches
621 ignore_target_hosts, the router declines.
623 7. Added expansion conditions match_domain, match_address, and match_local_
624 part (NOT match_host).
626 8. The placeholder for the Received: header didn't have a length field set.
628 9. Added code to Exim itself and to exim_lock to test for a specific race
629 condition that could lead to file corruption when using MBX delivery. The
630 issue is with the lockfile that is created in /tmp. If this file is removed
631 after a process has opened it but before that process has acquired a lock,
632 there is the potential for a second process to recreate the file and also
633 acquire a lock. This could lead to two Exim processes writing to the file
634 at the same time. The added code performs the same test as UW imapd; it
635 checks after acquiring the lock that its file descriptor still refers to
638 10. The buffer for building added header lines was of fixed size, 8192 bytes.
639 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
642 11. Added the smtp_active_hostname option. If used, this will typically be made
643 to depend on the incoming interface address. Because $interface_address is
644 not set up until the daemon has forked a reception process, error responses
645 that can happen earlier (such as "too many connections") no longer contain
648 12. If an expansion in a condition on a "warn" statement fails because a lookup
649 defers, the "warn" statement is abandoned, and the next ACL statement is
650 processed. Previously this caused the whole ACL to be aborted.
652 13. Added the iplsearch lookup type.
654 14. Added ident_timeout as a log selector.
656 15. Added tls_certificate_verified as a log selector.
658 16. Added a global option tls_require_ciphers (compare the smtp transport
659 option of the same name). This controls incoming TLS connections.
661 17. I finally figured out how to make tls_require_ciphers do a similar thing
662 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
663 before starting the TLS session.
665 18. Tabs are now shown as \t in -bP output.
667 19. If the log selector return_path_on_delivery was set, Exim crashed when
668 bouncing a message because it had too many Received: header lines.
670 20. If two routers both had headers_remove settings, and the first one included
671 a superfluous trailing colon, the final name in the first list and the
672 first name in the second list were incorrectly joined into one item (with a
673 colon in the middle).
679 1. Added -C and -D options to the exinext utility, mainly to make it easier
680 to include in the automated testing, but these could be helpful when
681 multiple configurations are in use.
683 2. The exinext utility was not formatting the output nicely when there was
684 an alternate port involved in the retry record key, nor when there was a
685 message id as well (for retries that were specific to a specific message
686 and a specific host). It was also confused by IPv6 addresses, because of
687 the additional colons they contain. I have fixed the IPv4 problem, and
688 patched it up to do a reasonable job for IPv6.
690 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
691 delivery, the log line now contains "pipelined" if PIPELINING was used.
693 4. An SMTP transport process used to panic and die if the bind() call to set
694 an explicit outgoing interface failed. This has been changed; it is now
695 treated in the same way as a connect() failure.
697 5. A reference to $sender_host_name in the part of a conditional expansion
698 that was being skipped was still causing a DNS lookup. This no longer
701 6. The def: expansion condition was not recognizing references to header lines
702 that used bh_ and bheader_.
704 7. Added the _cache feature to named lists.
706 8. The code for checking quota_filecount in the appendfile transport was
707 allowing one more file than it should have been.
709 9. For compatibility with Sendmail, the command line option
717 and sets the incoming protocol and host name (for trusted callers). The
718 host name and its colon can be omitted when only the protocol is to be set.
719 Note the Exim already has two private options, -pd and -ps, that refer to
720 embedded Perl. It is therefore impossible to set a protocol value of "d" or
721 "s", but I don't think that's a major issue.
723 10. A number of refactoring changes to the code, none of which should affect
726 (a) The number of logging options was getting close to filling up the
727 32-bit word that was used as a bit map. I have split them into two classes:
728 those that are passed in the argument to log_write(), and those that are
729 only ever tested independently outside of that function. These are now in
730 separate 32-bit words, so there is plenty of room for expansion again.
731 There is no change in the user interface or the logging behaviour.
733 (b) When building, for example, log lines, the code previously used a
734 macro that called string_cat() twice, in order to add two strings. This is
735 not really sufficiently general. Furthermore, there was one instance where
736 it was actually wrong because one of the argument was used twice, and in
737 one call a function was used. (As it happened, calling the function twice
738 did not affect the overall behaviour.) The macro has been replaced by a
739 function that can join an arbitrary number of extra strings onto a growing
742 (c) The code for expansion conditions now uses a table and a binary chop
743 instead of a serial search (which was left over from when there were very
744 few conditions). Also, it now recognizes conditions like "pam" even when
745 the relevant support is not compiled in: a suitably worded error message is
746 given if an attempt is made to use such a condition.
748 11. Added ${time_interval:xxxxx}.
750 12. A bug was causing one of the ddress fields not to be passed back correctly
751 from remote delivery subprocesses. The field in question was not being
752 subsequently used, so this caused to problems in practice.
754 13. Added new log selectors queue_time and deliver_time.
756 14. Might have fixed a bug in maildirsizefile handling that threw up
757 "unexpected character" debug warnings, and recalculated the data
758 unnecessarily. In any case, I expanded the warning message to give more
761 15. Added the message "Restricted characters in address" to the statements in
762 the default ACL that block characters like @ and % in local parts.
764 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
765 Three changes have been made:
767 (a) There was a serious bug; a negative response to MAIL caused the whole
768 recipient domain to be cached as invalid, thereby blocking all messages
769 to all local parts at the same domain, from all senders. This bug has
770 been fixed. The domain is no longer cached after a negative response to
771 MAIL if the sender used is not empty.
773 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
776 (c) A new callout option, "use_sender" has been added for people who want
777 the modified behaviour.
783 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
786 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
787 indeed breaks things for older releases.
789 3. Added additional logging to the case where there is a problem reading data
790 from a filter that is running in a subprocess using a pipe, in order to
791 try to track down a specific problem.
793 4. Testing facility fudge: when running in the test harness and attempting
794 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
795 getting "No route to host". Convert this to a timeout.
797 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
800 6. Some OS don't have socklen_t but use size_t instead. This affects the
801 fifth argument of getsockopt() amongst other things. This is now
802 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
803 can be set for individual OS. I have set it for SunOS5, OSF1, and
804 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
805 some earlier ones do not.
807 7. Change 4.30/15 was not doing the test caselessly.
809 8. The standard form for an IPv6 address literal was being rejected by address
810 parsing in, for example, MAIL and RCPT commands. An example of this kind of
811 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
812 this, as well as the form without the "IPv6" on the front (but only when
813 address literals are enabled, of course).
815 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
817 10. Exim crashed if a message with an empty sender address specified by -f
818 encountered a router with an errors_to setting. This could be provoked only
823 where an empty string was supplied; "<>" did not hit this bug.
825 11. Installed PCRE release 4.5.
827 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
828 remained set. It is now erased.
830 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
831 times from message ids (which are base 36 rather than the normal 62).
833 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
834 were being counted as actual protocol errors, and logged if the log
835 selector +smtp_protocol_error was set. One cannot be perfect in this test,
836 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
837 and DATA following a set of rejected RCPTs do not count as protocol errors.
