1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.15 2004/11/04 12:19:48 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
66 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
67 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
68 Now it returns an integer. A number of calls were still expecting a T/F
69 return. Fortuitously, in all cases, the tests worked in OK situations,
70 which is the norm. However, things would have gone wrong on any write
71 failures on the smtp file descriptor. This function is used when sending
72 messages over SMTP and also when doing verify callouts.
74 2. When Exim is called to do synchronous delivery of a locally submitted
75 message (the -odf or -odi options), it no longer closes stderr before doing
78 3. Implemented the mua_wrapper option.
80 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
82 5. Implemented the functions header_remove(), header_testname(),
83 header_add_at_position(), and receive_remove_recipient(), and exported them
86 6. If an ACL "warn" statement specified the addition of headers, Exim already
87 inserted X-ACL-Warn: at the start if there was no header name. However, it
88 was not making this test for the second and subsequent header lines if
89 there were newlines in the string. This meant that an invalid header could
90 be inserted if Exim was badly configured.
92 7. Allow an ACL "warn" statement to add header lines at the start or after all
93 the Received: headers, as well as at the end.
95 8. Added the rcpt_4xx retry error code.
97 9. Added postmaster_mailfrom=xxx to callout verification option.
99 10. Added mailfrom=xxxx to the callout verification option, for verify=
102 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
103 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
105 12. Inserted some casts to stop certain compilers warning when using pointer
106 differences as field lengths or precisions in printf-type calls (mostly
107 affecting debugging statements).
109 13. Added optional readline() support for -be (dynamically loaded).
111 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
112 same clock tick as a message's arrival, so that its received time was the
113 same as the "first fail" time on the retry record, and that message
114 remained on the queue past the ultimate address timeout, every queue runner
115 would try a delivery (because it was past the ultimate address timeout) but
116 after another failure, the ultimate address timeout, which should have then
117 bounced the address, did not kick in. This was a "< instead of <=" error;
118 in most cases the first failure would have been in the next clock tick
119 after the received time, and all would be well.
121 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
122 being recognized when the domain list was tested by the match_domain
123 condition in an expansion string.
125 16. Added the ${str2b64: operator.
127 17. Exim was always calling setrlimit() to set a large limit for the number of
128 processes, without checking whether the existing limit was already
129 adequate. (It did check for the limit on file descriptors.) Furthermore,
130 errors from getrlimit() and setrlimit() were being ignored. Now they are
131 logged to the main and panic logs, but Exim does carry on, to try to do its
132 job under whatever limits there are.
134 18. Imported PCRE 5.0.
136 19. Trivial typo in log message " temporarily refused connection" (the leading
139 20. If the log selector return_path_on_delivery was set and an address was
140 redirected to /dev/null, the delivery process crashed because it assumed
141 that a return path would always be set for a "successful" delivery. In this
142 case, the whole delivery is bypassed as an optimization, and therefore no
145 21. Internal re-arrangement: the function for sending a challenge and reading
146 a response while authentication was assuming a zero-terminated challenge
147 string. It's now changed to take a pointer and a length, to allow for
148 binary data in such strings.
150 22. Added the cyrus_sasl authenticator (code supplied by MBM).
152 23. Exim was not respecting finduser_retries when seeking the login of the
153 uid under which it was called; it was always trying 10 times. (The default
154 setting of finduser_retries is zero.) Also, it was sleeping after the final
155 failure, which is pointless.
157 24. Implemented tls_on_connect_ports.
159 25. Implemented acl_smtp_predata.
161 26. If the domain in control=submission is set empty, Exim assumes that the
162 authenticated id is a complete email address when it generates From: or
163 Sender: header lines.
165 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
166 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
167 chown and chgrp in /bin and hostname in /usr/bin.
169 28. Exim was keeping the "process log" file open after each use, just as it
170 does for the main log. This opens the possibility of it remaining open for
171 long periods when the USR1 signal hits a daemon. Occasional processlog
172 errors were reported, that could have been caused by this. Anyway, it seems
173 much more sensible not to leave this file open at all, so that is what now
176 29. The long-running daemon process does not normally write to the log once it
177 has entered its main loop, and it closes the log before doing so. This is
178 so that log files can straightforwardly be renamed and moved. However,
179 there are a couple of unusual error situations where the daemon does write
180 log entries, and I had neglected to close the log afterwards.
182 30. The text of an SMTP error response that was received during a remote
183 delivery was being truncated at 512 bytes. This is too short for some of
184 the long messages that one sometimes sees. I've increased the limit to
187 31. It is now possible to make retry rules that apply only when a message has a
188 specific sender, in particular, an empty sender.
190 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
191 it possible to be selective about when SMTP synchronization is enforced.
193 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
195 32. Implemented hosts_connection_nolog.
197 33. Added an ACL for QUIT.
199 34. Setting "delay_warning=" to disable warnings was not working; it gave a
202 35. Added mailbox_size and mailbox_filecount to appendfile.
204 36. Added control = no_multiline_responses to ACLs.
206 37. There was a bug in the logic of the code that waits for the clock to tick
207 in the case where the clock went backwards by a substantial amount such
208 that the microsecond fraction of "now" was more than the microsecond
209 fraction of "then" (but the whole seconds number was less).
211 38. Added support for the libradius Radius client library this is found on
212 FreeBSD (previously only the radiusclient library was supported).
218 1. When certain lookups returned multiple values in the form name=value, the
219 quoting of the values was not always being done properly. Specifically:
220 (a) If the value started with a double quote, but contained no whitespace,
222 (b) If the value contained whitespace other than a space character (i.e.
223 tabs or newlines or carriage returns) it was not quoted.
224 This fix has been applied to the mysql and pgsql lookups by writing a
225 separate quoting function and calling it from the lookup code. The fix
226 should probably also be applied to nisplus, ibase and oracle lookups, but
227 since I cannot test any of those, I have not disturbed their existing code.
229 2. A hit in the callout cache for a specific address caused a log line with no
230 reason for rejecting RCPT. Now it says "Previous (cached) callout
231 verification failure".
233 3. There was an off-by-one bug in the queryprogram router. An over-long
234 return line was truncated at 256 instead of 255 characters, thereby
235 overflowing its buffer with the terminating zero. As well as fixing this, I
236 have increased the buffer size to 1024 (and made a note to document this).
238 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
239 when Exim is waiting for an SMTP response from a remote server, Exim
240 restarts its select() call on the socket, thereby resetting its timeout.
241 This is not a problem when such interrupts are rare. Somebody set up a cron
242 job to run exiwhat every 2 minutes, which is less than the normal select()
243 timeout (5 or 10 minutes). This meant that the select() timeout never
244 kicked in because it was always reset. I have fixed this by comparing the
245 time when an interrupt arrives with the time at the start of the first call
246 to select(). If more time than the timeout has elapsed, the interrupt is
247 treated as a timeout.
249 5. Some internal re-factoring in preparation for the addition of Sieve
250 extensions (by MH). In particular, the "personal" test is moved to a
251 separate function, and given an option for scanning Cc: and Bcc: (which is
252 not set for Exim filters).
254 6. When Exim created an email address using the login of the caller as the
255 local part (e.g. when creating a From: or Sender: header line), it was not
256 quoting the local part when it contained special characters such as @.
258 7. Installed new OpenBSD configuration files.
260 8. Reworded some messages for syntax errors in "and" and "or" conditions to
261 try to make them clearer.
263 9. Callout options, other than the timeout value, were being ignored when
264 verifying sender addresses in header lines. For example, when using
266 verify = header_sender/callout=no_cache
268 the cache was (incorrectly) being used.
270 10. Added a missing instance of ${EXE} to the exim_install script; this affects
271 only the Cygwin environment.
273 11. When return_path_on_delivery was set as a log selector, if different remote
274 addresses in the same message used different return paths and parallel
275 remote delivery occurred, the wrong values would sometimes be logged.
276 (Whenever a remote delivery process finished, the return path value from
277 the most recently started remote delivery process was logged.)
279 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
280 lines when AUTH and/or TLS are in use. This is the "received protocol"
281 field. Exim used to use "asmtp" for authenticated SMTP, without any
282 indication (in the protocol name) for TLS use. Now it follows the RFC and
283 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
284 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
285 names appear in log lines as well as in Received: header lines.
287 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
288 extensions, and comparison tests, and to fix some bugs.
290 14. Changes to the "personal" filter test:
292 (1) The test was buggy in that it was just doing the equivalent of
293 "contains" tests on header lines. For example, if a user's address was
294 anne@some.where, the "personal" test would incorrectly be true for
296 To: susanne@some.where
298 This test is now done by extracting each address from the header in turn,
299 and checking the entire address. Other tests that are part of "personal"
300 are now done using regular expressions (for example, to check local parts
301 of addresses in From: header lines).
303 (2) The list of non-personal local parts in From: addresses has been
304 extended to include "listserv", "majordomo", "*-request", and "owner-*",
305 taken from the Sieve specification recommendations.
307 (3) If the message contains any header line starting with "List-" it is
308 treated as non-personal.
310 (4) The test for "circular" in the Subject: header line has been removed
311 because it now seems ill-conceived.
313 15. Minor typos in src/EDITME comments corrected.
315 16. Installed latest exipick from John Jetmore.
317 17. If headers_add on a router specified a text string that was too long for
318 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
319 of string_sprintf() is now avoided.
321 18. $message_body_size was not set (it was always zero) when running the DATA
322 ACL and the local_scan() function.
324 19. For the "mail" command in an Exim filter, no default was being set for
325 the once_repeat time, causing a random time value to be used if "once" was
326 specified. (If the value happened to be <= 0, no repeat happened.) The
327 default is now 0s, meaning "never repeat". The "vacation" command was OK
328 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
329 (I found it when inspecting the code).
331 20. There is now an overall timeout for performing a callout verification. It
332 defaults to 4 times the callout timeout, which applies to individual SMTP
333 commands during the callout. The overall timeout applies when there is more
334 than one host that can be tried. The timeout is checked before trying the
335 next host. This prevents very long delays if there are a large number of
336 hosts and all are timing out (e.g. when the network connections are timing
337 out). The value of the overall timeout can be changed by specifying an
338 additional sub-option for "callout", called "maxwait". For example:
340 verify = sender/callout=5s,maxwait=20s
342 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
343 to the end before writing, but this should make it even safer).
345 22. Exim was forgetting that it had advertised PIPELINING for the second and
346 subsequent messages on an SMTP connection. It was also not resetting its
347 memory on STARTTLS and an internal HELO.
349 23. When Exim logs an SMTP synchronization error within a session, it now
350 records whether PIPELINING has been advertised or not.
352 24. Added 3 instances of "(long int)" casts to time_t variables that were being
353 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
354 rather than long int.
356 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
358 26. Added the never_mail option to autoreply.
364 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
365 crash if the getsockname() call failed; this can happen if a connection is
366 closed very soon after it is established. The problem was simply in the
367 order in which certain operations were done, causing Exim to try to write
368 to the SMTP stream before it had set up the file descriptor. The bug has
369 been fixed by making things happen in the correct order.
375 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
376 before the connection was closed, thus losing the rejection response.
378 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
379 some early Solaris releases, but causes trouble in current releases where
380 socklen_t is defined.
382 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
385 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
388 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
389 that releasing the top of it at the end releases what was used for sub-
390 expansions (unless the block got too big). However, discard this block if
391 the first thing is a variable or header, so that we can use its block when
392 it is dynamic (useful for very large $message_headers, for example).
394 6. Lookups now cache *every* query, not just the most recent. A new, separate
395 store pool is used for this. It can be recovered when all lookup caches are
396 flushed. Lookups now release memory at the end of their result strings.
397 This has involved some general refactoring of the lookup sources.
399 7. Some code has been added to the store_xxx() functions to reduce the amount
400 of flapping under certain conditions.
402 8. log_incoming_interface used to affect only the <= reception log lines. Now
403 it causes the local interface and port to be added to several more SMTP log
404 lines, for example "SMTP connection from", and rejection lines.
406 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
408 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
410 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
411 could be overwritten at the end of the current message (or the start of a
412 new message if it was set in a HELO ACL). The value is now preserved for
413 the duration of the SMTP connection.
415 12. If a transport had a headers_rewrite setting, and a matching header line
416 contained an unqualified address, that address was qualified, even if it
417 did not match any rewriting rules. The underlying bug was that the values
418 of the flags that permit the existence of unqualified sender and recipient
419 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
420 non-local messages, and by -bnq for local messages) were not being
421 preserved with the message after it was received.
423 13. When Exim was logging an SMTP synchronization error, it could sometimes log
424 "next input=" as part of the text comprising the host identity instead of
425 the correct text. The code was using the same buffer for two different
426 strings. However, depending on which order the printing function evaluated
427 its arguments, the bug did not always show up. Under Linux, for example, my
428 test suite worked just fine.
430 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
431 doesn't work with some older versions of Perl. It has been changed to "my",
432 which in any case is probably the better facility to use.
434 15. A really picky compiler found some instances of statements for creating
435 error messages that either had too many or two few arguments for the format
438 16. The size of the buffer for calls to the DNS resolver has been increased
439 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
440 for addresses that have a lot of PTR records. This alleviates a problem; it
441 does not fully solve it.
443 17. A dnsdb lookup for PTR records that receives more data than will fit in the
444 buffer now truncates the list and logs the incident, which is the same
445 action as happens when Exim is looking up a host name and its aliases.
446 Previously in this situation something unpredictable would happen;
447 sometimes it was "internal error: store_reset failed".
449 18. If a server dropped the connection unexpectedly when an Exim client was
450 using GnuTLS and trying to read a response, the client delivery process
451 crashed while trying to generate an error log message.
453 19. If a "warn" verb in an ACL added multiple headers to a message in a single
456 warn message = H1: something\nH2: something
458 the text was added as a single header line from Exim's point of view
459 though it ended up OK in the delivered message. However, searching for the
460 second and subsequent header lines using $h_h2: did not work. This has been
461 fixed. Similarly, if a system filter added multiple headers in this way,
462 the routers could not see them.
464 20. Expanded the error message when iplsearch is called with an invalid key to
465 suggest using net-iplsearch in a host list.
467 21. When running tests using -bh, any delays imposed by "delay" modifiers in
468 ACLs are no longer actually imposed (and a message to that effect is
471 22. If a "gecos" field in a passwd entry contained escaped characters, in
472 particular, if it contained a \" sequence, Exim got it wrong when building
473 a From: or a Sender: header from that name. A second bug also caused
474 incorrect handling when an unquoted " was present following a character
477 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
478 was not being matched caselessly.
480 24. Arranged for all hyphens in the exim.8 source to be escaped with
483 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
484 properly. Recipient callout cache records were still being keyed to include
485 the sender, even when use_sender was set false. This led to far more
486 callouts that were necessary. The sender is no longer included in the key
487 when use_sender is false.
489 26. Added "control = submission" modifier to ACLs.
491 27. Added the ${base62d: operator to decode base 62 numbers.
493 28. dnsdb lookups can now access SRV records.
495 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
496 the configuration file.
498 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
499 (-v) mode. This makes the output for a verbose queue run more intelligible.
501 31. Added a use_postmaster feature to recipient callouts.
503 32. Added the $body_zerocount variable, containing the number of binary zero
504 bytes in the message body.
506 33. The time of last modification of the "new" subdirectory is now used as the
507 "mailbox time last read" when there is a quota error for a maildir
510 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
512 35. Added +ignore_unknown as a special item in host lists.
514 36. Code for decoding IPv6 addresses in host lists is now included, even if
515 IPv6 support is not being compiled. This fixes a bug in which an IPv6
516 address was recognized as an IP address, but was then not correctly decoded
517 into binary, causing unexpected and incorrect effects when compared with
524 1. Very minor rewording of debugging text in manualroute to say "list of
525 hosts" instead of "hostlist".
527 2. If verify=header_syntax was set, and a header line with an unqualified
528 address (no domain) and a large number of spaces between the end of the
529 name and the colon was received, the reception process suffered a buffer
530 overflow, and (when I tested it) crashed. This was caused by some obsolete
531 code that should have been removed. The fix is to remove it!
533 3. When running in the test harness, delay a bit after writing a bounce
534 message to get a bit more predictability in the log output.
536 4. Added a call to search_tidyup() just before forking a reception process. In
537 theory, someone could use a lookup in the expansion of smtp_accept_max_
538 per_host which, without the tidyup, could leave open a database connection.
540 5. Added the variables $recipient_data and $sender_data which get set from a
541 lookup success in an ACL "recipients" or "senders" condition, or a router
542 "senders" option, similar to $domain_data and $local_part_data.
544 6. Moved the writing of debug_print from before to after the "senders" test
547 7. Change 4.31/66 (moving the time when the Received: is generated) caused
548 problems for message scanning, either using a data ACL, or using
549 local_scan() because the Received: header was not generated till after they
550 were called (in order to set the time as the time of reception completion).
551 I have revised the way this works. The header is now generated after the
552 body is received, but before the ACL or local_scan() are called. After they
553 are run, the timestamp in the header is updated.
559 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
560 before starting a queue runner without re-exec. This happened only when
561 deliver_drop_privilege was set or when the Exim user was set to root. The
562 effect of the bug was that timeouts during subsequent deliveries caused
563 crashes instead of being properly handled. The handler is now left at its
564 default (and expected) setting.
566 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
567 message, again when deliver_drop_privilege is set or Exim is run as root.
568 The bug described in (1) was not present in this case, but the tidying up
569 of the other signals was missing. I have made the two cases consistent.
571 3. The ignore_target_hosts setting on a manualroute router was being ignored
572 for hosts that were looked up using the /MX notation.
574 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
577 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
578 operated on the sender address. After changing the $sender_address to <>
579 for the sender address verify, Exim was re-instated it as the original
580 (before rewriting) address, but remembering that it had rewritten it, so it
581 wasn't rewriting it again. This bug also had the effect of breaking the
582 sender address verification caching when the sender address was rewritten.
584 6. The ignore_target_hosts option was being ignored by the ipliteral router.
585 This has been changed so that if the ip literal address matches
586 ignore_target_hosts, the router declines.
588 7. Added expansion conditions match_domain, match_address, and match_local_
589 part (NOT match_host).
591 8. The placeholder for the Received: header didn't have a length field set.
593 9. Added code to Exim itself and to exim_lock to test for a specific race
594 condition that could lead to file corruption when using MBX delivery. The
595 issue is with the lockfile that is created in /tmp. If this file is removed
596 after a process has opened it but before that process has acquired a lock,
597 there is the potential for a second process to recreate the file and also
598 acquire a lock. This could lead to two Exim processes writing to the file
599 at the same time. The added code performs the same test as UW imapd; it
600 checks after acquiring the lock that its file descriptor still refers to
603 10. The buffer for building added header lines was of fixed size, 8192 bytes.
604 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
607 11. Added the smtp_active_hostname option. If used, this will typically be made
608 to depend on the incoming interface address. Because $interface_address is
609 not set up until the daemon has forked a reception process, error responses
610 that can happen earlier (such as "too many connections") no longer contain
613 12. If an expansion in a condition on a "warn" statement fails because a lookup
614 defers, the "warn" statement is abandoned, and the next ACL statement is
615 processed. Previously this caused the whole ACL to be aborted.
617 13. Added the iplsearch lookup type.
619 14. Added ident_timeout as a log selector.
621 15. Added tls_certificate_verified as a log selector.
623 16. Added a global option tls_require_ciphers (compare the smtp transport
624 option of the same name). This controls incoming TLS connections.
626 17. I finally figured out how to make tls_require_ciphers do a similar thing
627 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
628 before starting the TLS session.
630 18. Tabs are now shown as \t in -bP output.
632 19. If the log selector return_path_on_delivery was set, Exim crashed when
633 bouncing a message because it had too many Received: header lines.
635 20. If two routers both had headers_remove settings, and the first one included
636 a superfluous trailing colon, the final name in the first list and the
637 first name in the second list were incorrectly joined into one item (with a
638 colon in the middle).
644 1. Added -C and -D options to the exinext utility, mainly to make it easier
645 to include in the automated testing, but these could be helpful when
646 multiple configurations are in use.
648 2. The exinext utility was not formatting the output nicely when there was
649 an alternate port involved in the retry record key, nor when there was a
650 message id as well (for retries that were specific to a specific message
651 and a specific host). It was also confused by IPv6 addresses, because of
652 the additional colons they contain. I have fixed the IPv4 problem, and
653 patched it up to do a reasonable job for IPv6.
655 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
656 delivery, the log line now contains "pipelined" if PIPELINING was used.
658 4. An SMTP transport process used to panic and die if the bind() call to set
659 an explicit outgoing interface failed. This has been changed; it is now
660 treated in the same way as a connect() failure.
662 5. A reference to $sender_host_name in the part of a conditional expansion
663 that was being skipped was still causing a DNS lookup. This no longer
666 6. The def: expansion condition was not recognizing references to header lines
667 that used bh_ and bheader_.
669 7. Added the _cache feature to named lists.
671 8. The code for checking quota_filecount in the appendfile transport was
672 allowing one more file than it should have been.
674 9. For compatibility with Sendmail, the command line option
682 and sets the incoming protocol and host name (for trusted callers). The
683 host name and its colon can be omitted when only the protocol is to be set.
684 Note the Exim already has two private options, -pd and -ps, that refer to
685 embedded Perl. It is therefore impossible to set a protocol value of "d" or
686 "s", but I don't think that's a major issue.
688 10. A number of refactoring changes to the code, none of which should affect
691 (a) The number of logging options was getting close to filling up the
692 32-bit word that was used as a bit map. I have split them into two classes:
693 those that are passed in the argument to log_write(), and those that are
694 only ever tested independently outside of that function. These are now in
695 separate 32-bit words, so there is plenty of room for expansion again.
696 There is no change in the user interface or the logging behaviour.
698 (b) When building, for example, log lines, the code previously used a
699 macro that called string_cat() twice, in order to add two strings. This is
700 not really sufficiently general. Furthermore, there was one instance where
701 it was actually wrong because one of the argument was used twice, and in
702 one call a function was used. (As it happened, calling the function twice
703 did not affect the overall behaviour.) The macro has been replaced by a
704 function that can join an arbitrary number of extra strings onto a growing
707 (c) The code for expansion conditions now uses a table and a binary chop
708 instead of a serial search (which was left over from when there were very
709 few conditions). Also, it now recognizes conditions like "pam" even when
710 the relevant support is not compiled in: a suitably worded error message is
711 given if an attempt is made to use such a condition.
713 11. Added ${time_interval:xxxxx}.
715 12. A bug was causing one of the ddress fields not to be passed back correctly
716 from remote delivery subprocesses. The field in question was not being
717 subsequently used, so this caused to problems in practice.
719 13. Added new log selectors queue_time and deliver_time.
721 14. Might have fixed a bug in maildirsizefile handling that threw up
722 "unexpected character" debug warnings, and recalculated the data
723 unnecessarily. In any case, I expanded the warning message to give more
726 15. Added the message "Restricted characters in address" to the statements in
727 the default ACL that block characters like @ and % in local parts.
729 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
730 Three changes have been made:
732 (a) There was a serious bug; a negative response to MAIL caused the whole
733 recipient domain to be cached as invalid, thereby blocking all messages
734 to all local parts at the same domain, from all senders. This bug has
735 been fixed. The domain is no longer cached after a negative response to
736 MAIL if the sender used is not empty.
738 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
741 (c) A new callout option, "use_sender" has been added for people who want
742 the modified behaviour.
748 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
751 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
752 indeed breaks things for older releases.
754 3. Added additional logging to the case where there is a problem reading data
755 from a filter that is running in a subprocess using a pipe, in order to
756 try to track down a specific problem.
758 4. Testing facility fudge: when running in the test harness and attempting
759 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
760 getting "No route to host". Convert this to a timeout.
762 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
765 6. Some OS don't have socklen_t but use size_t instead. This affects the
766 fifth argument of getsockopt() amongst other things. This is now
767 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
768 can be set for individual OS. I have set it for SunOS5, OSF1, and
769 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
770 some earlier ones do not.
772 7. Change 4.30/15 was not doing the test caselessly.
774 8. The standard form for an IPv6 address literal was being rejected by address
775 parsing in, for example, MAIL and RCPT commands. An example of this kind of
776 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
777 this, as well as the form without the "IPv6" on the front (but only when
778 address literals are enabled, of course).
780 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
782 10. Exim crashed if a message with an empty sender address specified by -f
783 encountered a router with an errors_to setting. This could be provoked only
788 where an empty string was supplied; "<>" did not hit this bug.
790 11. Installed PCRE release 4.5.
792 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
793 remained set. It is now erased.
795 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
796 times from message ids (which are base 36 rather than the normal 62).
798 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
799 were being counted as actual protocol errors, and logged if the log
800 selector +smtp_protocol_error was set. One cannot be perfect in this test,
801 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
802 and DATA following a set of rejected RCPTs do not count as protocol errors.
803 In other words, Exim assumes they were pipelined, though this may not
804 actually be the case. Of course, in all cases the client gets an
805 appropriate error code.
807 15. If a lookup fails in an ACL condition, a message about the failure may
808 be available; it is used if testing the ACL cannot continue, because most
809 such messages specify what the cause of the deferral is. However, some
810 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
811 that caused an old message to be retained and used if a later statement
812 caused a defer, replacing the real cause of the deferral.
814 16. If an IP address had so many PTR records that the DNS lookup buffer
815 was not large enough to hold them, Exim could crash while trying to process
816 the truncated data. It now detects and logs this case.
818 17. Further to 4.21/58, another change has been made: if (and only if) the
819 first line of a message (the first header line) ends with CRLF, a bare LF
820 in a subsequent header line has a space inserted after it, so as not to
821 terminate the header.
823 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
824 unnecessarily, used atoi() instead of strtol(), and didn't check the
825 termination when getting file sizes from file names by regex.
827 19. Completely re-implemented the support for maildirsize files, in the light
828 of a number of problems with the previous contributed implementation
829 (4.30/29). In particular:
831 . If the quota is zero, the maildirsize file is maintained, but no quota is
834 . If the maildir directory does not exist, it is created before any attempt
835 to write a maildirsize file.
837 . The quota value in the file is just a cache; if the quota is changed in
838 the transport, the new value overrides.
840 . A regular expression is available for excluding directories from the
843 20. The autoreply transport checks the characters in options that define the
844 message's headers; it allows continued headers, but it was checking with
845 isspace() after an embedded newline instead of explicitly looking for a
848 21. If all the "regular" hosts to which an address was routed had passed their
849 expiry times, and had not reached their retry times, the address was
850 bounced, even if fallback hosts were defined. Now Exim should go on to try
853 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
854 equivalent code in the SMTP transport. Some hosts send humungous responses
855 to HELO/EHLO, more than 1024 it seems.
857 23. Refactoring: code in filter.c used (void *) for "any old type" but this
858 gives compiler warnings in some environments. I've now done it "properly",
861 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
862 (because of problems with the built-in one) was declared to return uschar *
863 instead of char *, causing compiler failure.
865 25. Fixed a file descriptor leak when processing alias/forward files.
867 26. Fixed a minor format string issue in dbfn.c.
869 27. Typo in exim.c: ("dmbnz" for "dbmnz").
871 28. If a filter file refered to $h_xxx or $message_headers, and the headers
872 contained RFC 2047 "words", Exim's memory could, under certain conditions,
875 29. When a sender address is verified, it is cached, to save repeating the test
876 when there is more than one recipient in a message. However, when the
877 verification involves a callout, it is possible for different callout
878 options to be set for different recipients. It is too complicated to keep
879 track of this in the cache, so now Exim always runs a verification when a
880 callout is required, relying on the callout cache for the optimization.
881 The overhead is duplication of the address routing, but this should not be
884 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
885 to be verified with callout=postmaster, and the main callout worked but the
886 postmaster check failed, the verification correctly failed. However, if a
887 subsequent RCPT command asked for sender verification *without* the
888 postmaster check, incorrect caching caused this verification also to fail,
891 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
892 it was not caching the DNS options (qualify_single, search_parents) that
893 were used when the lookup failed. A subsequent lookup with different
894 options therefore always gave the same answer, though there were cases
895 where it should not have. (Example: a "domains = !$mx_any" option on a
896 dnslookup router: the "domains" option is always processed without any
897 widening, but the router might have qualify_single set.) Now Exim uses the
898 cached value only when the same options are set.
900 32. Added John Jetmore's "exipick" utility to the distribution.
902 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
903 than a timeout (e.g. a certificate is required, and is not provided), an
904 Exim server now closes the connection immediately. Previously it waited for
905 the client to close - but if the client is SSL, it seems that they each
906 wait for each other, leading to a delay before one of them times out.
908 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
909 maintained 0.8.x compatibility because I don't think many are using it, and
910 it is clearly obsolete.
912 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
915 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
916 client certificate was expired. A simple patch fixes this, though I don't
917 understand the full logic of why the verify callback is called multiple
920 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
921 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
922 which causes problems with some clients (such as the Certicom SSL Plus
923 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
924 disables the coutermeasure allowing Eudora to connect."
926 38. Exim was not checking that a write() to a log file succeeded. This could
927 lead to Bad Things if a log got too big, in particular if it hit a file
928 size limit. Exim now panics and dies if it cannot write to a log file, just
929 as it does if it cannot open a log file.
931 39. Modified OS/Makefile-Linux so that it now contains
933 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
935 The two -D definitions ensure that Exim is compiled with large file
936 support, which makes it possible to handle log files that are bigger than
939 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
940 instance) a domain was checked against a named list that involved a lookup,
941 causing $domain_data to be set, then another domain was checked against the
942 same list, then the first domain was re-checked, the value of $domain_data
943 after the final check could be wrong. In particular, if the second check
944 failed, it could be set empty. This bug probably also applied to
947 41. The strip_trailing_dot option was not being applied to the address given
948 with the -f command-line option.
950 42. The code for reading a message's header from the spool was incrementing
951 $received_count, but never initializing it. This meant that the value was
952 incorrect (doubled) while delivering a message in the same process in which
953 it was received. In the most common configuration of Exim, this never
954 happens - a fresh exec is done - but it can happen when
955 deliver_drop_privilege is set.
957 43. When Exim logs an SMTP synchronization error - client data sent too soon -
958 it now includes up to 150 characters of the unexpected data in the log
961 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
962 and building data strings. The size of both of these buffers was 10 000
963 bytes - far larger than anybody would *ever* want, thought I. Needless to
964 say, somebody hit the limit. I have increased the maximum line length to
965 20 000 and the maximum data length of concatenated lines to 100 000. I have
966 also fixed two bugs, because there was no checking on these buffers. Tsk,
967 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
970 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
971 lsearch lookups. Now it does.
973 46. When parsing a route_list item in a manualroute router, a fixed-length
974 buffer was used for the list of hosts. I made this 1024 bytes long,
975 thinking that nobody would ever have a list of hosts that long. Wrong.
976 Somebody had a whole pile of complicated expansion conditions, and the
977 string was silently truncated, leading to an expansion error. It turns out
978 that it is easier to change to an unlimited length (owing to other changes
979 that have happened since this code was originally written) than to build
980 structure for giving a limitation error. The length of the item that
981 expands into the list of hosts is now unlimited.
983 47. The lsearch lookup could not handle data where the length of text line was
984 more than 4095 characters. Such lines were truncated, leading to shortened
985 data being returned. It should now handle lines of any length.
987 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
988 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
991 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
992 replacing strings such as BIN_DIRECTORY in a source file: the replacement
993 no longer happens in comment lines. A list of replacements is now placed
994 at the head of all of the source files, except those whose only change is
995 to replace PERL_COMMAND in the very first #! line.
997 50. Replaced the slow insertion sort in queue.c, for sorting the list of
998 messages on the queue, with a bottom-up merge sort, using code contributed
999 by Michael Haardt. This should make operations like -bp somewhat faster on
1000 large queues. It won't affect queue runners, except when queue_run_in_order
1003 51. Installed eximstats 1.31 in the distribution.
1005 52. Added support for SRV lookups to the dnslookup router.
1007 53. If an ACL referred to $message_body or $message_body_end, the value was not
1008 reset for any messages that followed in the same SMTP session.
1010 54. The store-handling optimization for building very long strings was not
1011 differentiating between the different store pools. I don't think this
1012 actually made any difference in practice, but I've tidied it.
1014 55. While running the routers to verify a sender address, $sender_address
1015 was still set to the sender address. This is wrong, because when routing to
1016 send a bounce to the sender, it would be empty. Therefore, I have changed
1017 it so that, while verifying a sender address, $sender_address is set to <>.
1018 (There is no change to what happens when verifying a recipient address.)
1020 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
1021 target A or AAAA records (if not already returned) without resetting the
1022 qualify_single or search_parents options of the DNS resolver. These are
1023 inappropriate in this case because the targets of MX and SRV records must
1024 be FQDNs. A broken DNS record could cause trouble if it happened to have a
1025 target that, when qualified, matched something in the local domain. These
1026 two options are now turned off when doing these lookups.
1028 57. It seems that at least some releases of Reiserfs (which does not have the
1029 concept of a fixed number of inodes) returns zero and not -1 for the
1030 number of available inodes. This interacted badly with check_spool_inodes,
1031 which assumed that -1 was the "no such thing" setting. What I have done is
1032 to check that the total number of inodes is greater than zero before doing
1033 the test of how many are available.
1035 58. When a "warn" ACL statement has a log_message modifier, the message is
1036 remembered, and not repeated. This is to avoid a lot of repetition when a
1037 message has many recipients that cause the same warning to be written.
1038 Howewer, Exim was preserving the list of already written lines for an
1039 entire SMTP session, which doesn't seem right. The memory is now reset if a
1040 new message is started.
1042 59. The "rewrite" debugging flag was not showing the result of rewriting in the
1043 debugging output unless log_rewrite was also set.
1045 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
1046 of (int)(handle) when we know that handle contains (void *)(-1).
1048 61. The Exim daemon panic-logs an error return when it closes the incoming
1049 connection. However "connection reset by peer" seems to be common, and
1050 isn't really an error worthy of noting specially, so that particular error
1053 62. When Exim is trying to find all the local interfaces, it used to panic and
1054 die if the ioctl to get the interface flags failed. However, it seems that
1055 on at least one OS (Solaris 9) it is possible to have an interface that is
1056 included in the list of interfaces, but for which you get a failure error
1057 for this call. This happens when the interface is not "plumbed" into a
1058 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
1059 failure of the "get flags" call assumes that the interface is down.
1061 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
1062 makes life easier for people who are doing arithmetic on fields extracted
1063 from dates, where you often get leading zeros that should not be
1064 interpreted as octal.
1066 64. Added qualify_domain to the redirect router, to override the global
1069 65. If a pathologically long header line contained very many addresses (the
1070 report of this problem mentioned 10 000) and each of them was rewritten,
1071 Exim could use up a very large amount of memory. (It kept on making new
1072 copies of the header line as it rewrote, and never released the old ones.)
1073 At the expense of a bit more processing, the header rewriting function has
1074 been changed so that it no longer eats memory in this way.
1076 66. The generation of the Received: header has been moved from the time that a
1077 message starts to be received, to the time that it finishes. The timestamp
1078 in the Received: header should now be very close to that of the <= log
1079 line. There are two side-effects of this change:
1081 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
1082 logged header lines no longer include the local Received: line, because
1083 it has not yet been created. The same applies to a copy of the message
1084 that is returned to a non-SMTP sender when a message is rejected.
1086 (b) When a filter file is tested using -bf, no additional Received: header
1087 is added to the test message. After some thought, I decided that this
1090 This change does not affect the value of $received_for. It is still set
1091 after address rewriting, but before local_scan() is called.
1093 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
1095 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
1096 gave an unhelpful panic error message, and a defer error. I have managed to
1097 change this behaviour so that it now rejects any supplied certificate,
1098 which seems right, as the list of acceptable certificates is empty.
1100 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
1101 gave an unhelpful defer error. I have not managed to make this reject any
1102 supplied certificates, but the error message it gives is "no certificate
1103 supplied", which is not helpful.
1105 70. exigrep's output now also includes lines that are not associated with any
1106 message, but which match the given pattern. Implemented by a patch from
1107 Martin Sluka, which also tidied up the Perl a bit.
1109 71. Recipient callout verification, like sender verification, was using <> in
1110 the MAIL FROM command. This isn't really the right thing, since the actual
1111 sender may affect whether the remote host accepts the recipient or not. I
1112 have changed it to use the actual sender in the callout; this means that
1113 the cache record is now keyed on a recipient/sender pair, not just the
1114 recipient address. There doesn't seem to be a real danger of callout loops,
1115 since a callout by the remote host to check the sender would use <>.
1116 [SEE ABOVE: changed after hitting problems.]
1118 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
1119 temporary errors. However, in the case of such a code being given after
1120 the end of a data transmission (i.e. after ".") Exim was failing to write
1121 a retry record for the message. (Yes, there was some broken host that was
1122 actually sending 8xx at this point.)
1124 73. An unknown lookup type in a host list could cause Exim to panic-die when
1125 the list was checked. (An example that provoked this was putting <; in the
1126 middle of a list instead of at the start.) If this happened during a DATA
1127 ACL check, a -D file could be left lying around. This kind of configuration
1128 error no longer causes Exim to die; instead it causes a defer errror. The
1129 incident is still logged to the main and panic logs.
1131 74. Buglet left over from Exim 3 conversion. The message "too many messages
1132 in one connection" was written to the rejectlog but not the mainlog, except
1133 when address rewriting (yes!) was being logged.
1135 75. Added write_rejectlog option.
1137 76. When a system filter was run not as root (that is, when system_filter_user
1138 was set), the values of the $n variables were not being returned to the
1139 main process; thus, they were not subsequently available in the $sn
1142 77. Added +return_path_on_delivery log selector.
1144 78. A connection timeout was being treated differently from recipients deferred
1145 when testing hosts_max_try with a message that was older than the host's
1146 retry timeout. (The host should not be counted, thus allowing all hosts to
1147 be tried at least once before bouncing.) This may have been the cause of an
1148 occasionally reported bug whereby a message would remain on the queue
1149 longer than the retry timeout, but would be bounced if a delivery was
1150 forced. I say "may" because I never totally pinned down the problem;
1151 setting up timeout/retry tests is difficult. See also the next item.
1153 79. The ultimate address timeout was not being applied to errors that involved
1154 a combination of host plus message (for example, a timeout on a MAIL
1155 command). When an address resolved to a number of possible hosts, and they
1156 were not all tried for each delivery (e.g. because of hosts_max_try), a
1157 message could remain on the queue longer than the retry timeout.
1159 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
1162 81. Fixed an obscure SMTP outgoing bug which required at least the following
1163 conditions: (a) there was another message waiting for the same server;
1164 (b) the server returned 5xx to all RCPT commands in the first message so
1165 that the message was not completed; (c) the server dropped the connection
1166 or gave a negative response to the RSET that Exim sends to abort the
1167 transaction. The observed case was a dropped connection after DATA that had
1168 been sent in pipelining mode. That is, the server had advertised PIPELINING
1169 but was not implementing it correctly. The effect of the bug was incorrect
1170 behaviour, such as trying another host, and this could lead to a crash.
1176 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
1177 and daemon.c were passed as pointers to ints; they should have been
1178 pointers to socklen_t variables (which are typically unsigned ints).
1180 2. Some signed/unsigned type warnings in the os.c file for Linux have been
1183 3. Fixed a really odd bug that affected only the testing scheme; patching a
1184 certain fixed string in the binary changed the value of another string that
1185 happened to be identical to the end of the original first string.
1187 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
1188 name", it returns that address as the IP address. On some operating
1189 systems (e.g. Solaris), it also passes back the IP address string as the
1190 "host name". However, on others (e.g. Linux), it passes back an empty
1191 string. Exim wasn't checking for this, and was changing the host name to an
1192 empty string, assuming it had been canonicized.
1194 5. Although rare, it is permitted to have more than one PTR record for a given
1195 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
1196 all the names associated with an address, because they do in Solaris.
1197 However, it seems that they do not in Linux for data that comes from the
1198 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
1199 I found this out when I moved to a new Linux workstation and tried to run
1200 the Exim test suite.
1202 To get round this problem I have changed the code so that it now does its
1203 own call to the DNS to look up PTR records when searching for a host name.
1204 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
1205 addresses that are only in /etc/hosts are still found.
1207 This behaviour is, however, controlled by an option called host_lookup_
1208 order, which defaults to "bydns:byaddr". If people want to use the other
1209 order, or indeed, just use one or the other means of lookup, they can
1210 specify it in this variable.
1212 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
1213 some operating systems, this comes back from gethostbyaddr() as an empty
1214 string, and this is what Exim used to test for. However, it seems that in
1215 other systems, "." is yielded. Exim now tests for this case too.
1217 7. The values of check_spool_space and check_log_space are now held internally
1218 as a number of kilobytes instead of an absolute number of bytes. If a
1219 numbers is specified without 'K' or 'M', it is rounded up to the nearest
1220 kilobyte. This means that much larger values can be stored.
1222 8. Exim monitor: an attempt to get the action menu when not actually pointing
1223 at a message produces an empty menu entitled "No message selected". This
1224 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
1225 no entries in it ("Shell widget menu has zero width and/or height"). So I
1226 have added a single, blank menu entry in this case.
1228 9. Added ${quote_local_part.
1230 10. MIME decoding is now applied to the contents of Subject: header lines when
1233 11. Now that a reference to $sender_host_address automatically causes a reverse
1234 lookup to occur if necessary (4.13/18), there is no need to arrange for a
1235 host lookup before query-style lookups in lists that might use this
1236 variable. This has therefore been abolished, and the "net-" prefix is no
1237 longer necessary for query-style lookups.
1239 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
1240 have been a typo for LFLAGS, so it has been changed.
1242 13. The install script calls Exim with "-C /dev/null" in order to find the
1243 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
1244 to be output. Howeve, since Exim outputs its version number before the
1245 error, it didn't break the script. It just looked ugly. I fixed this by
1246 always allowing "-C /dev/null" if the caller is root.
1248 14. Ignore overlarge ACL variable number when reading spool file - insurance
1249 against a later release with more variables having written the file.
1251 15. The standard form for an IPv6 address literal was being rejected by EHLO.
1252 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1253 this, as well as the form without the "IPv6" on the front.
1255 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
1256 OS/Makefile-Darwin file.
1258 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
1259 only to LDAP libraries that do not have LDAP_OPT_DEREF.
1261 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
1262 variable. However, some OS use ints for inodes. I've added cast to long int
1263 to get rid of the compiler warning.
1265 19. I had forgotten to lock out "/../" in configuration file names when
1266 ALT_CONFIG_PREFIX was set.
1268 20. Routers used for verification do not need to specify transports. However,
1269 if such a router generated a host list, and callout was configured, Exim
1270 crashed, because it could not find a port number from the (non-existent)
1271 transport. It now assumes port 25 in this circumstance.
1273 21. Added the -t option to exigrep.
1275 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
1276 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
1277 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
1278 them is set, LOOKUP_LSEARCH is forced.
1280 23. "exim -bV" now outputs a list of lookups that are included in the binary.
1282 24. Added sender and host information to the "rejected by local_scan()" log
1283 line; previously there was no indication of these.
1285 25. Added .include_if_exists.
1287 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
1288 Linux. It turns out that not all file systems support this. Apparently some
1289 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
1290 do it.) To cope with this, the error EINVAL, which means that sync-ing is
1291 not supported on the file descriptor, is now ignored when Exim is trying to
1292 sync a directory. This applies only to Linux.
1294 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
1296 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
1297 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
1300 29. Added support for maildirsize files from supplied patch (modified a bit).
1302 30. The use of :fail: followed by an empty string could lead Exim to respond to
1303 sender verification failures with (e.g.):
1305 550 Verification failed for <xxx>
1306 550 Sender verify failed
1308 where the first response line was missing the '-' that indicates it is not
1309 the final line of the response.
1311 31. The loop for finding the name of the user that called Exim had a hardwired
1312 limit of 10; it now uses the value of finduser_retries, which is used for
1313 all other user lookups.
1315 32. Added $received_count variable, available in data and not_smtp ACLs, and at
1318 33. Exim was neglecting to zero errno before one call of strtol() when
1319 expanding a string and expecting an integer value. On some systems this
1320 resulted in spurious "integer overflow" errors. Also, it was casting the
1321 result into an int without checking.
1323 34. Testing for a connection timeout using "timeout_connect" in the retry rules
1324 did not work. The code looks as if it has *never* worked, though it appears
1325 to have been documented since at least releast 1.62. I have made it work.
1327 35. The "timeout_DNS" error in retry rules, also documented since at least
1328 1.62, also never worked. As it isn't clear exactly what this means, and
1329 clearly it isn't a major issue, I have abolished the feature by treating it
1330 as "timeout", and writing a warning to the main and panic logs.
1332 36. The display of retry rules for -brt wasn't always showing the error code
1335 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
1336 timeout_connect_A, timeout_connect_MX.
1338 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
1339 to the empty sender.
1341 39. The daemon was not analysing the content of -oX till after it had closed
1342 stderr and disconnected from the controlling terminal. This meant that any
1343 syntax errors were only noted on the panic log, and the return code from
1344 the command was 0. By re-arranging the code a little, I've made the
1345 decoding happen first, so such errors now appear on stderr, and the return
1346 code is 1. However, the actual setting up of the sockets still happens in
1347 the disconnected process, so errors there are still only recorded on the
1350 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
1351 connections (as happens on some IP stacks) was logged at start up time as
1352 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
1353 it from "IPv6 and IPv4", which means that two separate sockets are being
1356 41. The debug output for gethostbyname2() or getipnodebyname() failures now
1357 says whether AF_INET or AF_INET6 was passed as an argument.
1359 42. Exiwhat output was messed up when time zones were included in log
1362 43. Exiwhat now gives more information about the daemon's listening ports,
1363 and whether -tls-on-connect was used.
1365 44. The "port" option of the smtp transport is now expanded.
1367 45. A "message" modifier in a "warn" statement in a non-message ACL was being
1368 silently ignored. Now an error message is written to the main and panic
1371 46. There's a new ACL modifier called "logwrite" which writes to a log file
1372 as soon as it is encountered.
1374 47. Added $local_user_uid and $local_user_gid at routing time.
1376 48. Exim crashed when trying to verify a sender address that was being
1379 49. Exim was recognizing only a space character after ".include". It now also
1380 recognizes a tab character.
1382 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
1383 extracting the relevant information from the specification. The man page no
1384 longer contains scrambled data for the -d option, and I've added a section
1385 at the front about calling Exim under different names.
1387 51. Added "extra_headers" argument to the "mail" command in filter files.
1389 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
1392 53. Installed eximstats 1.29.
1394 54. Added transport_filter_timeout as a generic transport option.
1396 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
1397 Cc: header lines. This was required by RFC 822, but it not required by RFC
1400 56. Exim used to add From:, Date:, and Message-Id: header lines to any
1401 incoming messages that did not have them. Now it does so only if the
1402 message originates locally, that is, if there is no associated remote host
1403 address. When Resent- header lines are present, this applies to the Resent-
1404 lines rather than the non-Resent- lines.
1406 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
1407 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
1409 58. Messages for configuration errors now include the name of the main
1410 configuration file - useful now that there may be more than one file in a
1411 list (.included file names were always shown).
1413 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
1414 for those rare installations that do not start the daemon as root or run it
1415 setuid root. I've cut out the call to initgroups() if the daemon is not
1418 60. The Exim user and group can now be bound into the binary as text strings
1419 that are looked up at the start of Exim's processing.
1421 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
1423 62. Added $mailstore_basename variable.
1425 63. Installed patch to sieve.c from Michael Haardt.
1427 64. When Exim failed to open the panic log after failing to open the main log,
1428 the original message it was trying to log was written to stderr and debug
1429 output, but if they were not available (the usual case in production), it
1430 was lost. Now it is written to syslog before the two lines that record the
1431 failures to open the logs.
1433 65. Users' Exim filters run in subprocesses under the user's uid. It is
1434 possible for a "deliver" command or an alias in a "personal" command to
1435 provoke an address rewrite. If logging of address rewriting is configured,
1436 this fails because the process is not running as root or exim. There may be
1437 a better way of dealing with this, but for the moment (because 4.30 needs
1438 to be released), I have disabled address rewrite logging when running a
1439 filter in a non-root, non-exim process.
1445 1. The buildconfig auxiliary program wasn't quoting the value set for
1446 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
1447 not defined. This bug was masked in 4.22 by the effect that was fixed in
1450 2. Some messages that were rejected after a message id was allocated were
1451 shown as "incomplete" by exigrep. It no longer does this for messages that
1452 are rejected by local_scan() or the DATA or non-SMTP ACLs.
1454 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
1455 have allow_domain_literals set, the ID did not get logged in the <= line.
1456 Domain literals are now always recognized in Message-ID: header lines.
1458 4. The first argument for a ${extract expansion item is the key name or field
1459 number. Leading and trailing spaces in this item were not being ignored,
1460 causing some misleading effects.
1462 5. When deliver_drop_privilege was set, single queue runner processes started
1463 manually (i.e. by the command "exim -q") or by the daemon (which uses the
1464 same command in the process it spins off) were not dropping privilege.
1466 6. When the daemon running as "exim" started a queue runner, it always
1467 re-executed Exim in the spun-off process. This is a waste of effort when
1468 deliver_drop_privilege is set. The new process now just calls the
1469 queue-runner function directly.
1475 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
1478 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
1479 ignored. Though the use of -oP was forcing the writing of a pid file, it
1480 was always written to the default place.
1482 3. If the message "no IP address found for host xxxx" is generated during
1483 incoming verification, it is now followed by identification of the incoming
1484 connection (so you can more easily find what provoked it).
1486 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
1488 5. Added some features to "harden" Exim a bit more against certain attacks:
1490 (a) There is now a build-time option called FIXED_NEVER_USERS that can
1491 be put in Local/Makefile. This is like the never_users runtime option,
1492 but it cannot be overridden. The default setting is "root".
1494 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
1495 prefix string with which any file named in a -C command line option
1498 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
1499 is retained for -C and -D only if the caller of Exim is root. Without
1500 it, the exim user may also use -C and -D and retain privilege.
1502 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
1503 command line option is disabled.
1505 6. Macro names set by the -D option must start with an upper case letter, just
1506 like macro names defined in the configuration file.
1508 7. Added "dereference=" facility to LDAP.
1510 8. Two instances of the typo "uknown" in the source files are fixed.
1512 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
1513 the Configure-Makefile script screwed up while processing it.
1515 10. Incorporated PCRE 4.4.
1517 11. The SMTP synchronization check was not operating right at the start of an
1518 SMTP session. For example, it could not catch a HELO sent before the client
1519 waited for the greeting. There is now a check for outstanding input at the
1520 point when the greeting is written. Because of the duplex, asynchronous
1521 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
1522 way, but not yet received, when the check is performed.
1524 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
1525 on TCP/IP sockets, because this apparently causes some broken clients to
1528 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
1529 unchanged) from the Cygwin maintainer.
1531 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
1532 was supported instead of testing for mbx. Effectively a typo.
1534 15. The spa authenticator server code was not checking that the input it
1535 received was valid base64.
1537 16. The debug output line for the "set" modifier in ACLs was not showing the
1538 name of the variable that was being set.
1540 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
1542 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
1543 lookup on its own. Something else was needed to trigger the lookup. For
1544 example, a match in host_lookup or the need for a host name in a host list.
1545 Now, if $sender_host_name is referenced and the host name has not yet been
1546 looked up, a lookup is performed. If the lookup fails, the variable remains
1547 empty, and $host_lookup_failed is set to "1".
1549 19. Added "eqi" as a case-independent comparison operator.
1551 20. The saslauthd authentication condition could segfault if neither service
1552 nor realm was specified.
1554 21. If an overflowing value such as "2048M" was set for message_size_limit, the
1555 error message that was logged was misleading, and incoming SMTP
1556 connections were dropped. The message is now more accurate, and temporary
1557 errors are given to SMTP connections.
1559 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
1560 (except RSET) with a 421 error, until QUIT is received. However, it was
1561 failing to send a response to QUIT.
1563 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
1564 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
1565 ACL is now run after the helo_try_verify_hosts code.
1567 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
1568 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
1569 case-independent, so other case variants are also recognized). Apparently
1570 some systems use these upper case variants.
1572 25. If more than two messages were waiting for the same host, and a transport
1573 filter was specified for the transport, Exim sent two messages over the
1574 same TCP/IP connection, and then failed with "socket operation on non-
1575 socket" when it tried to send the third.
1577 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
1579 27. The extern definition of crypt16() in expand.c was not being excluded when
1580 the OS had its own crypt16() function.
1582 28. Added bounce_return_body as a new option, and bounce_return_size_limit
1583 as a preferred synonym for return_size_limit, both as an option and as an
1586 29. Added LIBS=-liconv to OS/Makefile-OSF1.
1588 30. Changed the default configuration ACL to relax the local part checking rule
1589 for addresses that are not in any local domains. For these addresses,
1590 slashes and pipe symbols are allowed within local parts, but the sequence
1591 /../ is explicitly forbidden.
1593 31. SPA server authentication was not clearing the challenge buffer before
1596 32. log_message in a "warn" ACL statement was writing to the reject log as
1597 well as to the main log, which contradicts the documentation and doesn't
1598 seem right (because no rejection is happening). So I have stopped it.
1600 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
1601 However, I am unable to do any testing of this.
1603 34. Fixed an infelicity in the appendfile transport. When checking directories
1604 for a mailbox, to see if any needed to be created, it was accidentally
1605 using path names with one or more superfluous leading slashes; tracing
1606 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
1608 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
1609 contents are added to the log line that is written for every discarded
1610 recipient. (Previously a log_message setting was ignored.)
1612 36. The ${quote: operator now quotes the string if it is empty.
1614 37. The install script runs exim in order to find its version number. If for
1615 some reason other than non-existence or emptiness, which it checks, it
1616 could not run './exim', it was installing it with an empty version number,
1617 i.e. as "exim-". This error state is now caught, and the installation is
1620 38. An argument was missing from the function that creates an error message
1621 when Exim fails to connect to the socket for saslauthd authentication.
1622 This could cause Exim to crash, or give a corrupted message.
1624 39. Added isip, isip4, and isip6 to ${if conditions.
1626 40. The ACL variables $acl_xx are now saved with the message, and can be
1627 accessed later in routers, transports, and filters.
1629 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
1630 strings in the file are not string-expanded.
1632 42. If a MAIL command specified a SIZE value that was too large to fit into an
1633 int variable, the check against message_size_limit failed. Such values are
1634 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
1635 day this will have to be increased, but I don't think I want to be around
1636 when emails are that large.
1643 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
1644 iconv() is not standard in FreeBSD.
1646 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
1647 IPv6 enabled. The observed symptom was a segmentation fault on return from
1648 the function os_common_find_running_interfaces() in src/os.c.
1650 3. In the check_special_case() function in daemon.c I had used "errno" as an
1651 argument name, which causes warnings on some systems. This was basically a
1652 typo, since it was named "eno" in the comments!
1654 4. The code that waits for the clock to tick (at a resolution of some fraction
1655 of a second) so as to ensure message-id uniqueness was always waiting for
1656 at least one whole tick, when it could have waited for less. [This is
1657 almost certainly not relevant at current processor speeds, where it is
1658 unlikely to ever wait at all. But we try to future-proof.]
1660 5. The function that sleeps for a time interval that includes fractions of a
1661 second contained a race. It did not block SIGALRM between setting the
1662 timer, and suspending (a couple of lines later). If the interval was short
1663 and the sigsuspend() was delayed until after it had expired, the suspension
1664 never ended. On busy systems this could lead to processes getting stuck for
1667 6. Some uncommon configurations may cause a lookup to happen in a queue runner
1668 process, before it forks any delivery processes. The open lookup caching
1669 mechanism meant that the open file or database connection was passed into
1670 the delivery process. The problem was that delivery processes always tidy
1671 up cached lookup data. This could cause a problem for the next delivery
1672 process started by the queue runner, because the external queue runner
1673 process does not know about the closure. So the next delivery process
1674 still has data in the lookup cache. In the case of a file lookup, there was
1675 no problem because closing a file descriptor in a subprocess doesn't affect
1676 the parent. However, if the lookup was caching a connection to a database,
1677 the connection was closed, and the second delivery process was likely to
1678 see errors such as "PGSQL: query failed: server closed the connection
1679 unexpectedly". The problem has been fixed by closing all cached lookups
1680 in a queue runner before running a delivery process.
1682 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
1683 seem to have the "const" qualifier which it has on other OS. I've
1686 8. Change 4.21/2 was too strict. It is only if there are two authenticators
1687 *of the same type* (client or server) with the same public name that an
1688 error should be diagnosed.
1690 9. When Exim looked up a host name for an IP address, but failed to find the
1691 original IP address when looking up the host name (a safety check), it
1692 output the message "<ip address> does not match any IP for NULL", which was
1693 confusing, to say the least. The bug was that the host name should have
1694 appeared instead of "NULL".
1696 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
1697 user that is built into the binary, and the -C or -D options is used, root
1698 privilege is dropped before the configuration file is read. In addition,
1699 logging is switched to stderr instead of the normal log files. If the
1700 configuration then re-defines the Exim user, the unprivileged environment
1701 is probably not what is expected, so Exim logs a panic warning message (but
1704 However, if deliver_drop_privilege is set, the unprivileged state may well
1705 be exactly what is intended, so the warning has been cut out in that case,
1706 and Exim is allowed to try to write to its normal log files.
1712 1. smtp_return_error_details was not giving details for temporary sender
1713 or receiver verification errors.
1715 2. Diagnose a configuration error if two authenticators have the same public
1718 3. Exim used not to create the message log file for a message until the first
1719 delivery attempt. This could be confusing when incoming messages were held
1720 for policy or load reasons. The message log file is now created at the time
1721 the message is received, and an initial "Received" line is written to it.
1723 4. The automatically generated man page for command line options had a minor
1724 bug that caused no ill effects; however, a more serious problem was that
1725 the procedure for building the man page automatically didn't always
1726 operate. Consequently, release 4.20 contains an out-of-date version. This
1727 shouldn't happen again.
1729 5. When building Exim with embedded Perl support, the script that builds the
1730 Makefile was calling 'perl' to find its compile-time parameters, ignoring
1731 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
1733 6. The freeze_tell option was not being used for messages that were frozen on
1734 arrival, either by an ACL or by local_scan().
1736 7. Added the smtp_incomplete_transaction log selector.
1738 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
1739 was accepting AUTH without a new EHLO.
1741 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
1742 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
1745 10. Logging of TCP/IP connections (when configured) now happens in the main
1746 daemon process instead of the child process, so that the TCP/IP connection
1747 count is more accurate (but it can never be perfect).
1749 11. The use of "drop" in a nested ACL was not being handled correctly in the
1750 outer ACL. Now, if condition failure induced by the nested "drop" causes
1751 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
1752 or "require"), the connection is dropped.
1754 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
1755 that yield "discard" can now be used with an "accept" or a "discard" verb,
1756 but an error is generated for any others (because I can't see a useful way
1757 to define what should happen).
1759 13. When an ACL is read dynamically from a file (or anywhere else), the lines
1760 are now processed in the same way as lines in the Exim configuration file.
1761 In particular, continuation lines are supported.
1763 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
1765 15. Added -ti meaning -t -i.
1767 16. Check for letters, digits, hyphens, and dots in the names of dnslist
1768 domains, and warn by logging if others are found.
1770 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
1771 returned from GIFCONF when Exim is trying to find the list of interfaces on
1772 a host. The code in os.c has been modified to copy each ifreq to an aligned
1773 structure in all cases.
1775 Also, in some cases, the returned ifreq's were being copied to a 'struct
1776 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
1777 means the last couple of bytes of an IPv6 address could be chopped if the
1778 ifreq contained only a normal sockaddr (14 bytes storage).
1780 18. Named domain lists were not supported in the hosts_treat_as_local option.
1781 An entry such as +xxxx was not recognized, and was treated as a literal
1784 19. Ensure that header lines added by a DATA ACL are included in the reject log
1785 if the ACL subsequently rejects the message.
1787 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
1788 MD5 (which is deprecated).
1790 21. When testing a filter file using -bf, Exim was writing a message when it
1791 took the sender from a "From " line in the message, but it was not doing so
1792 when it took $return_path from a Return-Path: header line. It now does.
1794 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
1795 with a valid header line field name (a series of printing characters
1796 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
1798 23. Changed "disc" in the source to "disk" to conform to the documentation and
1799 the book and for uniformity.
1801 24. Ignore Sendmail's -Ooption=value command line item.
1803 25. When execve() failed while trying to run a command in a pipe transport,
1804 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
1805 could be confused with a return value of 69 from the command itself. This
1806 has been changed to 127, the value the shell returns if it is asked to run
1807 a non-existent command. The wording for the related log line suggests a
1808 non-existent command as the problem.
1810 26. If received_header_text expands to an empty string, do not add a Received:
1811 header line to the message. (Well, it adds a token one on the spool, but
1812 marks it "old" so that it doesn't get used or transmitted.)
1814 27. Installed eximstats 1.28 (addition of -nt option).
1816 28. There was no check for failure on the call to getsockname() in the daemon
1817 code. This can fail if there is a shortage of resources on the system, with
1818 ENOMEM, for example. A temporary error is now given on failure.
1820 29. Contrary to the C standard, it seems that in some environments, the
1821 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
1822 program. Exim now does this explicitly; it affects the formatting of
1823 timestamps using strftime().
1825 30. If exiqsumm was given junk data, it threw up some uninitialized variable
1826 complaints. I've now initialized all the variables, to avoid this.
1828 32. Header lines added by a system filter were not being "seen" during
1829 transport-time rewrites.
1831 33. The info_callback() function passed to OpenSSL is set up with type void
1832 (*)(SSL *, int, int), as described somewhere. However, when calling the
1833 function (actually a macro) that sets it up, the type void(*)() is
1834 expected. I've put in a cast to prevent warnings from picky compilers.
1836 34. If a DNS black list lookup found a CNAME record, but there were no A
1837 records associated with the domain it pointed at, Exim crashed.
1839 35. If a DNS black list lookup returned more than one A record, Exim ignored
1840 all but the first. It now scans all returned addresses if a particular IP
1841 value is being sought. In this situation, the contents of the
1842 $dnslist_value variable are a list of all the addresses, separated by a
1845 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
1846 to accept a host name and all its aliases if the forward lookup for any of
1847 them yielded the IP address of the incoming connection. Now it accepts only
1848 those names whose forward lookup yields the correct IP address. Any other
1849 names are discarded. This closes a loophole whereby a rogue DNS
1850 administrator could create reverse DNS records to break through a
1851 wildcarded host restriction in an ACL.
1853 37. If a user filter or a system filter that ran in a subprocess used any of
1854 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
1855 the wrong values were passed to the pipe command ($thisaddress had the
1856 value of $0, $0 had the value of $1, etc). This bug was introduced by
1857 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
1859 38. Improved the line breaking for long SMTP error messages from ACLs.
1860 Previously, if there was no break point between 40 and 75 characters, Exim
1861 left the rest of the message alone. Two changes have been made: (a) I've
1862 reduced the minimum length to 35 characters; (b) if it can't find a break
1863 point between 35 and 75 characters, it looks ahead and uses the first one
1864 that it finds. This may give the occasional overlong line, but at least the
1865 remaining text gets split now.
1867 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
1868 file descriptors might be low, and that setting 1000 would always raise it.
1869 It turns out that in some environments, the limit is already over 1000 and
1870 that lowering it causes trouble. So now Exim takes care not to decrease it.
1872 40. When delivering a message, the value of $return_path is set to $sender_
1873 address at the start of routing (routers may change the value). By an
1874 oversight, this default was not being set up when an address was tested by
1875 -bt or -bv, which affected the outcome if any router or filter referred to
1878 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
1879 log only when "message" or "log_message" are set. However, if one of the
1880 conditions was an address verification, or a call to a nested ACL, the
1881 messages generated by the underlying test were being passed through. This
1882 no longer happens. The underlying message is available in $acl_verify_
1883 message for both "message" and "log_message" expansions, so it can be
1884 passed through if needed.
1886 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
1887 new expansion $bh_ to give the encoded byte string without charset
1888 translation. Translation happens only if iconv() is available; HAVE_ICONV
1889 indicates this at build time. HEADERS_CHARSET gives the charset to
1890 translate to; headers_charset can change it in the configuration, and
1891 "headers charset" can change it in an individual filter file.
1893 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
1894 that creates RFC 2047 encoded "words" labels them as that charset instead
1895 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
1896 expansion operator; (ii) when Exim creates a From: line for a local
1897 message; (iii) when a header line is rewritten to include a "phrase" part.
1899 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
1900 buggy, causing it to skip the first lines of messages whose message ID
1901 ended in 'D'. This would not have bitten before Exim release 4.14, because
1902 message IDs were unlikely to end in 'D' before then. The effect was to have
1903 incorrect size information for certain domains.
1905 45. #include "config.h" was missing at the start of the crypt16.c module. This
1906 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
1909 46. If there was a timeout during a "random" callout check, Exim treated it as
1910 a failure of the random address, and carried on sending RSET and the real
1911 address. If the delay was just some slowness somewhere, the response to the
1912 original RCPT would be taken as a response to RSET and so on, causing
1913 mayhem of various kinds.
1915 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
1916 when I implemented it. It didn't allow for the fact that some option values
1917 may legitimatetly be negative (e.g. size_addition), and it didn't even do
1918 the right test for positive values.
1920 48. Domain names in DNS records are case-independent. Exim always looks them up
1921 in lower case. Some resolvers return domain names in exactly the case they
1922 appear in the zone file, that is, they may contain uppercase letters. Not
1923 all resolvers do this - some return always lower case. Exim was treating a
1924 change of case by a resolver as a change of domain, similar to a widening
1925 of a domain abbreviation. This triggered its re-routing code and so it was
1926 trying to route what was effectively the same domain again. This normally
1927 caused routing to fail (because the router wouldn't handle the domain
1928 twice). Now Exim checks for this case specially, and just changes the
1929 casing of the domain that it ultimately uses when it transmits the message
1932 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
1935 50. If a filter generated a file delivery with a non-absolute name (possible if
1936 no home directory exists for the router), the forbid_file option was not
1939 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
1940 the existing equality matching.
1942 52. Exim was using ints instead of ino_t variables in some places where it was
1943 dealing with inode numbers.
1945 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
1946 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
1947 called TMPDIR, and if it finds it is different, it changes its value.
1949 54. The smtp_printf() function is now made available to local_scan() so
1950 additional output lines can be written before returning. There is also an
1951 smtp_fflush() function to enable the detection of a dropped connection.
1952 The variables smtp_input and smtp_batched_input are exported to
1955 55. Changed the default runtime configuration: the message "Unknown user"
1956 has been removed from the ACL, and instead placed on the localuser router,
1957 using the cannot_route_message feature. This means that any verification
1958 failures that generate their own messages won't get overridden. Similarly,
1959 the "Unrouteable address" message that was in the ACL for unverifiable
1960 relay addresses has also been removed.
1962 56. Added hosts_avoid_esmtp to the smtp transport.
1964 57. The exicyclog script was not checking for the esoteric option
1965 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
1966 will work only if exicyclog is run under the appropriate euid.
1968 58. Following a discussion on the list, the rules by which Exim recognises line
1969 endings on incoming messages have been changed. The -dropcr and drop_cr
1970 options are now no-ops, retained only for backwards compatibility. The
1971 following line terminators are recognized: LF CRLF CR. However, special
1972 processing applies to CR:
1974 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
1975 nor a local message in the state where . is a terminator.
1977 (ii) If a bare CR is encountered in a header line, an extra space is added
1978 after the line terminator so as not to end the header. The reasoning
1979 behind this is that bare CRs in header lines are most likely either
1980 to be mistakes, or people trying to play silly games.
1982 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
1983 was being incorrectly given as 18 bytes larger than it should have been.
1984 This is a VOB (very old bug).
1986 60. This may never have affected anything current, but just in case it has:
1987 When the local host is found other than at the start of a list of hosts,
1988 the local host, those with the same MX, and any that follow, are discarded.
1989 When the list in question was part of a longer list of hosts, the following
1990 hosts (not currently being processed) were also being discarded. This no
1991 longer happens. I'm not sure if this situation could ever has previously
1994 61. Added the "/MX" feature to lists of hosts in the manualroute and query
1997 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
1998 header. This is something that is recommended in a new Internet Draft, and
1999 is something that is documented as being done by Sendmail. There are two
2000 possible values. For messages generated by the autoreply transport, Exim
2003 Auto-Submitted: auto-replied
2005 whereas for all other generated messages (e.g. bounces) it adds
2007 Auto-Submitted: auto-generated
2009 63. The "personal" condition in filters now includes a test for the
2010 Auto-Submitted: header. If it contains the string "auto-" the message it
2011 not considered personal.
2013 64. Added rcpt_include_affixes as a generic transport option.
2015 65. Added queue_only_override (default true).
2017 66. Added the syslog_duplication option.
2019 67. If what should have been the first header line of a message consisted of
2020 a space followed by a colon, Exim was mis-interpreting it as a header line.
2021 It isn't of course - it is syntactically invalid and should therefore be
2022 treated as the start of the message body. The misbehaviour could have
2023 caused a number of strange effects, including loss of data in subsequent
2024 header lines, and spool format errors.
2026 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
2027 client host had authenticated. This control can now be exercised by an ACL
2028 for more flexibility.
2030 69. By default, callouts do not happen when testing with -bh. There is now a
2031 variant, -bhc, which does actually run the callout code, including
2032 consulting and updating the callout cache.
2034 70. Added support for saslauthd authentication, courtesy of Alexander
2037 71. If statvfs() failed on the spool or log directories while checking their
2038 size for availability, Exim confusingly gave the error "space shortage".
2039 Furthermore, in debugging mode it crashed with a floating point exception.
2040 These checks are done if check_{spool,log}_{space,inodes} are set, and when
2041 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
2042 serious problem, Exim now writes to the main and panic logs when this
2043 happens, with details of the failure. It then refuses to accept the
2044 incoming message, giving the message "spool directory problem" or "log
2045 directory problem" with a 421 code for SMTP messages.
2047 72. When Exim is about to re-exec itself, it ensures that the file descriptors
2048 0, 1, and 2 exist, because some OS complain for execs without them (see
2049 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
2050 descriptors. However, the code omitted to check that the open succeeded,
2051 causing mysterious errors if for some reason the permissions on /dev/null
2052 got screwed. Now Exim writes a message to the main and panic logs, and
2053 bombs out if it can't open /dev/null.
2055 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
2056 interact so that it is all more flexible. It is supposed to remain
2057 backwards compatible. Also added extra_local_interfaces.
2059 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
2060 to bomb out with an assertion failure - to the client this appears as a
2061 connection drop. This problem occurs in the part of the code that was taken
2062 from the Samba project. Fortunately, the assertion is in a very simple
2063 function, so I have fixed this by reproducing the function inline in the
2064 one place where it is called, and arranging for authentication to fail
2065 instead of killing the process with assert().
2067 75. The SPA client code was not working when the server requested OEM rather
2068 than Unicode encoding.
2070 76. Added code to make require_files with a specific uid setting more usable in
2071 the case where statting the file as root fails - usually a non-root-mounted
2072 NFS file system. When this happens and the failure is EACCES, Exim now
2073 forks a subprocess and does the per-uid checking as the relevant uid.
2075 77. Added process_log_path.
2077 78. If log_file_path was not explicitly set, a setting of check_log_space or
2078 check_log_inodes was ignored.
2080 79. If a space check for the spool or log partitions fails, the incident is now
2081 logged. Of course, in the latter case the data may get lost...
2083 80. Added the %p formatting code to string_format() so that it can be used to
2084 print addresses in debug_print(). Adjusted all the address printing in the
2085 debugging in store.c to use %p rather than %d.
2087 81. There was a concern that a line of code in smtp_in.c could overflow a
2088 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
2089 initially expressed, the concern was not well-founded, because trailing
2090 spaces are removed early. However, if the trailing spaces were followed by
2091 a NULL, they did not get removed, so the overflow was possible. Two fixes
2094 (a) I re-wrote the offending code in a cleaner fashion.
2095 (b) If an incoming SMTP command contains a NULL character, it is rejected
2098 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
2099 runs initgroups(), so that if the Exim user is in any additional groups,
2100 they will be used during message reception.
2106 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob