fixed ordering of FAQ TOC page
[exim.git] / doc / doc-txt / ChangeLog
90af77f4 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.27 2004/11/12 15:03:40 ph10 Exp $
3Change log file for Exim from version 4.21
7Exim version 4.44
10 1. Minor wording change to the doc/README.SIEVE file.
139059f6 12 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 13 computation of the current number of files was incorrect.
495ae4b0 14
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
17 place.
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.
35edf2ff 34
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".
4310. $recipients is now available in the predata ACL (oversight).
4511. 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.
5012. 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
52 sender verification.
5413. The value of address_data from a sender verification is now available in
55 $sender_address_data in subsequent conditions in the ACL statement.
5714. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5915. Added a new option "connect=<time>" to callout options, to set a different
60 connection timeout.
6216. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
63 was its contents. (It was OK if the option was not defined at all.)
6517. A "Completed" log line is now written for messages that are removed from
66 the spool by the -Mrm option.
6818. New variables $sender_verify_failure and $recipient_verify_failure contain
69 information about exactly what failed.
7119. Added -dd to debug only the daemon process.
7320. Incorporated Michael Haardt's patch to ldap.c for improving the way it
74 handles timeouts, both on the server side and network timeouts. Renamed the
75 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
7721. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
78 It is now set to "smtps".
8022. $host_address is now set to the target address during the checking of
81 ignore_target_hosts.
8323. When checking ignore_target_hosts for an ipliteral router, no host name was
84 being passed; this would have caused $sender_host_name to have been used if
85 matching the list had actually called for a host name (not very likely,
86 since this list is usually IP addresses). A host name is now passed as
87 "[x.x.x.x]".
8924. Changed the calls that set up the SIGCHLD handler in the daemon to use the
90 code that specifies a non-restarting handler (typically sigaction() in
91 modern systems) in an attempt to fix a rare and obscure crash bug.
9325. Narrowed the window for a race in the daemon that could cause it to ignore
94 SIGCHLD signals. This is not a major problem, because they are used only to
95 wake it up if nothing else does.
9726. A malformed maildirsize file could cause Exim to calculate negative values
98 for the mailbox size or file count. Odd effects could occur as a result.
99 The maildirsize information is now recalculated if the size or filecount
100 end up negative.
10227. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
103 support for a long time. Removed HAVE_SYS_VFS_H.
10528. Installed the latest version of exipick from John Jetmore.
10729. In an address list, if the pattern was not a regular expression, an empty
108 subject address (from a bounce message) matched only if the pattern was an
109 empty string. Non-empty patterns were not even tested. This was the wrong
110 because it is perfectly reasonable to use an empty address as part of a
111 database query. An empty address is now tested by patterns that are
112 lookups. However, all the other forms of pattern expect the subject to
113 contain a local part and a domain, and therefore, for them, an empty
114 address still always fails if the pattern is not itself empty.
117Exim version 4.43
120 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
121 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
122 Now it returns an integer. A number of calls were still expecting a T/F
123 return. Fortuitously, in all cases, the tests worked in OK situations,
124 which is the norm. However, things would have gone wrong on any write
125 failures on the smtp file descriptor. This function is used when sending
126 messages over SMTP and also when doing verify callouts.
128 2. When Exim is called to do synchronous delivery of a locally submitted
129 message (the -odf or -odi options), it no longer closes stderr before doing
130 the delivery.
132 3. Implemented the mua_wrapper option.
134 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
136 5. Implemented the functions header_remove(), header_testname(),
137 header_add_at_position(), and receive_remove_recipient(), and exported them
138 to local_scan().
140 6. If an ACL "warn" statement specified the addition of headers, Exim already
141 inserted X-ACL-Warn: at the start if there was no header name. However, it
142 was not making this test for the second and subsequent header lines if
143 there were newlines in the string. This meant that an invalid header could
144 be inserted if Exim was badly configured.
146 7. Allow an ACL "warn" statement to add header lines at the start or after all
147 the Received: headers, as well as at the end.
149 8. Added the rcpt_4xx retry error code.
151 9. Added postmaster_mailfrom=xxx to callout verification option.
15310. Added mailfrom=xxxx to the callout verification option, for verify=
154 header_sender only.
15611. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
157 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
15912. Inserted some casts to stop certain compilers warning when using pointer
160 differences as field lengths or precisions in printf-type calls (mostly
161 affecting debugging statements).
16313. Added optional readline() support for -be (dynamically loaded).
16514. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
166 same clock tick as a message's arrival, so that its received time was the
167 same as the "first fail" time on the retry record, and that message
168 remained on the queue past the ultimate address timeout, every queue runner
169 would try a delivery (because it was past the ultimate address timeout) but
170 after another failure, the ultimate address timeout, which should have then
171 bounced the address, did not kick in. This was a "< instead of <=" error;
172 in most cases the first failure would have been in the next clock tick
173 after the received time, and all would be well.
17515. The special items beginning with @ in domain lists (e.g. @mx_any) were not
176 being recognized when the domain list was tested by the match_domain
177 condition in an expansion string.
17916. Added the ${str2b64: operator.
18117. Exim was always calling setrlimit() to set a large limit for the number of
182 processes, without checking whether the existing limit was already
183 adequate. (It did check for the limit on file descriptors.) Furthermore,
184 errors from getrlimit() and setrlimit() were being ignored. Now they are
185 logged to the main and panic logs, but Exim does carry on, to try to do its
186 job under whatever limits there are.
18818. Imported PCRE 5.0.
19019. Trivial typo in log message " temporarily refused connection" (the leading
191 space).
19320. If the log selector return_path_on_delivery was set and an address was
194 redirected to /dev/null, the delivery process crashed because it assumed
195 that a return path would always be set for a "successful" delivery. In this
196 case, the whole delivery is bypassed as an optimization, and therefore no
197 return path is set.
19921. Internal re-arrangement: the function for sending a challenge and reading
200 a response while authentication was assuming a zero-terminated challenge
201 string. It's now changed to take a pointer and a length, to allow for
202 binary data in such strings.
20422. Added the cyrus_sasl authenticator (code supplied by MBM).
20623. Exim was not respecting finduser_retries when seeking the login of the
207 uid under which it was called; it was always trying 10 times. (The default
208 setting of finduser_retries is zero.) Also, it was sleeping after the final
209 failure, which is pointless.
21124. Implemented tls_on_connect_ports.
21325. Implemented acl_smtp_predata.
21526. If the domain in control=submission is set empty, Exim assumes that the
216 authenticated id is a complete email address when it generates From: or
217 Sender: header lines.
21927. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
220 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
221 chown and chgrp in /bin and hostname in /usr/bin.
22328. Exim was keeping the "process log" file open after each use, just as it
224 does for the main log. This opens the possibility of it remaining open for
225 long periods when the USR1 signal hits a daemon. Occasional processlog
226 errors were reported, that could have been caused by this. Anyway, it seems
227 much more sensible not to leave this file open at all, so that is what now
228 happens.
23029. The long-running daemon process does not normally write to the log once it
231 has entered its main loop, and it closes the log before doing so. This is
232 so that log files can straightforwardly be renamed and moved. However,
233 there are a couple of unusual error situations where the daemon does write
234 log entries, and I had neglected to close the log afterwards.
23630. The text of an SMTP error response that was received during a remote
237 delivery was being truncated at 512 bytes. This is too short for some of
238 the long messages that one sometimes sees. I've increased the limit to
239 1024.
24131. It is now possible to make retry rules that apply only when a message has a
242 specific sender, in particular, an empty sender.
24432. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
245 it possible to be selective about when SMTP synchronization is enforced.
24733. Added "control = caseful_local_part" and "control = "caselower_local_part".
24932. Implemented hosts_connection_nolog.
25133. Added an ACL for QUIT.
25334. Setting "delay_warning=" to disable warnings was not working; it gave a
254 syntax error.
25635. Added mailbox_size and mailbox_filecount to appendfile.
25836. Added control = no_multiline_responses to ACLs.
26037. There was a bug in the logic of the code that waits for the clock to tick
261 in the case where the clock went backwards by a substantial amount such
262 that the microsecond fraction of "now" was more than the microsecond
263 fraction of "then" (but the whole seconds number was less).
26538. Added support for the libradius Radius client library this is found on
266 FreeBSD (previously only the radiusclient library was supported).
269Exim version 4.42
272 1. When certain lookups returned multiple values in the form name=value, the
273 quoting of the values was not always being done properly. Specifically:
274 (a) If the value started with a double quote, but contained no whitespace,
275 it was not quoted.
276 (b) If the value contained whitespace other than a space character (i.e.
277 tabs or newlines or carriage returns) it was not quoted.
278 This fix has been applied to the mysql and pgsql lookups by writing a
279 separate quoting function and calling it from the lookup code. The fix
280 should probably also be applied to nisplus, ibase and oracle lookups, but
281 since I cannot test any of those, I have not disturbed their existing code.
283 2. A hit in the callout cache for a specific address caused a log line with no
284 reason for rejecting RCPT. Now it says "Previous (cached) callout
285 verification failure".
287 3. There was an off-by-one bug in the queryprogram router. An over-long
288 return line was truncated at 256 instead of 255 characters, thereby
289 overflowing its buffer with the terminating zero. As well as fixing this, I
290 have increased the buffer size to 1024 (and made a note to document this).
292 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
293 when Exim is waiting for an SMTP response from a remote server, Exim
294 restarts its select() call on the socket, thereby resetting its timeout.
295 This is not a problem when such interrupts are rare. Somebody set up a cron
296 job to run exiwhat every 2 minutes, which is less than the normal select()
297 timeout (5 or 10 minutes). This meant that the select() timeout never
298 kicked in because it was always reset. I have fixed this by comparing the
299 time when an interrupt arrives with the time at the start of the first call
300 to select(). If more time than the timeout has elapsed, the interrupt is
301 treated as a timeout.
303 5. Some internal re-factoring in preparation for the addition of Sieve
304 extensions (by MH). In particular, the "personal" test is moved to a
305 separate function, and given an option for scanning Cc: and Bcc: (which is
306 not set for Exim filters).
308 6. When Exim created an email address using the login of the caller as the
309 local part (e.g. when creating a From: or Sender: header line), it was not
310 quoting the local part when it contained special characters such as @.
312 7. Installed new OpenBSD configuration files.
314 8. Reworded some messages for syntax errors in "and" and "or" conditions to
315 try to make them clearer.
317 9. Callout options, other than the timeout value, were being ignored when
318 verifying sender addresses in header lines. For example, when using
320 verify = header_sender/callout=no_cache
322 the cache was (incorrectly) being used.
32410. Added a missing instance of ${EXE} to the exim_install script; this affects
325 only the Cygwin environment.
32711. When return_path_on_delivery was set as a log selector, if different remote
328 addresses in the same message used different return paths and parallel
329 remote delivery occurred, the wrong values would sometimes be logged.
330 (Whenever a remote delivery process finished, the return path value from
331 the most recently started remote delivery process was logged.)
33312. RFC 3848 specifies standard names for the "with" phrase in Received: header
334 lines when AUTH and/or TLS are in use. This is the "received protocol"
335 field. Exim used to use "asmtp" for authenticated SMTP, without any
336 indication (in the protocol name) for TLS use. Now it follows the RFC and
337 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
338 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
339 names appear in log lines as well as in Received: header lines.
34113. Installed MH's patches for Sieve to add the "copy" and "vacation"
342 extensions, and comparison tests, and to fix some bugs.
34414. Changes to the "personal" filter test:
346 (1) The test was buggy in that it was just doing the equivalent of
347 "contains" tests on header lines. For example, if a user's address was
348 anne@some.where, the "personal" test would incorrectly be true for
350 To: susanne@some.where
352 This test is now done by extracting each address from the header in turn,
353 and checking the entire address. Other tests that are part of "personal"
354 are now done using regular expressions (for example, to check local parts
355 of addresses in From: header lines).
357 (2) The list of non-personal local parts in From: addresses has been
358 extended to include "listserv", "majordomo", "*-request", and "owner-*",
359 taken from the Sieve specification recommendations.
361 (3) If the message contains any header line starting with "List-" it is
362 treated as non-personal.
364 (4) The test for "circular" in the Subject: header line has been removed
365 because it now seems ill-conceived.
36715. Minor typos in src/EDITME comments corrected.
36916. Installed latest exipick from John Jetmore.
37117. If headers_add on a router specified a text string that was too long for
372 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
373 of string_sprintf() is now avoided.
37518. $message_body_size was not set (it was always zero) when running the DATA
376 ACL and the local_scan() function.
37819. For the "mail" command in an Exim filter, no default was being set for
379 the once_repeat time, causing a random time value to be used if "once" was
380 specified. (If the value happened to be <= 0, no repeat happened.) The
381 default is now 0s, meaning "never repeat". The "vacation" command was OK
382 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
383 (I found it when inspecting the code).
38520. There is now an overall timeout for performing a callout verification. It
386 defaults to 4 times the callout timeout, which applies to individual SMTP
387 commands during the callout. The overall timeout applies when there is more
388 than one host that can be tried. The timeout is checked before trying the
389 next host. This prevents very long delays if there are a large number of
390 hosts and all are timing out (e.g. when the network connections are timing
391 out). The value of the overall timeout can be changed by specifying an
392 additional sub-option for "callout", called "maxwait". For example:
394 verify = sender/callout=5s,maxwait=20s
39621. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
397 to the end before writing, but this should make it even safer).
39922. Exim was forgetting that it had advertised PIPELINING for the second and
400 subsequent messages on an SMTP connection. It was also not resetting its
401 memory on STARTTLS and an internal HELO.
40323. When Exim logs an SMTP synchronization error within a session, it now
404 records whether PIPELINING has been advertised or not.
40624. Added 3 instances of "(long int)" casts to time_t variables that were being
407 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
408 rather than long int.
41025. Installed the latest Cygwin configuration files from the Cygwin maintainer.
41226. Added the never_mail option to autoreply.
415Exim version 4.41
418 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
419 crash if the getsockname() call failed; this can happen if a connection is
420 closed very soon after it is established. The problem was simply in the
421 order in which certain operations were done, causing Exim to try to write
422 to the SMTP stream before it had set up the file descriptor. The bug has
423 been fixed by making things happen in the correct order.
426Exim version 4.40
429 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
430 before the connection was closed, thus losing the rejection response.
432 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
433 some early Solaris releases, but causes trouble in current releases where
434 socklen_t is defined.
436 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
437 always exist.
439 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
440 configured.
442 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
443 that releasing the top of it at the end releases what was used for sub-
444 expansions (unless the block got too big). However, discard this block if
445 the first thing is a variable or header, so that we can use its block when
446 it is dynamic (useful for very large $message_headers, for example).
448 6. Lookups now cache *every* query, not just the most recent. A new, separate
449 store pool is used for this. It can be recovered when all lookup caches are
450 flushed. Lookups now release memory at the end of their result strings.
451 This has involved some general refactoring of the lookup sources.
453 7. Some code has been added to the store_xxx() functions to reduce the amount
454 of flapping under certain conditions.
456 8. log_incoming_interface used to affect only the <= reception log lines. Now
457 it causes the local interface and port to be added to several more SMTP log
458 lines, for example "SMTP connection from", and rejection lines.
460 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
46210. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
46411. If $host_data was set by virtue of a hosts lookup in an ACL, its value
465 could be overwritten at the end of the current message (or the start of a
466 new message if it was set in a HELO ACL). The value is now preserved for
467 the duration of the SMTP connection.
46912. If a transport had a headers_rewrite setting, and a matching header line
470 contained an unqualified address, that address was qualified, even if it
471 did not match any rewriting rules. The underlying bug was that the values
472 of the flags that permit the existence of unqualified sender and recipient
473 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
474 non-local messages, and by -bnq for local messages) were not being
475 preserved with the message after it was received.
47713. When Exim was logging an SMTP synchronization error, it could sometimes log
478 "next input=" as part of the text comprising the host identity instead of
479 the correct text. The code was using the same buffer for two different
480 strings. However, depending on which order the printing function evaluated
481 its arguments, the bug did not always show up. Under Linux, for example, my
482 test suite worked just fine.
48414. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
485 doesn't work with some older versions of Perl. It has been changed to "my",
486 which in any case is probably the better facility to use.
48815. A really picky compiler found some instances of statements for creating
489 error messages that either had too many or two few arguments for the format
490 string.
49216. The size of the buffer for calls to the DNS resolver has been increased
493 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
494 for addresses that have a lot of PTR records. This alleviates a problem; it
495 does not fully solve it.
49717. A dnsdb lookup for PTR records that receives more data than will fit in the
498 buffer now truncates the list and logs the incident, which is the same
499 action as happens when Exim is looking up a host name and its aliases.
500 Previously in this situation something unpredictable would happen;
501 sometimes it was "internal error: store_reset failed".
50318. If a server dropped the connection unexpectedly when an Exim client was
504 using GnuTLS and trying to read a response, the client delivery process
505 crashed while trying to generate an error log message.
50719. If a "warn" verb in an ACL added multiple headers to a message in a single
508 string, for example:
510 warn message = H1: something\nH2: something
512 the text was added as a single header line from Exim's point of view
513 though it ended up OK in the delivered message. However, searching for the
514 second and subsequent header lines using $h_h2: did not work. This has been
515 fixed. Similarly, if a system filter added multiple headers in this way,
516 the routers could not see them.
51820. Expanded the error message when iplsearch is called with an invalid key to
519 suggest using net-iplsearch in a host list.
52121. When running tests using -bh, any delays imposed by "delay" modifiers in
522 ACLs are no longer actually imposed (and a message to that effect is
523 output).
52522. If a "gecos" field in a passwd entry contained escaped characters, in
526 particular, if it contained a \" sequence, Exim got it wrong when building
527 a From: or a Sender: header from that name. A second bug also caused
528 incorrect handling when an unquoted " was present following a character
529 that needed quoting.
53123. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
532 was not being matched caselessly.
53424. Arranged for all hyphens in the exim.8 source to be escaped with
535 backslashes.
53725. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
538 properly. Recipient callout cache records were still being keyed to include
539 the sender, even when use_sender was set false. This led to far more
540 callouts that were necessary. The sender is no longer included in the key
541 when use_sender is false.
54326. Added "control = submission" modifier to ACLs.
54527. Added the ${base62d: operator to decode base 62 numbers.
54728. dnsdb lookups can now access SRV records.
54929. CONFIGURE_OWNER can be set at build time to define an alternative owner for
550 the configuration file.
55230. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
553 (-v) mode. This makes the output for a verbose queue run more intelligible.
55531. Added a use_postmaster feature to recipient callouts.
55732. Added the $body_zerocount variable, containing the number of binary zero
558 bytes in the message body.
56033. The time of last modification of the "new" subdirectory is now used as the
561 "mailbox time last read" when there is a quota error for a maildir
562 delivery.
56434. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
56635. Added +ignore_unknown as a special item in host lists.
56836. Code for decoding IPv6 addresses in host lists is now included, even if
569 IPv6 support is not being compiled. This fixes a bug in which an IPv6
570 address was recognized as an IP address, but was then not correctly decoded
571 into binary, causing unexpected and incorrect effects when compared with
572 another IP address.
575Exim version 4.34
578 1. Very minor rewording of debugging text in manualroute to say "list of
579 hosts" instead of "hostlist".
581 2. If verify=header_syntax was set, and a header line with an unqualified
582 address (no domain) and a large number of spaces between the end of the
583 name and the colon was received, the reception process suffered a buffer
584 overflow, and (when I tested it) crashed. This was caused by some obsolete
585 code that should have been removed. The fix is to remove it!
587 3. When running in the test harness, delay a bit after writing a bounce
588 message to get a bit more predictability in the log output.
590 4. Added a call to search_tidyup() just before forking a reception process. In
591 theory, someone could use a lookup in the expansion of smtp_accept_max_
592 per_host which, without the tidyup, could leave open a database connection.
594 5. Added the variables $recipient_data and $sender_data which get set from a
595 lookup success in an ACL "recipients" or "senders" condition, or a router
596 "senders" option, similar to $domain_data and $local_part_data.
598 6. Moved the writing of debug_print from before to after the "senders" test
599 for routers.
601 7. Change 4.31/66 (moving the time when the Received: is generated) caused
602 problems for message scanning, either using a data ACL, or using
603 local_scan() because the Received: header was not generated till after they
604 were called (in order to set the time as the time of reception completion).
605 I have revised the way this works. The header is now generated after the
606 body is received, but before the ACL or local_scan() are called. After they
607 are run, the timestamp in the header is updated.
610Exim version 4.33
613 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
614 before starting a queue runner without re-exec. This happened only when
615 deliver_drop_privilege was set or when the Exim user was set to root. The
616 effect of the bug was that timeouts during subsequent deliveries caused
617 crashes instead of being properly handled. The handler is now left at its
618 default (and expected) setting.
620 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
621 message, again when deliver_drop_privilege is set or Exim is run as root.
622 The bug described in (1) was not present in this case, but the tidying up
623 of the other signals was missing. I have made the two cases consistent.
625 3. The ignore_target_hosts setting on a manualroute router was being ignored
626 for hosts that were looked up using the /MX notation.
628 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
629 in domain lists.
631 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
632 operated on the sender address. After changing the $sender_address to <>
633 for the sender address verify, Exim was re-instated it as the original
634 (before rewriting) address, but remembering that it had rewritten it, so it
635 wasn't rewriting it again. This bug also had the effect of breaking the
636 sender address verification caching when the sender address was rewritten.
638 6. The ignore_target_hosts option was being ignored by the ipliteral router.
639 This has been changed so that if the ip literal address matches
640 ignore_target_hosts, the router declines.
642 7. Added expansion conditions match_domain, match_address, and match_local_
643 part (NOT match_host).
645 8. The placeholder for the Received: header didn't have a length field set.
647 9. Added code to Exim itself and to exim_lock to test for a specific race
648 condition that could lead to file corruption when using MBX delivery. The
649 issue is with the lockfile that is created in /tmp. If this file is removed
650 after a process has opened it but before that process has acquired a lock,
651 there is the potential for a second process to recreate the file and also
652 acquire a lock. This could lead to two Exim processes writing to the file
653 at the same time. The added code performs the same test as UW imapd; it
654 checks after acquiring the lock that its file descriptor still refers to
655 the same named file.
65710. The buffer for building added header lines was of fixed size, 8192 bytes.
658 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
659 when Exim is built.
66111. Added the smtp_active_hostname option. If used, this will typically be made
662 to depend on the incoming interface address. Because $interface_address is
663 not set up until the daemon has forked a reception process, error responses
664 that can happen earlier (such as "too many connections") no longer contain
665 a host name.
66712. If an expansion in a condition on a "warn" statement fails because a lookup
668 defers, the "warn" statement is abandoned, and the next ACL statement is
669 processed. Previously this caused the whole ACL to be aborted.
67113. Added the iplsearch lookup type.
67314. Added ident_timeout as a log selector.
67515. Added tls_certificate_verified as a log selector.
67716. Added a global option tls_require_ciphers (compare the smtp transport
678 option of the same name). This controls incoming TLS connections.
68017. I finally figured out how to make tls_require_ciphers do a similar thing
681 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
682 before starting the TLS session.
68418. Tabs are now shown as \t in -bP output.
68619. If the log selector return_path_on_delivery was set, Exim crashed when
687 bouncing a message because it had too many Received: header lines.
68920. If two routers both had headers_remove settings, and the first one included
690 a superfluous trailing colon, the final name in the first list and the
691 first name in the second list were incorrectly joined into one item (with a
692 colon in the middle).
695Exim version 4.32
698 1. Added -C and -D options to the exinext utility, mainly to make it easier
699 to include in the automated testing, but these could be helpful when
700 multiple configurations are in use.
702 2. The exinext utility was not formatting the output nicely when there was
703 an alternate port involved in the retry record key, nor when there was a
704 message id as well (for retries that were specific to a specific message
705 and a specific host). It was also confused by IPv6 addresses, because of
706 the additional colons they contain. I have fixed the IPv4 problem, and
707 patched it up to do a reasonable job for IPv6.
709 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
710 delivery, the log line now contains "pipelined" if PIPELINING was used.
712 4. An SMTP transport process used to panic and die if the bind() call to set
713 an explicit outgoing interface failed. This has been changed; it is now
714 treated in the same way as a connect() failure.
716 5. A reference to $sender_host_name in the part of a conditional expansion
717 that was being skipped was still causing a DNS lookup. This no longer
718 occurs.
720 6. The def: expansion condition was not recognizing references to header lines
721 that used bh_ and bheader_.
723 7. Added the _cache feature to named lists.
725 8. The code for checking quota_filecount in the appendfile transport was
726 allowing one more file than it should have been.
728 9. For compatibility with Sendmail, the command line option
730 -prval:sval
732 is equivalent to
734 -oMr rval -oMs sval
736 and sets the incoming protocol and host name (for trusted callers). The
737 host name and its colon can be omitted when only the protocol is to be set.
738 Note the Exim already has two private options, -pd and -ps, that refer to
739 embedded Perl. It is therefore impossible to set a protocol value of "d" or
740 "s", but I don't think that's a major issue.
74210. A number of refactoring changes to the code, none of which should affect
743 Exim's behaviour:
745 (a) The number of logging options was getting close to filling up the
746 32-bit word that was used as a bit map. I have split them into two classes:
747 those that are passed in the argument to log_write(), and those that are
748 only ever tested independently outside of that function. These are now in
749 separate 32-bit words, so there is plenty of room for expansion again.
750 There is no change in the user interface or the logging behaviour.
752 (b) When building, for example, log lines, the code previously used a
753 macro that called string_cat() twice, in order to add two strings. This is
754 not really sufficiently general. Furthermore, there was one instance where
755 it was actually wrong because one of the argument was used twice, and in
756 one call a function was used. (As it happened, calling the function twice
757 did not affect the overall behaviour.) The macro has been replaced by a
758 function that can join an arbitrary number of extra strings onto a growing
759 string.
761 (c) The code for expansion conditions now uses a table and a binary chop
762 instead of a serial search (which was left over from when there were very
763 few conditions). Also, it now recognizes conditions like "pam" even when
764 the relevant support is not compiled in: a suitably worded error message is
765 given if an attempt is made to use such a condition.
76711. Added ${time_interval:xxxxx}.
76912. A bug was causing one of the ddress fields not to be passed back correctly
770 from remote delivery subprocesses. The field in question was not being
771 subsequently used, so this caused to problems in practice.
77313. Added new log selectors queue_time and deliver_time.
77514. Might have fixed a bug in maildirsizefile handling that threw up
776 "unexpected character" debug warnings, and recalculated the data
777 unnecessarily. In any case, I expanded the warning message to give more
778 information.
78015. Added the message "Restricted characters in address" to the statements in
781 the default ACL that block characters like @ and % in local parts.
78316. Change 71 for release 4.31 proved to be much less benign that I imagined.
784 Three changes have been made:
786 (a) There was a serious bug; a negative response to MAIL caused the whole
787 recipient domain to be cached as invalid, thereby blocking all messages
788 to all local parts at the same domain, from all senders. This bug has
789 been fixed. The domain is no longer cached after a negative response to
790 MAIL if the sender used is not empty.
792 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
793 been restored.
795 (c) A new callout option, "use_sender" has been added for people who want
796 the modified behaviour.
799Exim version 4.31
802 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
803 Larry Rosenman.
805 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
806 indeed breaks things for older releases.
808 3. Added additional logging to the case where there is a problem reading data
809 from a filter that is running in a subprocess using a pipe, in order to
810 try to track down a specific problem.
812 4. Testing facility fudge: when running in the test harness and attempting
813 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
814 getting "No route to host". Convert this to a timeout.
816 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
817 warning.
819 6. Some OS don't have socklen_t but use size_t instead. This affects the
820 fifth argument of getsockopt() amongst other things. This is now
821 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
822 can be set for individual OS. I have set it for SunOS5, OSF1, and
823 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
824 some earlier ones do not.
826 7. Change 4.30/15 was not doing the test caselessly.
828 8. The standard form for an IPv6 address literal was being rejected by address
829 parsing in, for example, MAIL and RCPT commands. An example of this kind of
830 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
831 this, as well as the form without the "IPv6" on the front (but only when
832 address literals are enabled, of course).
834 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
83610. Exim crashed if a message with an empty sender address specified by -f
837 encountered a router with an errors_to setting. This could be provoked only
838 by a command such as
840 exim -f "" ...
842 where an empty string was supplied; "<>" did not hit this bug.
84411. Installed PCRE release 4.5.
84612. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
847 remained set. It is now erased.
84913. exiqgrep wasn't working on MacOS X because it didn't correctly compute
850 times from message ids (which are base 36 rather than the normal 62).
85214. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
853 were being counted as actual protocol errors, and logged if the log
854 selector +smtp_protocol_error was set. One cannot be perfect in this test,
855 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
856 and DATA following a set of rejected RCPTs do not count as protocol errors.
857 In other words, Exim assumes they were pipelined, though this may not
858 actually be the case. Of course, in all cases the client gets an
859 appropriate error code.
86115. If a lookup fails in an ACL condition, a message about the failure may
862 be available; it is used if testing the ACL cannot continue, because most
863 such messages specify what the cause of the deferral is. However, some
864 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
865 that caused an old message to be retained and used if a later statement
866 caused a defer, replacing the real cause of the deferral.
86816. If an IP address had so many PTR records that the DNS lookup buffer
869 was not large enough to hold them, Exim could crash while trying to process
870 the truncated data. It now detects and logs this case.
87217. Further to 4.21/58, another change has been made: if (and only if) the
873 first line of a message (the first header line) ends with CRLF, a bare LF
874 in a subsequent header line has a space inserted after it, so as not to
875 terminate the header.
87718. Refactoring: tidied an ugly bit of code in appendfile that copied data
878 unnecessarily, used atoi() instead of strtol(), and didn't check the
879 termination when getting file sizes from file names by regex.
88119. Completely re-implemented the support for maildirsize files, in the light
882 of a number of problems with the previous contributed implementation
883 (4.30/29). In particular:
885 . If the quota is zero, the maildirsize file is maintained, but no quota is
886 imposed.
888 . If the maildir directory does not exist, it is created before any attempt
889 to write a maildirsize file.
891 . The quota value in the file is just a cache; if the quota is changed in
892 the transport, the new value overrides.
894 . A regular expression is available for excluding directories from the
895 count.
89720. The autoreply transport checks the characters in options that define the
898 message's headers; it allows continued headers, but it was checking with
899 isspace() after an embedded newline instead of explicitly looking for a
900 space or a tab.
90221. If all the "regular" hosts to which an address was routed had passed their
903 expiry times, and had not reached their retry times, the address was
904 bounced, even if fallback hosts were defined. Now Exim should go on to try
905 the fallback hosts.
90722. Increased buffer sizes in the callout code from 1024 to 4096 to match the
908 equivalent code in the SMTP transport. Some hosts send humungous responses
909 to HELO/EHLO, more than 1024 it seems.
91123. Refactoring: code in filter.c used (void *) for "any old type" but this
912 gives compiler warnings in some environments. I've now done it "properly",
913 using a union.
91524. The replacement for inet_ntoa() that is used with gcc on IRIX systems
916 (because of problems with the built-in one) was declared to return uschar *
917 instead of char *, causing compiler failure.
91925. Fixed a file descriptor leak when processing alias/forward files.
92126. Fixed a minor format string issue in dbfn.c.
92327. Typo in exim.c: ("dmbnz" for "dbmnz").
92528. If a filter file refered to $h_xxx or $message_headers, and the headers
926 contained RFC 2047 "words", Exim's memory could, under certain conditions,
927 become corrupted.
92929. When a sender address is verified, it is cached, to save repeating the test
930 when there is more than one recipient in a message. However, when the
931 verification involves a callout, it is possible for different callout
932 options to be set for different recipients. It is too complicated to keep
933 track of this in the cache, so now Exim always runs a verification when a
934 callout is required, relying on the callout cache for the optimization.
935 The overhead is duplication of the address routing, but this should not be
936 too great.
93830. Fixed a bug in callout caching. If a RCPT command caused the sender address
939 to be verified with callout=postmaster, and the main callout worked but the
940 postmaster check failed, the verification correctly failed. However, if a
941 subsequent RCPT command asked for sender verification *without* the
942 postmaster check, incorrect caching caused this verification also to fail,
943 incorrectly.
94531. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
946 it was not caching the DNS options (qualify_single, search_parents) that
947 were used when the lookup failed. A subsequent lookup with different
948 options therefore always gave the same answer, though there were cases
949 where it should not have. (Example: a "domains = !$mx_any" option on a
950 dnslookup router: the "domains" option is always processed without any
951 widening, but the router might have qualify_single set.) Now Exim uses the
952 cached value only when the same options are set.
95432. Added John Jetmore's "exipick" utility to the distribution.
95633. GnuTLS: When an attempt to start a TLS session fails for any reason other
957 than a timeout (e.g. a certificate is required, and is not provided), an
958 Exim server now closes the connection immediately. Previously it waited for
959 the client to close - but if the client is SSL, it seems that they each
960 wait for each other, leading to a delay before one of them times out.
96234: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
963 maintained 0.8.x compatibility because I don't think many are using it, and
964 it is clearly obsolete.
96635. Added TLS support for CRLs: a tls_crl global option and one for the smtp
967 transport.
96936. OpenSSL: $tls_certificate_verified was being set to 1 even if the
970 client certificate was expired. A simple patch fixes this, though I don't
971 understand the full logic of why the verify callback is called multiple
972 times.
97437. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
975 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
976 which causes problems with some clients (such as the Certicom SSL Plus
977 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
978 disables the coutermeasure allowing Eudora to connect."
98038. Exim was not checking that a write() to a log file succeeded. This could
981 lead to Bad Things if a log got too big, in particular if it hit a file
982 size limit. Exim now panics and dies if it cannot write to a log file, just
983 as it does if it cannot open a log file.
98539. Modified OS/Makefile-Linux so that it now contains
989 The two -D definitions ensure that Exim is compiled with large file
990 support, which makes it possible to handle log files that are bigger than
991 2^31.
99340. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
994 instance) a domain was checked against a named list that involved a lookup,
995 causing $domain_data to be set, then another domain was checked against the
996 same list, then the first domain was re-checked, the value of $domain_data
997 after the final check could be wrong. In particular, if the second check
998 failed, it could be set empty. This bug probably also applied to
999 $localpart_data.
100141. The strip_trailing_dot option was not being applied to the address given
1002 with the -f command-line option.
100442. The code for reading a message's header from the spool was incrementing
1005 $received_count, but never initializing it. This meant that the value was
1006 incorrect (doubled) while delivering a message in the same process in which
1007 it was received. In the most common configuration of Exim, this never
1008 happens - a fresh exec is done - but it can happen when
1009 deliver_drop_privilege is set.
101143. When Exim logs an SMTP synchronization error - client data sent too soon -
1012 it now includes up to 150 characters of the unexpected data in the log
1013 line.
101544. The exim_dbmbuild utility uses fixed size buffers for reading input lines
1016 and building data strings. The size of both of these buffers was 10 000
1017 bytes - far larger than anybody would *ever* want, thought I. Needless to
1018 say, somebody hit the limit. I have increased the maximum line length to
1019 20 000 and the maximum data length of concatenated lines to 100 000. I have
1020 also fixed two bugs, because there was no checking on these buffers. Tsk,
1021 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1022 buffer is too small.
102445. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1025 lsearch lookups. Now it does.
102746. When parsing a route_list item in a manualroute router, a fixed-length
1028 buffer was used for the list of hosts. I made this 1024 bytes long,
1029 thinking that nobody would ever have a list of hosts that long. Wrong.
1030 Somebody had a whole pile of complicated expansion conditions, and the
1031 string was silently truncated, leading to an expansion error. It turns out
1032 that it is easier to change to an unlimited length (owing to other changes
1033 that have happened since this code was originally written) than to build
1034 structure for giving a limitation error. The length of the item that
1035 expands into the list of hosts is now unlimited.
103747. The lsearch lookup could not handle data where the length of text line was
1038 more than 4095 characters. Such lines were truncated, leading to shortened
1039 data being returned. It should now handle lines of any length.
104148. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
1042 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
1043 ACL").
104549. Cosmetic tidy to scripts like exicyclog that are generated by globally
1046 replacing strings such as BIN_DIRECTORY in a source file: the replacement
1047 no longer happens in comment lines. A list of replacements is now placed
1048 at the head of all of the source files, except those whose only change is
1049 to replace PERL_COMMAND in the very first #! line.
105150. Replaced the slow insertion sort in queue.c, for sorting the list of
1052 messages on the queue, with a bottom-up merge sort, using code contributed
1053 by Michael Haardt. This should make operations like -bp somewhat faster on
1054 large queues. It won't affect queue runners, except when queue_run_in_order
1055 is set.
105751. Installed eximstats 1.31 in the distribution.
105952. Added support for SRV lookups to the dnslookup router.
106153. If an ACL referred to $message_body or $message_body_end, the value was not
1062 reset for any messages that followed in the same SMTP session.
106454. The store-handling optimization for building very long strings was not
1065 differentiating between the different store pools. I don't think this
1066 actually made any difference in practice, but I've tidied it.
106855. While running the routers to verify a sender address, $sender_address
1069 was still set to the sender address. This is wrong, because when routing to
1070 send a bounce to the sender, it would be empty. Therefore, I have changed
1071 it so that, while verifying a sender address, $sender_address is set to <>.
1072 (There is no change to what happens when verifying a recipient address.)
107456. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
1075 target A or AAAA records (if not already returned) without resetting the
1076 qualify_single or search_parents options of the DNS resolver. These are
1077 inappropriate in this case because the targets of MX and SRV records must
1078 be FQDNs. A broken DNS record could cause trouble if it happened to have a
1079 target that, when qualified, matched something in the local domain. These
1080 two options are now turned off when doing these lookups.
108257. It seems that at least some releases of Reiserfs (which does not have the
1083 concept of a fixed number of inodes) returns zero and not -1 for the
1084 number of available inodes. This interacted badly with check_spool_inodes,
1085 which assumed that -1 was the "no such thing" setting. What I have done is
1086 to check that the total number of inodes is greater than zero before doing
1087 the test of how many are available.
108958. When a "warn" ACL statement has a log_message modifier, the message is
1090 remembered, and not repeated. This is to avoid a lot of repetition when a
1091 message has many recipients that cause the same warning to be written.
1092 Howewer, Exim was preserving the list of already written lines for an
1093 entire SMTP session, which doesn't seem right. The memory is now reset if a
1094 new message is started.
109659. The "rewrite" debugging flag was not showing the result of rewriting in the
1097 debugging output unless log_rewrite was also set.
109960. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
1100 of (int)(handle) when we know that handle contains (void *)(-1).
110261. The Exim daemon panic-logs an error return when it closes the incoming
1103 connection. However "connection reset by peer" seems to be common, and
1104 isn't really an error worthy of noting specially, so that particular error
1105 is no long logged.
110762. When Exim is trying to find all the local interfaces, it used to panic and
1108 die if the ioctl to get the interface flags failed. However, it seems that
1109 on at least one OS (Solaris 9) it is possible to have an interface that is
1110 included in the list of interfaces, but for which you get a failure error
1111 for this call. This happens when the interface is not "plumbed" into a
1112 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
1113 failure of the "get flags" call assumes that the interface is down.
111563. Added a ${eval10: operator, which assumes all numbers are decimal. This
1116 makes life easier for people who are doing arithmetic on fields extracted
1117 from dates, where you often get leading zeros that should not be
1118 interpreted as octal.
112064. Added qualify_domain to the redirect router, to override the global
1121 setting.
112365. If a pathologically long header line contained very many addresses (the
1124 report of this problem mentioned 10 000) and each of them was rewritten,
1125 Exim could use up a very large amount of memory. (It kept on making new
1126 copies of the header line as it rewrote, and never released the old ones.)
1127 At the expense of a bit more processing, the header rewriting function has
1128 been changed so that it no longer eats memory in this way.
113066. The generation of the Received: header has been moved from the time that a
1131 message starts to be received, to the time that it finishes. The timestamp
1132 in the Received: header should now be very close to that of the <= log
1133 line. There are two side-effects of this change:
1135 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
1136 logged header lines no longer include the local Received: line, because
1137 it has not yet been created. The same applies to a copy of the message
1138 that is returned to a non-SMTP sender when a message is rejected.
1140 (b) When a filter file is tested using -bf, no additional Received: header
1141 is added to the test message. After some thought, I decided that this
1142 is a bug fix.
1144 This change does not affect the value of $received_for. It is still set
1145 after address rewriting, but before local_scan() is called.
114767. Installed the latest Cygwin-specific files from the Cygwin maintainer.
114968. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
1150 gave an unhelpful panic error message, and a defer error. I have managed to
1151 change this behaviour so that it now rejects any supplied certificate,
1152 which seems right, as the list of acceptable certificates is empty.
115469. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
1155 gave an unhelpful defer error. I have not managed to make this reject any
1156 supplied certificates, but the error message it gives is "no certificate
1157 supplied", which is not helpful.
115970. exigrep's output now also includes lines that are not associated with any
1160 message, but which match the given pattern. Implemented by a patch from
1161 Martin Sluka, which also tidied up the Perl a bit.
116371. Recipient callout verification, like sender verification, was using <> in
1164 the MAIL FROM command. This isn't really the right thing, since the actual
1165 sender may affect whether the remote host accepts the recipient or not. I
1166 have changed it to use the actual sender in the callout; this means that
1167 the cache record is now keyed on a recipient/sender pair, not just the
1168 recipient address. There doesn't seem to be a real danger of callout loops,
1169 since a callout by the remote host to check the sender would use <>.
1170 [SEE ABOVE: changed after hitting problems.]
117272. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
1173 temporary errors. However, in the case of such a code being given after
1174 the end of a data transmission (i.e. after ".") Exim was failing to write
1175 a retry record for the message. (Yes, there was some broken host that was
1176 actually sending 8xx at this point.)
117873. An unknown lookup type in a host list could cause Exim to panic-die when
1179 the list was checked. (An example that provoked this was putting <; in the
1180 middle of a list instead of at the start.) If this happened during a DATA
1181 ACL check, a -D file could be left lying around. This kind of configuration
1182 error no longer causes Exim to die; instead it causes a defer errror. The
1183 incident is still logged to the main and panic logs.
118574. Buglet left over from Exim 3 conversion. The message "too many messages
1186 in one connection" was written to the rejectlog but not the mainlog, except
1187 when address rewriting (yes!) was being logged.
118975. Added write_rejectlog option.
119176. When a system filter was run not as root (that is, when system_filter_user
1192 was set), the values of the $n variables were not being returned to the
1193 main process; thus, they were not subsequently available in the $sn
1194 variables.
119677. Added +return_path_on_delivery log selector.
119878. A connection timeout was being treated differently from recipients deferred
1199 when testing hosts_max_try with a message that was older than the host's
1200 retry timeout. (The host should not be counted, thus allowing all hosts to
1201 be tried at least once before bouncing.) This may have been the cause of an
1202 occasionally reported bug whereby a message would remain on the queue
1203 longer than the retry timeout, but would be bounced if a delivery was
1204 forced. I say "may" because I never totally pinned down the problem;
1205 setting up timeout/retry tests is difficult. See also the next item.
120779. The ultimate address timeout was not being applied to errors that involved
1208 a combination of host plus message (for example, a timeout on a MAIL
1209 command). When an address resolved to a number of possible hosts, and they
1210 were not all tried for each delivery (e.g. because of hosts_max_try), a
1211 message could remain on the queue longer than the retry timeout.
121380. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
1214 Haardt.
121681. Fixed an obscure SMTP outgoing bug which required at least the following
1217 conditions: (a) there was another message waiting for the same server;
1218 (b) the server returned 5xx to all RCPT commands in the first message so
1219 that the message was not completed; (c) the server dropped the connection
1220 or gave a negative response to the RSET that Exim sends to abort the
1221 transaction. The observed case was a dropped connection after DATA that had
1222 been sent in pipelining mode. That is, the server had advertised PIPELINING
1223 but was not implementing it correctly. The effect of the bug was incorrect
1224 behaviour, such as trying another host, and this could lead to a crash.
1227Exim version 4.30
1230 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
1231 and daemon.c were passed as pointers to ints; they should have been
1232 pointers to socklen_t variables (which are typically unsigned ints).
1234 2. Some signed/unsigned type warnings in the os.c file for Linux have been
1235 fixed.
1237 3. Fixed a really odd bug that affected only the testing scheme; patching a
1238 certain fixed string in the binary changed the value of another string that
1239 happened to be identical to the end of the original first string.
1241 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
1242 name", it returns that address as the IP address. On some operating
1243 systems (e.g. Solaris), it also passes back the IP address string as the
1244 "host name". However, on others (e.g. Linux), it passes back an empty
1245 string. Exim wasn't checking for this, and was changing the host name to an
1246 empty string, assuming it had been canonicized.
1248 5. Although rare, it is permitted to have more than one PTR record for a given
1249 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
1250 all the names associated with an address, because they do in Solaris.
1251 However, it seems that they do not in Linux for data that comes from the
1252 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
1253 I found this out when I moved to a new Linux workstation and tried to run
1254 the Exim test suite.
1256 To get round this problem I have changed the code so that it now does its
1257 own call to the DNS to look up PTR records when searching for a host name.
1258 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
1259 addresses that are only in /etc/hosts are still found.
1261 This behaviour is, however, controlled by an option called host_lookup_
1262 order, which defaults to "bydns:byaddr". If people want to use the other
1263 order, or indeed, just use one or the other means of lookup, they can
1264 specify it in this variable.
1266 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
1267 some operating systems, this comes back from gethostbyaddr() as an empty
1268 string, and this is what Exim used to test for. However, it seems that in
1269 other systems, "." is yielded. Exim now tests for this case too.
1271 7. The values of check_spool_space and check_log_space are now held internally
1272 as a number of kilobytes instead of an absolute number of bytes. If a
1273 numbers is specified without 'K' or 'M', it is rounded up to the nearest
1274 kilobyte. This means that much larger values can be stored.
1276 8. Exim monitor: an attempt to get the action menu when not actually pointing
1277 at a message produces an empty menu entitled "No message selected". This
1278 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
1279 no entries in it ("Shell widget menu has zero width and/or height"). So I
1280 have added a single, blank menu entry in this case.
1282 9. Added ${quote_local_part.
128410. MIME decoding is now applied to the contents of Subject: header lines when
1285 they are logged.
128711. Now that a reference to $sender_host_address automatically causes a reverse
1288 lookup to occur if necessary (4.13/18), there is no need to arrange for a
1289 host lookup before query-style lookups in lists that might use this
1290 variable. This has therefore been abolished, and the "net-" prefix is no
1291 longer necessary for query-style lookups.
129312. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
1294 have been a typo for LFLAGS, so it has been changed.
129613. The install script calls Exim with "-C /dev/null" in order to find the
1297 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
1298 to be output. Howeve, since Exim outputs its version number before the
1299 error, it didn't break the script. It just looked ugly. I fixed this by
1300 always allowing "-C /dev/null" if the caller is root.
130214. Ignore overlarge ACL variable number when reading spool file - insurance
1303 against a later release with more variables having written the file.
130515. The standard form for an IPv6 address literal was being rejected by EHLO.
1306 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1307 this, as well as the form without the "IPv6" on the front.
130916. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
1310 OS/Makefile-Darwin file.
131217. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
1313 only to LDAP libraries that do not have LDAP_OPT_DEREF.
131518. After change 4.21/52, "%ld" was used to format the contents of the $inode
1316 variable. However, some OS use ints for inodes. I've added cast to long int
1317 to get rid of the compiler warning.
131919. I had forgotten to lock out "/../" in configuration file names when
1320 ALT_CONFIG_PREFIX was set.
132220. Routers used for verification do not need to specify transports. However,
1323 if such a router generated a host list, and callout was configured, Exim
1324 crashed, because it could not find a port number from the (non-existent)
1325 transport. It now assumes port 25 in this circumstance.
132721. Added the -t option to exigrep.
132922. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
1330 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
1331 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
1332 them is set, LOOKUP_LSEARCH is forced.
133423. "exim -bV" now outputs a list of lookups that are included in the binary.
133624. Added sender and host information to the "rejected by local_scan()" log
1337 line; previously there was no indication of these.
133925. Added .include_if_exists.
134126. Change 3.952/11 added an explicit directory sync on top of a file sync for
1342 Linux. It turns out that not all file systems support this. Apparently some
1343 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
1344 do it.) To cope with this, the error EINVAL, which means that sync-ing is
1345 not supported on the file descriptor, is now ignored when Exim is trying to
1346 sync a directory. This applies only to Linux.
134827. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
135028. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
1351 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
1352 this.
135429. Added support for maildirsize files from supplied patch (modified a bit).
135630. The use of :fail: followed by an empty string could lead Exim to respond to
1357 sender verification failures with (e.g.):
1359 550 Verification failed for <xxx>
1360 550 Sender verify failed
1362 where the first response line was missing the '-' that indicates it is not
1363 the final line of the response.
136531. The loop for finding the name of the user that called Exim had a hardwired
1366 limit of 10; it now uses the value of finduser_retries, which is used for
1367 all other user lookups.
136932. Added $received_count variable, available in data and not_smtp ACLs, and at
1370 delivery time.
137233. Exim was neglecting to zero errno before one call of strtol() when
1373 expanding a string and expecting an integer value. On some systems this
1374 resulted in spurious "integer overflow" errors. Also, it was casting the
1375 result into an int without checking.
137734. Testing for a connection timeout using "timeout_connect" in the retry rules
1378 did not work. The code looks as if it has *never* worked, though it appears
1379 to have been documented since at least releast 1.62. I have made it work.
138135. The "timeout_DNS" error in retry rules, also documented since at least
1382 1.62, also never worked. As it isn't clear exactly what this means, and
1383 clearly it isn't a major issue, I have abolished the feature by treating it
1384 as "timeout", and writing a warning to the main and panic logs.
138636. The display of retry rules for -brt wasn't always showing the error code
1387 correctly.
138937. Added new error conditions to retry rules: timeout_A, timeout_MX,
1390 timeout_connect_A, timeout_connect_MX.
139238. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
1393 to the empty sender.
139539. The daemon was not analysing the content of -oX till after it had closed
1396 stderr and disconnected from the controlling terminal. This meant that any
1397 syntax errors were only noted on the panic log, and the return code from
1398 the command was 0. By re-arranging the code a little, I've made the
1399 decoding happen first, so such errors now appear on stderr, and the return
1400 code is 1. However, the actual setting up of the sockets still happens in
1401 the disconnected process, so errors there are still only recorded on the
1402 panic log.
140440. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
1405 connections (as happens on some IP stacks) was logged at start up time as
1406 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
1407 it from "IPv6 and IPv4", which means that two separate sockets are being
1408 used.
141041. The debug output for gethostbyname2() or getipnodebyname() failures now
1411 says whether AF_INET or AF_INET6 was passed as an argument.
141342. Exiwhat output was messed up when time zones were included in log
1414 timestamps.
141643. Exiwhat now gives more information about the daemon's listening ports,
1417 and whether -tls-on-connect was used.
141944. The "port" option of the smtp transport is now expanded.
142145. A "message" modifier in a "warn" statement in a non-message ACL was being
1422 silently ignored. Now an error message is written to the main and panic
1423 logs.
142546. There's a new ACL modifier called "logwrite" which writes to a log file
1426 as soon as it is encountered.
142847. Added $local_user_uid and $local_user_gid at routing time.
143048. Exim crashed when trying to verify a sender address that was being
1431 rewritten to "<>".
143349. Exim was recognizing only a space character after ".include". It now also
1434 recognizes a tab character.
143650. Fixed several bugs in the Perl script that creates the exim.8 man page by
1437 extracting the relevant information from the specification. The man page no
1438 longer contains scrambled data for the -d option, and I've added a section
1439 at the front about calling Exim under different names.
144151. Added "extra_headers" argument to the "mail" command in filter files.
144352. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
1444 crash.
144653. Installed eximstats 1.29.
144854. Added transport_filter_timeout as a generic transport option.
145055. Exim no longer adds an empty Bcc: header to messages that have no To: or
1451 Cc: header lines. This was required by RFC 822, but it not required by RFC
1452 2822.
145456. Exim used to add From:, Date:, and Message-Id: header lines to any
1455 incoming messages that did not have them. Now it does so only if the
1456 message originates locally, that is, if there is no associated remote host
1457 address. When Resent- header lines are present, this applies to the Resent-
1458 lines rather than the non-Resent- lines.
146057. Drop incoming SMTP connection after too many syntax or protocol errors. The
1461 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
146358. Messages for configuration errors now include the name of the main
1464 configuration file - useful now that there may be more than one file in a
1465 list (.included file names were always shown).
146759. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
1468 for those rare installations that do not start the daemon as root or run it
1469 setuid root. I've cut out the call to initgroups() if the daemon is not
1470 root at that time.
147260. The Exim user and group can now be bound into the binary as text strings
1473 that are looked up at the start of Exim's processing.
147561. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
147762. Added $mailstore_basename variable.
147963. Installed patch to sieve.c from Michael Haardt.
148164. When Exim failed to open the panic log after failing to open the main log,
1482 the original message it was trying to log was written to stderr and debug
1483 output, but if they were not available (the usual case in production), it
1484 was lost. Now it is written to syslog before the two lines that record the
1485 failures to open the logs.
148765. Users' Exim filters run in subprocesses under the user's uid. It is
1488 possible for a "deliver" command or an alias in a "personal" command to
1489 provoke an address rewrite. If logging of address rewriting is configured,
1490 this fails because the process is not running as root or exim. There may be
1491 a better way of dealing with this, but for the moment (because 4.30 needs
1492 to be released), I have disabled address rewrite logging when running a
1493 filter in a non-root, non-exim process.
1496Exim version 4.24
1499 1. The buildconfig auxiliary program wasn't quoting the value set for
1500 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
1501 not defined. This bug was masked in 4.22 by the effect that was fixed in
1502 change 4.23/1.
1504 2. Some messages that were rejected after a message id was allocated were
1505 shown as "incomplete" by exigrep. It no longer does this for messages that
1506 are rejected by local_scan() or the DATA or non-SMTP ACLs.
1508 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
1509 have allow_domain_literals set, the ID did not get logged in the <= line.
1510 Domain literals are now always recognized in Message-ID: header lines.
1512 4. The first argument for a ${extract expansion item is the key name or field
1513 number. Leading and trailing spaces in this item were not being ignored,
1514 causing some misleading effects.
1516 5. When deliver_drop_privilege was set, single queue runner processes started
1517 manually (i.e. by the command "exim -q") or by the daemon (which uses the
1518 same command in the process it spins off) were not dropping privilege.
1520 6. When the daemon running as "exim" started a queue runner, it always
1521 re-executed Exim in the spun-off process. This is a waste of effort when
1522 deliver_drop_privilege is set. The new process now just calls the
1523 queue-runner function directly.
1526Exim version 4.23
1529 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
1532 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
1533 ignored. Though the use of -oP was forcing the writing of a pid file, it
1534 was always written to the default place.
1536 3. If the message "no IP address found for host xxxx" is generated during
1537 incoming verification, it is now followed by identification of the incoming
1538 connection (so you can more easily find what provoked it).
1540 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
1542 5. Added some features to "harden" Exim a bit more against certain attacks:
1544 (a) There is now a build-time option called FIXED_NEVER_USERS that can
1545 be put in Local/Makefile. This is like the never_users runtime option,
1546 but it cannot be overridden. The default setting is "root".
1548 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
1549 prefix string with which any file named in a -C command line option
1550 must start.
1552 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
1553 is retained for -C and -D only if the caller of Exim is root. Without
1554 it, the exim user may also use -C and -D and retain privilege.
1556 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
1557 command line option is disabled.
1559 6. Macro names set by the -D option must start with an upper case letter, just
1560 like macro names defined in the configuration file.
1562 7. Added "dereference=" facility to LDAP.
1564 8. Two instances of the typo "uknown" in the source files are fixed.
1566 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
1567 the Configure-Makefile script screwed up while processing it.
156910. Incorporated PCRE 4.4.
157111. The SMTP synchronization check was not operating right at the start of an
1572 SMTP session. For example, it could not catch a HELO sent before the client
1573 waited for the greeting. There is now a check for outstanding input at the
1574 point when the greeting is written. Because of the duplex, asynchronous
1575 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
1576 way, but not yet received, when the check is performed.
157812. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
1579 on TCP/IP sockets, because this apparently causes some broken clients to
1580 timeout.
158213. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
1583 unchanged) from the Cygwin maintainer.
158514. The code for -bV that shows what is in the binary showed "mbx" when maildir
1586 was supported instead of testing for mbx. Effectively a typo.
158815. The spa authenticator server code was not checking that the input it
1589 received was valid base64.
159116. The debug output line for the "set" modifier in ACLs was not showing the
1592 name of the variable that was being set.
159417. Code tidy: the variable type "vtype_string" was never used. Removed it.
159618. Previously, a reference to $sender_host_name did not cause a DNS reverse
1597 lookup on its own. Something else was needed to trigger the lookup. For
1598 example, a match in host_lookup or the need for a host name in a host list.
1599 Now, if $sender_host_name is referenced and the host name has not yet been
1600 looked up, a lookup is performed. If the lookup fails, the variable remains
1601 empty, and $host_lookup_failed is set to "1".
160319. Added "eqi" as a case-independent comparison operator.
160520. The saslauthd authentication condition could segfault if neither service
1606 nor realm was specified.
160821. If an overflowing value such as "2048M" was set for message_size_limit, the
1609 error message that was logged was misleading, and incoming SMTP
1610 connections were dropped. The message is now more accurate, and temporary
1611 errors are given to SMTP connections.
161322. In some error situations (such as 21 above) Exim rejects all SMTP commands
1614 (except RSET) with a 421 error, until QUIT is received. However, it was
1615 failing to send a response to QUIT.
161723. The HELO ACL was being run before the code for helo_try_verify_hosts,
1618 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
1619 ACL is now run after the helo_try_verify_hosts code.
162124. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
1622 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
1623 case-independent, so other case variants are also recognized). Apparently
1624 some systems use these upper case variants.
162625. If more than two messages were waiting for the same host, and a transport
1627 filter was specified for the transport, Exim sent two messages over the
1628 same TCP/IP connection, and then failed with "socket operation on non-
1629 socket" when it tried to send the third.
163126. Added Exim::debug_write and Exim::log_write for embedded Perl use.
163327. The extern definition of crypt16() in expand.c was not being excluded when
1634 the OS had its own crypt16() function.
163628. Added bounce_return_body as a new option, and bounce_return_size_limit
1637 as a preferred synonym for return_size_limit, both as an option and as an
1638 expansion variable.
164029. Added LIBS=-liconv to OS/Makefile-OSF1.
164230. Changed the default configuration ACL to relax the local part checking rule
1643 for addresses that are not in any local domains. For these addresses,
1644 slashes and pipe symbols are allowed within local parts, but the sequence
1645 /../ is explicitly forbidden.
164731. SPA server authentication was not clearing the challenge buffer before
1648 using it.
165032. log_message in a "warn" ACL statement was writing to the reject log as
1651 well as to the main log, which contradicts the documentation and doesn't
1652 seem right (because no rejection is happening). So I have stopped it.
165433. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
1655 However, I am unable to do any testing of this.
165734. Fixed an infelicity in the appendfile transport. When checking directories
1658 for a mailbox, to see if any needed to be created, it was accidentally
1659 using path names with one or more superfluous leading slashes; tracing
1660 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
166235. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
1663 contents are added to the log line that is written for every discarded
1664 recipient. (Previously a log_message setting was ignored.)
166636. The ${quote: operator now quotes the string if it is empty.
166837. The install script runs exim in order to find its version number. If for
1669 some reason other than non-existence or emptiness, which it checks, it
1670 could not run './exim', it was installing it with an empty version number,
1671 i.e. as "exim-". This error state is now caught, and the installation is
1672 aborted.
167438. An argument was missing from the function that creates an error message
1675 when Exim fails to connect to the socket for saslauthd authentication.
1676 This could cause Exim to crash, or give a corrupted message.
167839. Added isip, isip4, and isip6 to ${if conditions.
168040. The ACL variables $acl_xx are now saved with the message, and can be
1681 accessed later in routers, transports, and filters.
168341. The new lookup type nwildlsearch is like wildlsearch, except that the key
1684 strings in the file are not string-expanded.
168642. If a MAIL command specified a SIZE value that was too large to fit into an
1687 int variable, the check against message_size_limit failed. Such values are
1688 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
1689 day this will have to be increased, but I don't think I want to be around
1690 when emails are that large.
1694Exim version 4.22
1697 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
1698 iconv() is not standard in FreeBSD.
1700 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
1701 IPv6 enabled. The observed symptom was a segmentation fault on return from
1702 the function os_common_find_running_interfaces() in src/os.c.
1704 3. In the check_special_case() function in daemon.c I had used "errno" as an
1705 argument name, which causes warnings on some systems. This was basically a
1706 typo, since it was named "eno" in the comments!
1708 4. The code that waits for the clock to tick (at a resolution of some fraction
1709 of a second) so as to ensure message-id uniqueness was always waiting for
1710 at least one whole tick, when it could have waited for less. [This is
1711 almost certainly not relevant at current processor speeds, where it is
1712 unlikely to ever wait at all. But we try to future-proof.]
1714 5. The function that sleeps for a time interval that includes fractions of a
1715 second contained a race. It did not block SIGALRM between setting the
1716 timer, and suspending (a couple of lines later). If the interval was short
1717 and the sigsuspend() was delayed until after it had expired, the suspension
1718 never ended. On busy systems this could lead to processes getting stuck for
1719 ever.
1721 6. Some uncommon configurations may cause a lookup to happen in a queue runner
1722 process, before it forks any delivery processes. The open lookup caching
1723 mechanism meant that the open file or database connection was passed into
1724 the delivery process. The problem was that delivery processes always tidy
1725 up cached lookup data. This could cause a problem for the next delivery
1726 process started by the queue runner, because the external queue runner
1727 process does not know about the closure. So the next delivery process
1728 still has data in the lookup cache. In the case of a file lookup, there was
1729 no problem because closing a file descriptor in a subprocess doesn't affect
1730 the parent. However, if the lookup was caching a connection to a database,
1731 the connection was closed, and the second delivery process was likely to
1732 see errors such as "PGSQL: query failed: server closed the connection
1733 unexpectedly". The problem has been fixed by closing all cached lookups
1734 in a queue runner before running a delivery process.
1736 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
1737 seem to have the "const" qualifier which it has on other OS. I've
1738 parameterised it.
1740 8. Change 4.21/2 was too strict. It is only if there are two authenticators
1741 *of the same type* (client or server) with the same public name that an
1742 error should be diagnosed.
1744 9. When Exim looked up a host name for an IP address, but failed to find the
1745 original IP address when looking up the host name (a safety check), it
1746 output the message "<ip address> does not match any IP for NULL", which was
1747 confusing, to say the least. The bug was that the host name should have
1748 appeared instead of "NULL".
175010. Since release 3.03, if Exim is called by a uid other than root or the Exim
1751 user that is built into the binary, and the -C or -D options is used, root
1752 privilege is dropped before the configuration file is read. In addition,
1753 logging is switched to stderr instead of the normal log files. If the
1754 configuration then re-defines the Exim user, the unprivileged environment
1755 is probably not what is expected, so Exim logs a panic warning message (but
1756 proceeds).
1758 However, if deliver_drop_privilege is set, the unprivileged state may well
1759 be exactly what is intended, so the warning has been cut out in that case,
1760 and Exim is allowed to try to write to its normal log files.
1763Exim version 4.21
1766 1. smtp_return_error_details was not giving details for temporary sender
1767 or receiver verification errors.
1769 2. Diagnose a configuration error if two authenticators have the same public
1770 name.
1772 3. Exim used not to create the message log file for a message until the first
1773 delivery attempt. This could be confusing when incoming messages were held
1774 for policy or load reasons. The message log file is now created at the time
1775 the message is received, and an initial "Received" line is written to it.
1777 4. The automatically generated man page for command line options had a minor
1778 bug that caused no ill effects; however, a more serious problem was that
1779 the procedure for building the man page automatically didn't always
1780 operate. Consequently, release 4.20 contains an out-of-date version. This
1781 shouldn't happen again.
1783 5. When building Exim with embedded Perl support, the script that builds the
1784 Makefile was calling 'perl' to find its compile-time parameters, ignoring
1785 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
1787 6. The freeze_tell option was not being used for messages that were frozen on
1788 arrival, either by an ACL or by local_scan().
1790 7. Added the smtp_incomplete_transaction log selector.
1792 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
1793 was accepting AUTH without a new EHLO.
1795 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
1796 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
1797 RFC.
179910. Logging of TCP/IP connections (when configured) now happens in the main
1800 daemon process instead of the child process, so that the TCP/IP connection
1801 count is more accurate (but it can never be perfect).
180311. The use of "drop" in a nested ACL was not being handled correctly in the
1804 outer ACL. Now, if condition failure induced by the nested "drop" causes
1805 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
1806 or "require"), the connection is dropped.
180812. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
1809 that yield "discard" can now be used with an "accept" or a "discard" verb,
1810 but an error is generated for any others (because I can't see a useful way
1811 to define what should happen).
181313. When an ACL is read dynamically from a file (or anywhere else), the lines
1814 are now processed in the same way as lines in the Exim configuration file.
1815 In particular, continuation lines are supported.
181714. Added the "dnslists = a.b.c!=n.n.n.n" feature.
181915. Added -ti meaning -t -i.
182116. Check for letters, digits, hyphens, and dots in the names of dnslist
1822 domains, and warn by logging if others are found.
182417. At least on BSD, alignment is not guarenteed for the array of ifreq's
1825 returned from GIFCONF when Exim is trying to find the list of interfaces on
1826 a host. The code in os.c has been modified to copy each ifreq to an aligned
1827 structure in all cases.
1829 Also, in some cases, the returned ifreq's were being copied to a 'struct
1830 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
1831 means the last couple of bytes of an IPv6 address could be chopped if the
1832 ifreq contained only a normal sockaddr (14 bytes storage).
183418. Named domain lists were not supported in the hosts_treat_as_local option.
1835 An entry such as +xxxx was not recognized, and was treated as a literal
1836 domain name.
183819. Ensure that header lines added by a DATA ACL are included in the reject log
1839 if the ACL subsequently rejects the message.
184120. Upgrade the utility script to use Digest::MD5 instead of just
1842 MD5 (which is deprecated).
184421. When testing a filter file using -bf, Exim was writing a message when it
1845 took the sender from a "From " line in the message, but it was not doing so
1846 when it took $return_path from a Return-Path: header line. It now does.
184822. If the contents of a "message" modifier for a "warn" ACL verb do not begin
1849 with a valid header line field name (a series of printing characters
1850 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
185223. Changed "disc" in the source to "disk" to conform to the documentation and
1853 the book and for uniformity.
185524. Ignore Sendmail's -Ooption=value command line item.
185725. When execve() failed while trying to run a command in a pipe transport,
1858 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
1859 could be confused with a return value of 69 from the command itself. This
1860 has been changed to 127, the value the shell returns if it is asked to run
1861 a non-existent command. The wording for the related log line suggests a
1862 non-existent command as the problem.
186426. If received_header_text expands to an empty string, do not add a Received:
1865 header line to the message. (Well, it adds a token one on the spool, but
1866 marks it "old" so that it doesn't get used or transmitted.)
186827. Installed eximstats 1.28 (addition of -nt option).
187028. There was no check for failure on the call to getsockname() in the daemon
1871 code. This can fail if there is a shortage of resources on the system, with
1872 ENOMEM, for example. A temporary error is now given on failure.
187429. Contrary to the C standard, it seems that in some environments, the
1875 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
1876 program. Exim now does this explicitly; it affects the formatting of
1877 timestamps using strftime().
187930. If exiqsumm was given junk data, it threw up some uninitialized variable
1880 complaints. I've now initialized all the variables, to avoid this.
188232. Header lines added by a system filter were not being "seen" during
1883 transport-time rewrites.
188533. The info_callback() function passed to OpenSSL is set up with type void
1886 (*)(SSL *, int, int), as described somewhere. However, when calling the
1887 function (actually a macro) that sets it up, the type void(*)() is
1888 expected. I've put in a cast to prevent warnings from picky compilers.
189034. If a DNS black list lookup found a CNAME record, but there were no A
1891 records associated with the domain it pointed at, Exim crashed.
189335. If a DNS black list lookup returned more than one A record, Exim ignored
1894 all but the first. It now scans all returned addresses if a particular IP
1895 value is being sought. In this situation, the contents of the
1896 $dnslist_value variable are a list of all the addresses, separated by a
1897 comma and a space.
189936. Tightened up the rules for host name lookups using reverse DNS. Exim used
1900 to accept a host name and all its aliases if the forward lookup for any of
1901 them yielded the IP address of the incoming connection. Now it accepts only
1902 those names whose forward lookup yields the correct IP address. Any other
1903 names are discarded. This closes a loophole whereby a rogue DNS
1904 administrator could create reverse DNS records to break through a
1905 wildcarded host restriction in an ACL.
190737. If a user filter or a system filter that ran in a subprocess used any of
1908 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
1909 the wrong values were passed to the pipe command ($thisaddress had the
1910 value of $0, $0 had the value of $1, etc). This bug was introduced by
1911 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
191338. Improved the line breaking for long SMTP error messages from ACLs.
1914 Previously, if there was no break point between 40 and 75 characters, Exim
1915 left the rest of the message alone. Two changes have been made: (a) I've
1916 reduced the minimum length to 35 characters; (b) if it can't find a break
1917 point between 35 and 75 characters, it looks ahead and uses the first one
1918 that it finds. This may give the occasional overlong line, but at least the
1919 remaining text gets split now.
192139. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
1922 file descriptors might be low, and that setting 1000 would always raise it.
1923 It turns out that in some environments, the limit is already over 1000 and
1924 that lowering it causes trouble. So now Exim takes care not to decrease it.
192640. When delivering a message, the value of $return_path is set to $sender_
1927 address at the start of routing (routers may change the value). By an
1928 oversight, this default was not being set up when an address was tested by
1929 -bt or -bv, which affected the outcome if any router or filter referred to
1930 $return_path.
193241. The idea of the "warn" ACL verb is that it adds a header or writes to the
1933 log only when "message" or "log_message" are set. However, if one of the
1934 conditions was an address verification, or a call to a nested ACL, the
1935 messages generated by the underlying test were being passed through. This
1936 no longer happens. The underlying message is available in $acl_verify_
1937 message for both "message" and "log_message" expansions, so it can be
1938 passed through if needed.
194042. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
1941 new expansion $bh_ to give the encoded byte string without charset
1942 translation. Translation happens only if iconv() is available; HAVE_ICONV
1943 indicates this at build time. HEADERS_CHARSET gives the charset to
1944 translate to; headers_charset can change it in the configuration, and
1945 "headers charset" can change it in an individual filter file.
194743. Now that we have a default RFC 2047 charset (see above), the code in Exim
1948 that creates RFC 2047 encoded "words" labels them as that charset instead
1949 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
1950 expansion operator; (ii) when Exim creates a From: line for a local
1951 message; (iii) when a header line is rewritten to include a "phrase" part.
195344. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
1954 buggy, causing it to skip the first lines of messages whose message ID
1955 ended in 'D'. This would not have bitten before Exim release 4.14, because
1956 message IDs were unlikely to end in 'D' before then. The effect was to have
1957 incorrect size information for certain domains.
195945. #include "config.h" was missing at the start of the crypt16.c module. This
1960 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
1961 noticed.
196346. If there was a timeout during a "random" callout check, Exim treated it as
1964 a failure of the random address, and carried on sending RSET and the real
1965 address. If the delay was just some slowness somewhere, the response to the
1966 original RCPT would be taken as a response to RSET and so on, causing
1967 mayhem of various kinds.
196947. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
1970 when I implemented it. It didn't allow for the fact that some option values
1971 may legitimatetly be negative (e.g. size_addition), and it didn't even do
1972 the right test for positive values.
197448. Domain names in DNS records are case-independent. Exim always looks them up
1975 in lower case. Some resolvers return domain names in exactly the case they
1976 appear in the zone file, that is, they may contain uppercase letters. Not
1977 all resolvers do this - some return always lower case. Exim was treating a
1978 change of case by a resolver as a change of domain, similar to a widening
1979 of a domain abbreviation. This triggered its re-routing code and so it was
1980 trying to route what was effectively the same domain again. This normally
1981 caused routing to fail (because the router wouldn't handle the domain
1982 twice). Now Exim checks for this case specially, and just changes the
1983 casing of the domain that it ultimately uses when it transmits the message
1984 envelope.
198649. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
1987 module.
198950. If a filter generated a file delivery with a non-absolute name (possible if
1990 no home directory exists for the router), the forbid_file option was not
1991 forbidding it.
199351. Added '&' feature to dnslists, to provide bit mask matching in addition to
1994 the existing equality matching.
199652. Exim was using ints instead of ino_t variables in some places where it was
1997 dealing with inode numbers.
199953. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
2000 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
2001 called TMPDIR, and if it finds it is different, it changes its value.
200354. The smtp_printf() function is now made available to local_scan() so
2004 additional output lines can be written before returning. There is also an
2005 smtp_fflush() function to enable the detection of a dropped connection.
2006 The variables smtp_input and smtp_batched_input are exported to
2007 local_scan().
200955. Changed the default runtime configuration: the message "Unknown user"
2010 has been removed from the ACL, and instead placed on the localuser router,
2011 using the cannot_route_message feature. This means that any verification
2012 failures that generate their own messages won't get overridden. Similarly,
2013 the "Unrouteable address" message that was in the ACL for unverifiable
2014 relay addresses has also been removed.
201656. Added hosts_avoid_esmtp to the smtp transport.
201857. The exicyclog script was not checking for the esoteric option
2019 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2020 will work only if exicyclog is run under the appropriate euid.
202258. Following a discussion on the list, the rules by which Exim recognises line
2023 endings on incoming messages have been changed. The -dropcr and drop_cr
2024 options are now no-ops, retained only for backwards compatibility. The
2025 following line terminators are recognized: LF CRLF CR. However, special
2026 processing applies to CR:
2028 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2029 nor a local message in the state where . is a terminator.
2031 (ii) If a bare CR is encountered in a header line, an extra space is added
2032 after the line terminator so as not to end the header. The reasoning
2033 behind this is that bare CRs in header lines are most likely either
2034 to be mistakes, or people trying to play silly games.
203659. The size of a message, as listed by "-bp" or in the Exim monitor window,
2037 was being incorrectly given as 18 bytes larger than it should have been.
2038 This is a VOB (very old bug).
204060. This may never have affected anything current, but just in case it has:
2041 When the local host is found other than at the start of a list of hosts,
2042 the local host, those with the same MX, and any that follow, are discarded.
2043 When the list in question was part of a longer list of hosts, the following
2044 hosts (not currently being processed) were also being discarded. This no
2045 longer happens. I'm not sure if this situation could ever has previously
2046 arisen.
204861. Added the "/MX" feature to lists of hosts in the manualroute and query
2049 program routers.
205162. Whenever Exim generates a new message, it now adds an Auto-Submitted:
2052 header. This is something that is recommended in a new Internet Draft, and
2053 is something that is documented as being done by Sendmail. There are two
2054 possible values. For messages generated by the autoreply transport, Exim
2055 adds:
2057 Auto-Submitted: auto-replied
2059 whereas for all other generated messages (e.g. bounces) it adds
2061 Auto-Submitted: auto-generated
206363. The "personal" condition in filters now includes a test for the
2064 Auto-Submitted: header. If it contains the string "auto-" the message it
2065 not considered personal.
206764. Added rcpt_include_affixes as a generic transport option.
206965. Added queue_only_override (default true).
207166. Added the syslog_duplication option.
207367. If what should have been the first header line of a message consisted of
2074 a space followed by a colon, Exim was mis-interpreting it as a header line.
2075 It isn't of course - it is syntactically invalid and should therefore be
2076 treated as the start of the message body. The misbehaviour could have
2077 caused a number of strange effects, including loss of data in subsequent
2078 header lines, and spool format errors.
208068. Formerly, the AUTH parameter on a MAIL command was trusted only if the
2081 client host had authenticated. This control can now be exercised by an ACL
2082 for more flexibility.
208469. By default, callouts do not happen when testing with -bh. There is now a
2085 variant, -bhc, which does actually run the callout code, including
2086 consulting and updating the callout cache.
208870. Added support for saslauthd authentication, courtesy of Alexander
2089 Sabourenkov.
209171. If statvfs() failed on the spool or log directories while checking their
2092 size for availability, Exim confusingly gave the error "space shortage".
2093 Furthermore, in debugging mode it crashed with a floating point exception.
2094 These checks are done if check_{spool,log}_{space,inodes} are set, and when
2095 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
2096 serious problem, Exim now writes to the main and panic logs when this
2097 happens, with details of the failure. It then refuses to accept the
2098 incoming message, giving the message "spool directory problem" or "log
2099 directory problem" with a 421 code for SMTP messages.
210172. When Exim is about to re-exec itself, it ensures that the file descriptors
2102 0, 1, and 2 exist, because some OS complain for execs without them (see
2103 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
2104 descriptors. However, the code omitted to check that the open succeeded,
2105 causing mysterious errors if for some reason the permissions on /dev/null
2106 got screwed. Now Exim writes a message to the main and panic logs, and
2107 bombs out if it can't open /dev/null.
210973. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
2110 interact so that it is all more flexible. It is supposed to remain
2111 backwards compatible. Also added extra_local_interfaces.
211374. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
2114 to bomb out with an assertion failure - to the client this appears as a
2115 connection drop. This problem occurs in the part of the code that was taken
2116 from the Samba project. Fortunately, the assertion is in a very simple
2117 function, so I have fixed this by reproducing the function inline in the
2118 one place where it is called, and arranging for authentication to fail
2119 instead of killing the process with assert().
212175. The SPA client code was not working when the server requested OEM rather
2122 than Unicode encoding.
212476. Added code to make require_files with a specific uid setting more usable in
2125 the case where statting the file as root fails - usually a non-root-mounted
2126 NFS file system. When this happens and the failure is EACCES, Exim now
2127 forks a subprocess and does the per-uid checking as the relevant uid.
212977. Added process_log_path.
213178. If log_file_path was not explicitly set, a setting of check_log_space or
2132 check_log_inodes was ignored.
213479. If a space check for the spool or log partitions fails, the incident is now
2135 logged. Of course, in the latter case the data may get lost...
213780. Added the %p formatting code to string_format() so that it can be used to
2138 print addresses in debug_print(). Adjusted all the address printing in the
2139 debugging in store.c to use %p rather than %d.
214181. There was a concern that a line of code in smtp_in.c could overflow a
2142 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
2143 initially expressed, the concern was not well-founded, because trailing
2144 spaces are removed early. However, if the trailing spaces were followed by
2145 a NULL, they did not get removed, so the overflow was possible. Two fixes
2146 were applied:
2148 (a) I re-wrote the offending code in a cleaner fashion.
2149 (b) If an incoming SMTP command contains a NULL character, it is rejected
2150 as invalid.
215282. When Exim changes uid/gid to the Exim user at daemon start time, it now
2153 runs initgroups(), so that if the Exim user is in any additional groups,
2154 they will be used during message reception.
2157Exim version 4.20
2160The change log for 4.20 and earlier releases has been archived.