838 In other words, Exim assumes they were pipelined, though this may not
839 actually be the case. Of course, in all cases the client gets an
840 appropriate error code.
842 15. If a lookup fails in an ACL condition, a message about the failure may
843 be available; it is used if testing the ACL cannot continue, because most
844 such messages specify what the cause of the deferral is. However, some
845 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
846 that caused an old message to be retained and used if a later statement
847 caused a defer, replacing the real cause of the deferral.
849 16. If an IP address had so many PTR records that the DNS lookup buffer
850 was not large enough to hold them, Exim could crash while trying to process
851 the truncated data. It now detects and logs this case.
853 17. Further to 4.21/58, another change has been made: if (and only if) the
854 first line of a message (the first header line) ends with CRLF, a bare LF
855 in a subsequent header line has a space inserted after it, so as not to
856 terminate the header.
858 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
859 unnecessarily, used atoi() instead of strtol(), and didn't check the
860 termination when getting file sizes from file names by regex.
862 19. Completely re-implemented the support for maildirsize files, in the light
863 of a number of problems with the previous contributed implementation
864 (4.30/29). In particular:
866 . If the quota is zero, the maildirsize file is maintained, but no quota is
869 . If the maildir directory does not exist, it is created before any attempt
870 to write a maildirsize file.
872 . The quota value in the file is just a cache; if the quota is changed in
873 the transport, the new value overrides.
875 . A regular expression is available for excluding directories from the
878 20. The autoreply transport checks the characters in options that define the
879 message's headers; it allows continued headers, but it was checking with
880 isspace() after an embedded newline instead of explicitly looking for a
883 21. If all the "regular" hosts to which an address was routed had passed their
884 expiry times, and had not reached their retry times, the address was
885 bounced, even if fallback hosts were defined. Now Exim should go on to try
888 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
889 equivalent code in the SMTP transport. Some hosts send humungous responses
890 to HELO/EHLO, more than 1024 it seems.
892 23. Refactoring: code in filter.c used (void *) for "any old type" but this
893 gives compiler warnings in some environments. I've now done it "properly",
896 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
897 (because of problems with the built-in one) was declared to return uschar *
898 instead of char *, causing compiler failure.
900 25. Fixed a file descriptor leak when processing alias/forward files.
902 26. Fixed a minor format string issue in dbfn.c.
904 27. Typo in exim.c: ("dmbnz" for "dbmnz").
906 28. If a filter file refered to $h_xxx or $message_headers, and the headers
907 contained RFC 2047 "words", Exim's memory could, under certain conditions,
910 29. When a sender address is verified, it is cached, to save repeating the test
911 when there is more than one recipient in a message. However, when the
912 verification involves a callout, it is possible for different callout
913 options to be set for different recipients. It is too complicated to keep
914 track of this in the cache, so now Exim always runs a verification when a
915 callout is required, relying on the callout cache for the optimization.
916 The overhead is duplication of the address routing, but this should not be
919 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
920 to be verified with callout=postmaster, and the main callout worked but the
921 postmaster check failed, the verification correctly failed. However, if a
922 subsequent RCPT command asked for sender verification *without* the
923 postmaster check, incorrect caching caused this verification also to fail,
926 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
927 it was not caching the DNS options (qualify_single, search_parents) that
928 were used when the lookup failed. A subsequent lookup with different
929 options therefore always gave the same answer, though there were cases
930 where it should not have. (Example: a "domains = !$mx_any" option on a
931 dnslookup router: the "domains" option is always processed without any
932 widening, but the router might have qualify_single set.) Now Exim uses the
933 cached value only when the same options are set.
935 32. Added John Jetmore's "exipick" utility to the distribution.
937 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
938 than a timeout (e.g. a certificate is required, and is not provided), an
939 Exim server now closes the connection immediately. Previously it waited for
940 the client to close - but if the client is SSL, it seems that they each
941 wait for each other, leading to a delay before one of them times out.
943 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
944 maintained 0.8.x compatibility because I don't think many are using it, and
945 it is clearly obsolete.
947 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
950 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
951 client certificate was expired. A simple patch fixes this, though I don't
952 understand the full logic of why the verify callback is called multiple
955 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
956 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
957 which causes problems with some clients (such as the Certicom SSL Plus
958 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
959 disables the coutermeasure allowing Eudora to connect."
961 38. Exim was not checking that a write() to a log file succeeded. This could
962 lead to Bad Things if a log got too big, in particular if it hit a file
963 size limit. Exim now panics and dies if it cannot write to a log file, just
964 as it does if it cannot open a log file.
966 39. Modified OS/Makefile-Linux so that it now contains
968 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
970 The two -D definitions ensure that Exim is compiled with large file
971 support, which makes it possible to handle log files that are bigger than
974 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
975 instance) a domain was checked against a named list that involved a lookup,
976 causing $domain_data to be set, then another domain was checked against the
977 same list, then the first domain was re-checked, the value of $domain_data
978 after the final check could be wrong. In particular, if the second check
979 failed, it could be set empty. This bug probably also applied to
982 41. The strip_trailing_dot option was not being applied to the address given
983 with the -f command-line option.
985 42. The code for reading a message's header from the spool was incrementing
986 $received_count, but never initializing it. This meant that the value was
987 incorrect (doubled) while delivering a message in the same process in which
988 it was received. In the most common configuration of Exim, this never
989 happens - a fresh exec is done - but it can happen when
990 deliver_drop_privilege is set.
992 43. When Exim logs an SMTP synchronization error - client data sent too soon -
993 it now includes up to 150 characters of the unexpected data in the log
996 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
997 and building data strings. The size of both of these buffers was 10 000
998 bytes - far larger than anybody would *ever* want, thought I. Needless to
999 say, somebody hit the limit. I have increased the maximum line length to
1000 20 000 and the maximum data length of concatenated lines to 100 000. I have
1001 also fixed two bugs, because there was no checking on these buffers. Tsk,
1002 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1003 buffer is too small.
1005 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1006 lsearch lookups. Now it does.
1008 46. When parsing a route_list item in a manualroute router, a fixed-length
1009 buffer was used for the list of hosts. I made this 1024 bytes long,
1010 thinking that nobody would ever have a list of hosts that long. Wrong.
1011 Somebody had a whole pile of complicated expansion conditions, and the
1012 string was silently truncated, leading to an expansion error. It turns out
1013 that it is easier to change to an unlimited length (owing to other changes
1014 that have happened since this code was originally written) than to build
1015 structure for giving a limitation error. The length of the item that
1016 expands into the list of hosts is now unlimited.
1018 47. The lsearch lookup could not handle data where the length of text line was
1019 more than 4095 characters. Such lines were truncated, leading to shortened
1020 data being returned. It should now handle lines of any length.
1022 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
1023 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
1026 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
1027 replacing strings such as BIN_DIRECTORY in a source file: the replacement
1028 no longer happens in comment lines. A list of replacements is now placed
1029 at the head of all of the source files, except those whose only change is
1030 to replace PERL_COMMAND in the very first #! line.
1032 50. Replaced the slow insertion sort in queue.c, for sorting the list of
1033 messages on the queue, with a bottom-up merge sort, using code contributed
1034 by Michael Haardt. This should make operations like -bp somewhat faster on
1035 large queues. It won't affect queue runners, except when queue_run_in_order
1038 51. Installed eximstats 1.31 in the distribution.
1040 52. Added support for SRV lookups to the dnslookup router.
1042 53. If an ACL referred to $message_body or $message_body_end, the value was not
1043 reset for any messages that followed in the same SMTP session.
1045 54. The store-handling optimization for building very long strings was not
1046 differentiating between the different store pools. I don't think this
1047 actually made any difference in practice, but I've tidied it.
1049 55. While running the routers to verify a sender address, $sender_address
1050 was still set to the sender address. This is wrong, because when routing to
1051 send a bounce to the sender, it would be empty. Therefore, I have changed
1052 it so that, while verifying a sender address, $sender_address is set to <>.
1053 (There is no change to what happens when verifying a recipient address.)
1055 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
1056 target A or AAAA records (if not already returned) without resetting the
1057 qualify_single or search_parents options of the DNS resolver. These are
1058 inappropriate in this case because the targets of MX and SRV records must
1059 be FQDNs. A broken DNS record could cause trouble if it happened to have a
1060 target that, when qualified, matched something in the local domain. These
1061 two options are now turned off when doing these lookups.
1063 57. It seems that at least some releases of Reiserfs (which does not have the
1064 concept of a fixed number of inodes) returns zero and not -1 for the
1065 number of available inodes. This interacted badly with check_spool_inodes,
1066 which assumed that -1 was the "no such thing" setting. What I have done is
1067 to check that the total number of inodes is greater than zero before doing
1068 the test of how many are available.
1070 58. When a "warn" ACL statement has a log_message modifier, the message is
1071 remembered, and not repeated. This is to avoid a lot of repetition when a
1072 message has many recipients that cause the same warning to be written.
1073 Howewer, Exim was preserving the list of already written lines for an
1074 entire SMTP session, which doesn't seem right. The memory is now reset if a
1075 new message is started.
1077 59. The "rewrite" debugging flag was not showing the result of rewriting in the
1078 debugging output unless log_rewrite was also set.
1080 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
1081 of (int)(handle) when we know that handle contains (void *)(-1).
1083 61. The Exim daemon panic-logs an error return when it closes the incoming
1084 connection. However "connection reset by peer" seems to be common, and
1085 isn't really an error worthy of noting specially, so that particular error
1088 62. When Exim is trying to find all the local interfaces, it used to panic and
1089 die if the ioctl to get the interface flags failed. However, it seems that
1090 on at least one OS (Solaris 9) it is possible to have an interface that is
1091 included in the list of interfaces, but for which you get a failure error
1092 for this call. This happens when the interface is not "plumbed" into a
1093 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
1094 failure of the "get flags" call assumes that the interface is down.
1096 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
1097 makes life easier for people who are doing arithmetic on fields extracted
1098 from dates, where you often get leading zeros that should not be
1099 interpreted as octal.
1101 64. Added qualify_domain to the redirect router, to override the global
1104 65. If a pathologically long header line contained very many addresses (the
1105 report of this problem mentioned 10 000) and each of them was rewritten,
1106 Exim could use up a very large amount of memory. (It kept on making new
1107 copies of the header line as it rewrote, and never released the old ones.)
1108 At the expense of a bit more processing, the header rewriting function has
1109 been changed so that it no longer eats memory in this way.
1111 66. The generation of the Received: header has been moved from the time that a
1112 message starts to be received, to the time that it finishes. The timestamp
1113 in the Received: header should now be very close to that of the <= log
1114 line. There are two side-effects of this change:
1116 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
1117 logged header lines no longer include the local Received: line, because
1118 it has not yet been created. The same applies to a copy of the message
1119 that is returned to a non-SMTP sender when a message is rejected.
1121 (b) When a filter file is tested using -bf, no additional Received: header
1122 is added to the test message. After some thought, I decided that this
1125 This change does not affect the value of $received_for. It is still set
1126 after address rewriting, but before local_scan() is called.
1128 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
1130 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
1131 gave an unhelpful panic error message, and a defer error. I have managed to
1132 change this behaviour so that it now rejects any supplied certificate,
1133 which seems right, as the list of acceptable certificates is empty.
1135 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
1136 gave an unhelpful defer error. I have not managed to make this reject any
1137 supplied certificates, but the error message it gives is "no certificate
1138 supplied", which is not helpful.
1140 70. exigrep's output now also includes lines that are not associated with any
1141 message, but which match the given pattern. Implemented by a patch from
1142 Martin Sluka, which also tidied up the Perl a bit.
1144 71. Recipient callout verification, like sender verification, was using <> in
1145 the MAIL FROM command. This isn't really the right thing, since the actual
1146 sender may affect whether the remote host accepts the recipient or not. I
1147 have changed it to use the actual sender in the callout; this means that
1148 the cache record is now keyed on a recipient/sender pair, not just the
1149 recipient address. There doesn't seem to be a real danger of callout loops,
1150 since a callout by the remote host to check the sender would use <>.
1151 [SEE ABOVE: changed after hitting problems.]
1153 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
1154 temporary errors. However, in the case of such a code being given after
1155 the end of a data transmission (i.e. after ".") Exim was failing to write
1156 a retry record for the message. (Yes, there was some broken host that was
1157 actually sending 8xx at this point.)
1159 73. An unknown lookup type in a host list could cause Exim to panic-die when
1160 the list was checked. (An example that provoked this was putting <; in the
1161 middle of a list instead of at the start.) If this happened during a DATA
1162 ACL check, a -D file could be left lying around. This kind of configuration
1163 error no longer causes Exim to die; instead it causes a defer errror. The
1164 incident is still logged to the main and panic logs.
1166 74. Buglet left over from Exim 3 conversion. The message "too many messages
1167 in one connection" was written to the rejectlog but not the mainlog, except
1168 when address rewriting (yes!) was being logged.
1170 75. Added write_rejectlog option.
1172 76. When a system filter was run not as root (that is, when system_filter_user
1173 was set), the values of the $n variables were not being returned to the
1174 main process; thus, they were not subsequently available in the $sn
1177 77. Added +return_path_on_delivery log selector.
1179 78. A connection timeout was being treated differently from recipients deferred
1180 when testing hosts_max_try with a message that was older than the host's
1181 retry timeout. (The host should not be counted, thus allowing all hosts to
1182 be tried at least once before bouncing.) This may have been the cause of an
1183 occasionally reported bug whereby a message would remain on the queue
1184 longer than the retry timeout, but would be bounced if a delivery was
1185 forced. I say "may" because I never totally pinned down the problem;
1186 setting up timeout/retry tests is difficult. See also the next item.
1188 79. The ultimate address timeout was not being applied to errors that involved
1189 a combination of host plus message (for example, a timeout on a MAIL
1190 command). When an address resolved to a number of possible hosts, and they
1191 were not all tried for each delivery (e.g. because of hosts_max_try), a
1192 message could remain on the queue longer than the retry timeout.
1194 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
1197 81. Fixed an obscure SMTP outgoing bug which required at least the following
1198 conditions: (a) there was another message waiting for the same server;
1199 (b) the server returned 5xx to all RCPT commands in the first message so
1200 that the message was not completed; (c) the server dropped the connection
1201 or gave a negative response to the RSET that Exim sends to abort the
1202 transaction. The observed case was a dropped connection after DATA that had
1203 been sent in pipelining mode. That is, the server had advertised PIPELINING
1204 but was not implementing it correctly. The effect of the bug was incorrect
1205 behaviour, such as trying another host, and this could lead to a crash.
1211 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
1212 and daemon.c were passed as pointers to ints; they should have been
1213 pointers to socklen_t variables (which are typically unsigned ints).
1215 2. Some signed/unsigned type warnings in the os.c file for Linux have been
1218 3. Fixed a really odd bug that affected only the testing scheme; patching a
1219 certain fixed string in the binary changed the value of another string that
1220 happened to be identical to the end of the original first string.
1222 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
1223 name", it returns that address as the IP address. On some operating
1224 systems (e.g. Solaris), it also passes back the IP address string as the
1225 "host name". However, on others (e.g. Linux), it passes back an empty
1226 string. Exim wasn't checking for this, and was changing the host name to an
1227 empty string, assuming it had been canonicized.
1229 5. Although rare, it is permitted to have more than one PTR record for a given
1230 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
1231 all the names associated with an address, because they do in Solaris.
1232 However, it seems that they do not in Linux for data that comes from the
1233 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
1234 I found this out when I moved to a new Linux workstation and tried to run
1235 the Exim test suite.
1237 To get round this problem I have changed the code so that it now does its
1238 own call to the DNS to look up PTR records when searching for a host name.
1239 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
1240 addresses that are only in /etc/hosts are still found.
1242 This behaviour is, however, controlled by an option called host_lookup_
1243 order, which defaults to "bydns:byaddr". If people want to use the other
1244 order, or indeed, just use one or the other means of lookup, they can
1245 specify it in this variable.
1247 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
1248 some operating systems, this comes back from gethostbyaddr() as an empty
1249 string, and this is what Exim used to test for. However, it seems that in
1250 other systems, "." is yielded. Exim now tests for this case too.
1252 7. The values of check_spool_space and check_log_space are now held internally
1253 as a number of kilobytes instead of an absolute number of bytes. If a
1254 numbers is specified without 'K' or 'M', it is rounded up to the nearest
1255 kilobyte. This means that much larger values can be stored.
1257 8. Exim monitor: an attempt to get the action menu when not actually pointing
1258 at a message produces an empty menu entitled "No message selected". This
1259 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
1260 no entries in it ("Shell widget menu has zero width and/or height"). So I
1261 have added a single, blank menu entry in this case.
1263 9. Added ${quote_local_part.
1265 10. MIME decoding is now applied to the contents of Subject: header lines when
1268 11. Now that a reference to $sender_host_address automatically causes a reverse
1269 lookup to occur if necessary (4.13/18), there is no need to arrange for a
1270 host lookup before query-style lookups in lists that might use this
1271 variable. This has therefore been abolished, and the "net-" prefix is no
1272 longer necessary for query-style lookups.
1274 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
1275 have been a typo for LFLAGS, so it has been changed.
1277 13. The install script calls Exim with "-C /dev/null" in order to find the
1278 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
1279 to be output. Howeve, since Exim outputs its version number before the
1280 error, it didn't break the script. It just looked ugly. I fixed this by
1281 always allowing "-C /dev/null" if the caller is root.
1283 14. Ignore overlarge ACL variable number when reading spool file - insurance
1284 against a later release with more variables having written the file.
1286 15. The standard form for an IPv6 address literal was being rejected by EHLO.
1287 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1288 this, as well as the form without the "IPv6" on the front.
1290 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
1291 OS/Makefile-Darwin file.
1293 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
1294 only to LDAP libraries that do not have LDAP_OPT_DEREF.
1296 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
1297 variable. However, some OS use ints for inodes. I've added cast to long int
1298 to get rid of the compiler warning.
1300 19. I had forgotten to lock out "/../" in configuration file names when
1301 ALT_CONFIG_PREFIX was set.
1303 20. Routers used for verification do not need to specify transports. However,
1304 if such a router generated a host list, and callout was configured, Exim
1305 crashed, because it could not find a port number from the (non-existent)
1306 transport. It now assumes port 25 in this circumstance.
1308 21. Added the -t option to exigrep.
1310 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
1311 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
1312 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
1313 them is set, LOOKUP_LSEARCH is forced.
1315 23. "exim -bV" now outputs a list of lookups that are included in the binary.
1317 24. Added sender and host information to the "rejected by local_scan()" log
1318 line; previously there was no indication of these.
1320 25. Added .include_if_exists.
1322 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
1323 Linux. It turns out that not all file systems support this. Apparently some
1324 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
1325 do it.) To cope with this, the error EINVAL, which means that sync-ing is
1326 not supported on the file descriptor, is now ignored when Exim is trying to
1327 sync a directory. This applies only to Linux.
1329 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
1331 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
1332 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
1335 29. Added support for maildirsize files from supplied patch (modified a bit).
1337 30. The use of :fail: followed by an empty string could lead Exim to respond to
1338 sender verification failures with (e.g.):
1340 550 Verification failed for <xxx>
1341 550 Sender verify failed
1343 where the first response line was missing the '-' that indicates it is not
1344 the final line of the response.
1346 31. The loop for finding the name of the user that called Exim had a hardwired
1347 limit of 10; it now uses the value of finduser_retries, which is used for
1348 all other user lookups.
1350 32. Added $received_count variable, available in data and not_smtp ACLs, and at
1353 33. Exim was neglecting to zero errno before one call of strtol() when
1354 expanding a string and expecting an integer value. On some systems this
1355 resulted in spurious "integer overflow" errors. Also, it was casting the
1356 result into an int without checking.
1358 34. Testing for a connection timeout using "timeout_connect" in the retry rules
1359 did not work. The code looks as if it has *never* worked, though it appears
1360 to have been documented since at least releast 1.62. I have made it work.
1362 35. The "timeout_DNS" error in retry rules, also documented since at least
1363 1.62, also never worked. As it isn't clear exactly what this means, and
1364 clearly it isn't a major issue, I have abolished the feature by treating it
1365 as "timeout", and writing a warning to the main and panic logs.
1367 36. The display of retry rules for -brt wasn't always showing the error code
1370 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
1371 timeout_connect_A, timeout_connect_MX.
1373 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
1374 to the empty sender.
1376 39. The daemon was not analysing the content of -oX till after it had closed
1377 stderr and disconnected from the controlling terminal. This meant that any
1378 syntax errors were only noted on the panic log, and the return code from
1379 the command was 0. By re-arranging the code a little, I've made the
1380 decoding happen first, so such errors now appear on stderr, and the return
1381 code is 1. However, the actual setting up of the sockets still happens in
1382 the disconnected process, so errors there are still only recorded on the
1385 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
1386 connections (as happens on some IP stacks) was logged at start up time as
1387 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
1388 it from "IPv6 and IPv4", which means that two separate sockets are being
1391 41. The debug output for gethostbyname2() or getipnodebyname() failures now
1392 says whether AF_INET or AF_INET6 was passed as an argument.
1394 42. Exiwhat output was messed up when time zones were included in log
1397 43. Exiwhat now gives more information about the daemon's listening ports,
1398 and whether -tls-on-connect was used.
1400 44. The "port" option of the smtp transport is now expanded.
1402 45. A "message" modifier in a "warn" statement in a non-message ACL was being
1403 silently ignored. Now an error message is written to the main and panic
1406 46. There's a new ACL modifier called "logwrite" which writes to a log file
1407 as soon as it is encountered.
1409 47. Added $local_user_uid and $local_user_gid at routing time.
1411 48. Exim crashed when trying to verify a sender address that was being
1414 49. Exim was recognizing only a space character after ".include". It now also
1415 recognizes a tab character.
1417 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
1418 extracting the relevant information from the specification. The man page no
1419 longer contains scrambled data for the -d option, and I've added a section
1420 at the front about calling Exim under different names.
1422 51. Added "extra_headers" argument to the "mail" command in filter files.
1424 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
1427 53. Installed eximstats 1.29.
1429 54. Added transport_filter_timeout as a generic transport option.
1431 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
1432 Cc: header lines. This was required by RFC 822, but it not required by RFC
1435 56. Exim used to add From:, Date:, and Message-Id: header lines to any
1436 incoming messages that did not have them. Now it does so only if the
1437 message originates locally, that is, if there is no associated remote host
1438 address. When Resent- header lines are present, this applies to the Resent-
1439 lines rather than the non-Resent- lines.
1441 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
1442 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
1444 58. Messages for configuration errors now include the name of the main
1445 configuration file - useful now that there may be more than one file in a
1446 list (.included file names were always shown).
1448 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
1449 for those rare installations that do not start the daemon as root or run it
1450 setuid root. I've cut out the call to initgroups() if the daemon is not
1453 60. The Exim user and group can now be bound into the binary as text strings
1454 that are looked up at the start of Exim's processing.
1456 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
1458 62. Added $mailstore_basename variable.
1460 63. Installed patch to sieve.c from Michael Haardt.
1462 64. When Exim failed to open the panic log after failing to open the main log,
1463 the original message it was trying to log was written to stderr and debug
1464 output, but if they were not available (the usual case in production), it
1465 was lost. Now it is written to syslog before the two lines that record the
1466 failures to open the logs.
1468 65. Users' Exim filters run in subprocesses under the user's uid. It is
1469 possible for a "deliver" command or an alias in a "personal" command to
1470 provoke an address rewrite. If logging of address rewriting is configured,
1471 this fails because the process is not running as root or exim. There may be
1472 a better way of dealing with this, but for the moment (because 4.30 needs
1473 to be released), I have disabled address rewrite logging when running a
1474 filter in a non-root, non-exim process.
1480 1. The buildconfig auxiliary program wasn't quoting the value set for
1481 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
1482 not defined. This bug was masked in 4.22 by the effect that was fixed in
1485 2. Some messages that were rejected after a message id was allocated were
1486 shown as "incomplete" by exigrep. It no longer does this for messages that
1487 are rejected by local_scan() or the DATA or non-SMTP ACLs.
1489 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
1490 have allow_domain_literals set, the ID did not get logged in the <= line.
1491 Domain literals are now always recognized in Message-ID: header lines.
1493 4. The first argument for a ${extract expansion item is the key name or field
1494 number. Leading and trailing spaces in this item were not being ignored,
1495 causing some misleading effects.
1497 5. When deliver_drop_privilege was set, single queue runner processes started
1498 manually (i.e. by the command "exim -q") or by the daemon (which uses the
1499 same command in the process it spins off) were not dropping privilege.
1501 6. When the daemon running as "exim" started a queue runner, it always
1502 re-executed Exim in the spun-off process. This is a waste of effort when
1503 deliver_drop_privilege is set. The new process now just calls the
1504 queue-runner function directly.
1510 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
1513 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
1514 ignored. Though the use of -oP was forcing the writing of a pid file, it
1515 was always written to the default place.
1517 3. If the message "no IP address found for host xxxx" is generated during
1518 incoming verification, it is now followed by identification of the incoming
1519 connection (so you can more easily find what provoked it).
1521 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
1523 5. Added some features to "harden" Exim a bit more against certain attacks:
1525 (a) There is now a build-time option called FIXED_NEVER_USERS that can
1526 be put in Local/Makefile. This is like the never_users runtime option,
1527 but it cannot be overridden. The default setting is "root".
1529 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
1530 prefix string with which any file named in a -C command line option
1533 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
1534 is retained for -C and -D only if the caller of Exim is root. Without
1535 it, the exim user may also use -C and -D and retain privilege.
1537 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
1538 command line option is disabled.
1540 6. Macro names set by the -D option must start with an upper case letter, just
1541 like macro names defined in the configuration file.
1543 7. Added "dereference=" facility to LDAP.
1545 8. Two instances of the typo "uknown" in the source files are fixed.
1547 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
1548 the Configure-Makefile script screwed up while processing it.
1550 10. Incorporated PCRE 4.4.
1552 11. The SMTP synchronization check was not operating right at the start of an
1553 SMTP session. For example, it could not catch a HELO sent before the client
1554 waited for the greeting. There is now a check for outstanding input at the
1555 point when the greeting is written. Because of the duplex, asynchronous
1556 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
1557 way, but not yet received, when the check is performed.
1559 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
1560 on TCP/IP sockets, because this apparently causes some broken clients to
1563 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
1564 unchanged) from the Cygwin maintainer.
1566 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
1567 was supported instead of testing for mbx. Effectively a typo.
1569 15. The spa authenticator server code was not checking that the input it
1570 received was valid base64.
1572 16. The debug output line for the "set" modifier in ACLs was not showing the
1573 name of the variable that was being set.
1575 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
1577 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
1578 lookup on its own. Something else was needed to trigger the lookup. For
1579 example, a match in host_lookup or the need for a host name in a host list.
1580 Now, if $sender_host_name is referenced and the host name has not yet been
1581 looked up, a lookup is performed. If the lookup fails, the variable remains
1582 empty, and $host_lookup_failed is set to "1".
1584 19. Added "eqi" as a case-independent comparison operator.
1586 20. The saslauthd authentication condition could segfault if neither service
1587 nor realm was specified.
1589 21. If an overflowing value such as "2048M" was set for message_size_limit, the
1590 error message that was logged was misleading, and incoming SMTP
1591 connections were dropped. The message is now more accurate, and temporary
1592 errors are given to SMTP connections.
1594 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
1595 (except RSET) with a 421 error, until QUIT is received. However, it was
1596 failing to send a response to QUIT.
1598 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
1599 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
1600 ACL is now run after the helo_try_verify_hosts code.
1602 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
1603 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
1604 case-independent, so other case variants are also recognized). Apparently
1605 some systems use these upper case variants.
1607 25. If more than two messages were waiting for the same host, and a transport
1608 filter was specified for the transport, Exim sent two messages over the
1609 same TCP/IP connection, and then failed with "socket operation on non-
1610 socket" when it tried to send the third.
1612 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
1614 27. The extern definition of crypt16() in expand.c was not being excluded when
1615 the OS had its own crypt16() function.
1617 28. Added bounce_return_body as a new option, and bounce_return_size_limit
1618 as a preferred synonym for return_size_limit, both as an option and as an
1621 29. Added LIBS=-liconv to OS/Makefile-OSF1.
1623 30. Changed the default configuration ACL to relax the local part checking rule
1624 for addresses that are not in any local domains. For these addresses,
1625 slashes and pipe symbols are allowed within local parts, but the sequence
1626 /../ is explicitly forbidden.
1628 31. SPA server authentication was not clearing the challenge buffer before
1631 32. log_message in a "warn" ACL statement was writing to the reject log as
1632 well as to the main log, which contradicts the documentation and doesn't
1633 seem right (because no rejection is happening). So I have stopped it.
1635 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
1636 However, I am unable to do any testing of this.
1638 34. Fixed an infelicity in the appendfile transport. When checking directories
1639 for a mailbox, to see if any needed to be created, it was accidentally
1640 using path names with one or more superfluous leading slashes; tracing
1641 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
1643 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
1644 contents are added to the log line that is written for every discarded
1645 recipient. (Previously a log_message setting was ignored.)
1647 36. The ${quote: operator now quotes the string if it is empty.
1649 37. The install script runs exim in order to find its version number. If for
1650 some reason other than non-existence or emptiness, which it checks, it
1651 could not run './exim', it was installing it with an empty version number,
1652 i.e. as "exim-". This error state is now caught, and the installation is
1655 38. An argument was missing from the function that creates an error message
1656 when Exim fails to connect to the socket for saslauthd authentication.
1657 This could cause Exim to crash, or give a corrupted message.
1659 39. Added isip, isip4, and isip6 to ${if conditions.
1661 40. The ACL variables $acl_xx are now saved with the message, and can be
1662 accessed later in routers, transports, and filters.
1664 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
1665 strings in the file are not string-expanded.
1667 42. If a MAIL command specified a SIZE value that was too large to fit into an
1668 int variable, the check against message_size_limit failed. Such values are
1669 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
1670 day this will have to be increased, but I don't think I want to be around
1671 when emails are that large.
1678 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
1679 iconv() is not standard in FreeBSD.
1681 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
1682 IPv6 enabled. The observed symptom was a segmentation fault on return from
1683 the function os_common_find_running_interfaces() in src/os.c.
1685 3. In the check_special_case() function in daemon.c I had used "errno" as an
1686 argument name, which causes warnings on some systems. This was basically a
1687 typo, since it was named "eno" in the comments!
1689 4. The code that waits for the clock to tick (at a resolution of some fraction
1690 of a second) so as to ensure message-id uniqueness was always waiting for
1691 at least one whole tick, when it could have waited for less. [This is
1692 almost certainly not relevant at current processor speeds, where it is
1693 unlikely to ever wait at all. But we try to future-proof.]
1695 5. The function that sleeps for a time interval that includes fractions of a
1696 second contained a race. It did not block SIGALRM between setting the
1697 timer, and suspending (a couple of lines later). If the interval was short
1698 and the sigsuspend() was delayed until after it had expired, the suspension
1699 never ended. On busy systems this could lead to processes getting stuck for
1702 6. Some uncommon configurations may cause a lookup to happen in a queue runner
1703 process, before it forks any delivery processes. The open lookup caching
1704 mechanism meant that the open file or database connection was passed into
1705 the delivery process. The problem was that delivery processes always tidy
1706 up cached lookup data. This could cause a problem for the next delivery
1707 process started by the queue runner, because the external queue runner
1708 process does not know about the closure. So the next delivery process
1709 still has data in the lookup cache. In the case of a file lookup, there was
1710 no problem because closing a file descriptor in a subprocess doesn't affect
1711 the parent. However, if the lookup was caching a connection to a database,
1712 the connection was closed, and the second delivery process was likely to
1713 see errors such as "PGSQL: query failed: server closed the connection
1714 unexpectedly". The problem has been fixed by closing all cached lookups
1715 in a queue runner before running a delivery process.
1717 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
1718 seem to have the "const" qualifier which it has on other OS. I've
1721 8. Change 4.21/2 was too strict. It is only if there are two authenticators
1722 *of the same type* (client or server) with the same public name that an
1723 error should be diagnosed.
1725 9. When Exim looked up a host name for an IP address, but failed to find the
1726 original IP address when looking up the host name (a safety check), it
1727 output the message "<ip address> does not match any IP for NULL", which was
1728 confusing, to say the least. The bug was that the host name should have
1729 appeared instead of "NULL".
1731 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
1732 user that is built into the binary, and the -C or -D options is used, root
1733 privilege is dropped before the configuration file is read. In addition,
1734 logging is switched to stderr instead of the normal log files. If the
1735 configuration then re-defines the Exim user, the unprivileged environment
1736 is probably not what is expected, so Exim logs a panic warning message (but
1739 However, if deliver_drop_privilege is set, the unprivileged state may well
1740 be exactly what is intended, so the warning has been cut out in that case,
1741 and Exim is allowed to try to write to its normal log files.
1747 1. smtp_return_error_details was not giving details for temporary sender
1748 or receiver verification errors.
1750 2. Diagnose a configuration error if two authenticators have the same public
1753 3. Exim used not to create the message log file for a message until the first
1754 delivery attempt. This could be confusing when incoming messages were held
1755 for policy or load reasons. The message log file is now created at the time
1756 the message is received, and an initial "Received" line is written to it.
1758 4. The automatically generated man page for command line options had a minor
1759 bug that caused no ill effects; however, a more serious problem was that
1760 the procedure for building the man page automatically didn't always
1761 operate. Consequently, release 4.20 contains an out-of-date version. This
1762 shouldn't happen again.
1764 5. When building Exim with embedded Perl support, the script that builds the
1765 Makefile was calling 'perl' to find its compile-time parameters, ignoring
1766 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
1768 6. The freeze_tell option was not being used for messages that were frozen on
1769 arrival, either by an ACL or by local_scan().
1771 7. Added the smtp_incomplete_transaction log selector.
1773 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
1774 was accepting AUTH without a new EHLO.
1776 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
1777 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
1780 10. Logging of TCP/IP connections (when configured) now happens in the main
1781 daemon process instead of the child process, so that the TCP/IP connection
1782 count is more accurate (but it can never be perfect).
1784 11. The use of "drop" in a nested ACL was not being handled correctly in the
1785 outer ACL. Now, if condition failure induced by the nested "drop" causes
1786 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
1787 or "require"), the connection is dropped.
1789 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
1790 that yield "discard" can now be used with an "accept" or a "discard" verb,
1791 but an error is generated for any others (because I can't see a useful way
1792 to define what should happen).
1794 13. When an ACL is read dynamically from a file (or anywhere else), the lines
1795 are now processed in the same way as lines in the Exim configuration file.
1796 In particular, continuation lines are supported.
1798 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
1800 15. Added -ti meaning -t -i.
1802 16. Check for letters, digits, hyphens, and dots in the names of dnslist
1803 domains, and warn by logging if others are found.
1805 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
1806 returned from GIFCONF when Exim is trying to find the list of interfaces on
1807 a host. The code in os.c has been modified to copy each ifreq to an aligned
1808 structure in all cases.
1810 Also, in some cases, the returned ifreq's were being copied to a 'struct
1811 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
1812 means the last couple of bytes of an IPv6 address could be chopped if the
1813 ifreq contained only a normal sockaddr (14 bytes storage).
1815 18. Named domain lists were not supported in the hosts_treat_as_local option.
1816 An entry such as +xxxx was not recognized, and was treated as a literal
1819 19. Ensure that header lines added by a DATA ACL are included in the reject log
1820 if the ACL subsequently rejects the message.
1822 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
1823 MD5 (which is deprecated).
1825 21. When testing a filter file using -bf, Exim was writing a message when it
1826 took the sender from a "From " line in the message, but it was not doing so
1827 when it took $return_path from a Return-Path: header line. It now does.
1829 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
1830 with a valid header line field name (a series of printing characters
1831 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
1833 23. Changed "disc" in the source to "disk" to conform to the documentation and
1834 the book and for uniformity.
1836 24. Ignore Sendmail's -Ooption=value command line item.
1838 25. When execve() failed while trying to run a command in a pipe transport,
1839 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
1840 could be confused with a return value of 69 from the command itself. This
1841 has been changed to 127, the value the shell returns if it is asked to run
1842 a non-existent command. The wording for the related log line suggests a
1843 non-existent command as the problem.
1845 26. If received_header_text expands to an empty string, do not add a Received:
1846 header line to the message. (Well, it adds a token one on the spool, but
1847 marks it "old" so that it doesn't get used or transmitted.)
1849 27. Installed eximstats 1.28 (addition of -nt option).
1851 28. There was no check for failure on the call to getsockname() in the daemon
1852 code. This can fail if there is a shortage of resources on the system, with
1853 ENOMEM, for example. A temporary error is now given on failure.
1855 29. Contrary to the C standard, it seems that in some environments, the
1856 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
1857 program. Exim now does this explicitly; it affects the formatting of
1858 timestamps using strftime().
1860 30. If exiqsumm was given junk data, it threw up some uninitialized variable
1861 complaints. I've now initialized all the variables, to avoid this.
1863 32. Header lines added by a system filter were not being "seen" during
1864 transport-time rewrites.
1866 33. The info_callback() function passed to OpenSSL is set up with type void
1867 (*)(SSL *, int, int), as described somewhere. However, when calling the
1868 function (actually a macro) that sets it up, the type void(*)() is
1869 expected. I've put in a cast to prevent warnings from picky compilers.
1871 34. If a DNS black list lookup found a CNAME record, but there were no A
1872 records associated with the domain it pointed at, Exim crashed.
1874 35. If a DNS black list lookup returned more than one A record, Exim ignored
1875 all but the first. It now scans all returned addresses if a particular IP
1876 value is being sought. In this situation, the contents of the
1877 $dnslist_value variable are a list of all the addresses, separated by a
1880 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
1881 to accept a host name and all its aliases if the forward lookup for any of
1882 them yielded the IP address of the incoming connection. Now it accepts only
1883 those names whose forward lookup yields the correct IP address. Any other
1884 names are discarded. This closes a loophole whereby a rogue DNS
1885 administrator could create reverse DNS records to break through a
1886 wildcarded host restriction in an ACL.
1888 37. If a user filter or a system filter that ran in a subprocess used any of
1889 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
1890 the wrong values were passed to the pipe command ($thisaddress had the
1891 value of $0, $0 had the value of $1, etc). This bug was introduced by
1892 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
1894 38. Improved the line breaking for long SMTP error messages from ACLs.
1895 Previously, if there was no break point between 40 and 75 characters, Exim
1896 left the rest of the message alone. Two changes have been made: (a) I've
1897 reduced the minimum length to 35 characters; (b) if it can't find a break
1898 point between 35 and 75 characters, it looks ahead and uses the first one
1899 that it finds. This may give the occasional overlong line, but at least the
1900 remaining text gets split now.
1902 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
1903 file descriptors might be low, and that setting 1000 would always raise it.
1904 It turns out that in some environments, the limit is already over 1000 and
1905 that lowering it causes trouble. So now Exim takes care not to decrease it.
1907 40. When delivering a message, the value of $return_path is set to $sender_
1908 address at the start of routing (routers may change the value). By an
1909 oversight, this default was not being set up when an address was tested by
1910 -bt or -bv, which affected the outcome if any router or filter referred to
1913 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
1914 log only when "message" or "log_message" are set. However, if one of the
1915 conditions was an address verification, or a call to a nested ACL, the
1916 messages generated by the underlying test were being passed through. This
1917 no longer happens. The underlying message is available in $acl_verify_
1918 message for both "message" and "log_message" expansions, so it can be
1919 passed through if needed.
1921 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
1922 new expansion $bh_ to give the encoded byte string without charset
1923 translation. Translation happens only if iconv() is available; HAVE_ICONV
1924 indicates this at build time. HEADERS_CHARSET gives the charset to
1925 translate to; headers_charset can change it in the configuration, and
1926 "headers charset" can change it in an individual filter file.
1928 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
1929 that creates RFC 2047 encoded "words" labels them as that charset instead
1930 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
1931 expansion operator; (ii) when Exim creates a From: line for a local
1932 message; (iii) when a header line is rewritten to include a "phrase" part.
1934 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
1935 buggy, causing it to skip the first lines of messages whose message ID
1936 ended in 'D'. This would not have bitten before Exim release 4.14, because
1937 message IDs were unlikely to end in 'D' before then. The effect was to have
1938 incorrect size information for certain domains.
1940 45. #include "config.h" was missing at the start of the crypt16.c module. This
1941 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
1944 46. If there was a timeout during a "random" callout check, Exim treated it as
1945 a failure of the random address, and carried on sending RSET and the real
1946 address. If the delay was just some slowness somewhere, the response to the
1947 original RCPT would be taken as a response to RSET and so on, causing
1948 mayhem of various kinds.
1950 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
1951 when I implemented it. It didn't allow for the fact that some option values
1952 may legitimatetly be negative (e.g. size_addition), and it didn't even do
1953 the right test for positive values.
1955 48. Domain names in DNS records are case-independent. Exim always looks them up
1956 in lower case. Some resolvers return domain names in exactly the case they
1957 appear in the zone file, that is, they may contain uppercase letters. Not
1958 all resolvers do this - some return always lower case. Exim was treating a
1959 change of case by a resolver as a change of domain, similar to a widening
1960 of a domain abbreviation. This triggered its re-routing code and so it was
1961 trying to route what was effectively the same domain again. This normally
1962 caused routing to fail (because the router wouldn't handle the domain
1963 twice). Now Exim checks for this case specially, and just changes the
1964 casing of the domain that it ultimately uses when it transmits the message
1967 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
1970 50. If a filter generated a file delivery with a non-absolute name (possible if
1971 no home directory exists for the router), the forbid_file option was not
1974 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
1975 the existing equality matching.
1977 52. Exim was using ints instead of ino_t variables in some places where it was
1978 dealing with inode numbers.
1980 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
1981 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
1982 called TMPDIR, and if it finds it is different, it changes its value.
1984 54. The smtp_printf() function is now made available to local_scan() so
1985 additional output lines can be written before returning. There is also an
1986 smtp_fflush() function to enable the detection of a dropped connection.
1987 The variables smtp_input and smtp_batched_input are exported to
1990 55. Changed the default runtime configuration: the message "Unknown user"
1991 has been removed from the ACL, and instead placed on the localuser router,
1992 using the cannot_route_message feature. This means that any verification
1993 failures that generate their own messages won't get overridden. Similarly,
1994 the "Unrouteable address" message that was in the ACL for unverifiable
1995 relay addresses has also been removed.
1997 56. Added hosts_avoid_esmtp to the smtp transport.
1999 57. The exicyclog script was not checking for the esoteric option
2000 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2001 will work only if exicyclog is run under the appropriate euid.
2003 58. Following a discussion on the list, the rules by which Exim recognises line
2004 endings on incoming messages have been changed. The -dropcr and drop_cr
2005 options are now no-ops, retained only for backwards compatibility. The
2006 following line terminators are recognized: LF CRLF CR. However, special
2007 processing applies to CR:
2009 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2010 nor a local message in the state where . is a terminator.
2012 (ii) If a bare CR is encountered in a header line, an extra space is added
2013 after the line terminator so as not to end the header. The reasoning
2014 behind this is that bare CRs in header lines are most likely either
2015 to be mistakes, or people trying to play silly games.
2017 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
2018 was being incorrectly given as 18 bytes larger than it should have been.
2019 This is a VOB (very old bug).
2021 60. This may never have affected anything current, but just in case it has:
2022 When the local host is found other than at the start of a list of hosts,
2023 the local host, those with the same MX, and any that follow, are discarded.
2024 When the list in question was part of a longer list of hosts, the following
2025 hosts (not currently being processed) were also being discarded. This no
2026 longer happens. I'm not sure if this situation could ever has previously
2029 61. Added the "/MX" feature to lists of hosts in the manualroute and query
2032 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
2033 header. This is something that is recommended in a new Internet Draft, and
2034 is something that is documented as being done by Sendmail. There are two
2035 possible values. For messages generated by the autoreply transport, Exim
2038 Auto-Submitted: auto-replied
2040 whereas for all other generated messages (e.g. bounces) it adds
2042 Auto-Submitted: auto-generated
2044 63. The "personal" condition in filters now includes a test for the
2045 Auto-Submitted: header. If it contains the string "auto-" the message it
2046 not considered personal.
2048 64. Added rcpt_include_affixes as a generic transport option.
2050 65. Added queue_only_override (default true).
2052 66. Added the syslog_duplication option.
2054 67. If what should have been the first header line of a message consisted of
2055 a space followed by a colon, Exim was mis-interpreting it as a header line.
2056 It isn't of course - it is syntactically invalid and should therefore be
2057 treated as the start of the message body. The misbehaviour could have
2058 caused a number of strange effects, including loss of data in subsequent
2059 header lines, and spool format errors.
2061 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
2062 client host had authenticated. This control can now be exercised by an ACL
2063 for more flexibility.
2065 69. By default, callouts do not happen when testing with -bh. There is now a
2066 variant, -bhc, which does actually run the callout code, including
2067 consulting and updating the callout cache.
2069 70. Added support for saslauthd authentication, courtesy of Alexander
2072 71. If statvfs() failed on the spool or log directories while checking their
2073 size for availability, Exim confusingly gave the error "space shortage".
2074 Furthermore, in debugging mode it crashed with a floating point exception.
2075 These checks are done if check_{spool,log}_{space,inodes} are set, and when
2076 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
2077 serious problem, Exim now writes to the main and panic logs when this
2078 happens, with details of the failure. It then refuses to accept the
2079 incoming message, giving the message "spool directory problem" or "log
2080 directory problem" with a 421 code for SMTP messages.
2082 72. When Exim is about to re-exec itself, it ensures that the file descriptors
2083 0, 1, and 2 exist, because some OS complain for execs without them (see
2084 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
2085 descriptors. However, the code omitted to check that the open succeeded,
2086 causing mysterious errors if for some reason the permissions on /dev/null
2087 got screwed. Now Exim writes a message to the main and panic logs, and
2088 bombs out if it can't open /dev/null.
2090 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
2091 interact so that it is all more flexible. It is supposed to remain
2092 backwards compatible. Also added extra_local_interfaces.
2094 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
2095 to bomb out with an assertion failure - to the client this appears as a
2096 connection drop. This problem occurs in the part of the code that was taken
2097 from the Samba project. Fortunately, the assertion is in a very simple
2098 function, so I have fixed this by reproducing the function inline in the
2099 one place where it is called, and arranging for authentication to fail
2100 instead of killing the process with assert().
2102 75. The SPA client code was not working when the server requested OEM rather
2103 than Unicode encoding.
2105 76. Added code to make require_files with a specific uid setting more usable in
2106 the case where statting the file as root fails - usually a non-root-mounted
2107 NFS file system. When this happens and the failure is EACCES, Exim now
2108 forks a subprocess and does the per-uid checking as the relevant uid.
2110 77. Added process_log_path.
2112 78. If log_file_path was not explicitly set, a setting of check_log_space or
2113 check_log_inodes was ignored.
2115 79. If a space check for the spool or log partitions fails, the incident is now
2116 logged. Of course, in the latter case the data may get lost...
2118 80. Added the %p formatting code to string_format() so that it can be used to
2119 print addresses in debug_print(). Adjusted all the address printing in the
2120 debugging in store.c to use %p rather than %d.
2122 81. There was a concern that a line of code in smtp_in.c could overflow a
2123 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
2124 initially expressed, the concern was not well-founded, because trailing
2125 spaces are removed early. However, if the trailing spaces were followed by
2126 a NULL, they did not get removed, so the overflow was possible. Two fixes
2129 (a) I re-wrote the offending code in a cleaner fashion.
2130 (b) If an incoming SMTP command contains a NULL character, it is rejected
2133 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
2134 runs initgroups(), so that if the Exim user is in any additional groups,
2135 they will be used during message reception.
2141 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob