Additions to dnsdb lookups: (a) list of domains (b) change output
[exim.git] / doc / doc-txt / ChangeLog
7bb56e1f 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.35 2004/11/19 15:18:57 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.
11630. Exim went into a mad DNS loop when attempting to do a callout where the
117 host was specified on an smtp transport, and looking it up yielded more
118 than one IP address.
12031. Re-factored the code for checking spool and log partition space into a
121 function that finds that data and another that does the check. The former
122 is then used to implement four new variables: $spool_space, $log_space,
123 $spool_inodes, and $log_inodes.
12532. The RFC2047 encoding function was originally intended for short strings
126 such as real names; it was not keeping to the 75-character limit for
127 encoded words that the RFC imposes. It now respects the limit, and
128 generates multiple encoded words if necessary. To be on the safe side, I
129 have increased the buffer size for the ${rfc2047: expansion operator from
130 1024 to 2048 bytes.
13233. It is now permitted to omit both strings after an "if" condition; if the
133 condition is true, the result is "true". As before, when the second string
134 is omitted, a false condition yields an empty string. This makes it less
135 cumbersome to write custom ACL and router conditions.
13734. Failure to deliver a bounce message always caused it to be frozen, even if
138 there was an errors_to setting on the router. The errors_to setting is now
139 respected.
14135. If an IPv6 address is given for -bh or -bhc, it is now converted to the
142 canonical form (fully expanded) before being placed in
143 $sender_host_address.
14536. The table in the code that translates DNS record types into text (T_A to
146 "A" for instance) was missing entries for NS and CNAME. It is just possible
147 that this could have caused confusion if both these types were looked up
148 for the same domain, because the text type is used as part of Exim's
149 per-process caching. But the chance of anyone hitting this buglet seems
150 very small.
15237. The dnsdb lookup has been extended in a number of ways.
154 (1) There is a new type, "zns", which walks up the domain tree until it
155 finds some nameserver records. It should be used with care.
157 (2) It is now possible to give a list of domains (or IP addresses) to be
158 looked up.
160 (3) It is now possible to specify the separator character for use when
161 multiple records are returned.
33397d19 162
164Exim version 4.43
167 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
168 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
169 Now it returns an integer. A number of calls were still expecting a T/F
170 return. Fortuitously, in all cases, the tests worked in OK situations,
171 which is the norm. However, things would have gone wrong on any write
172 failures on the smtp file descriptor. This function is used when sending
173 messages over SMTP and also when doing verify callouts.
175 2. When Exim is called to do synchronous delivery of a locally submitted
176 message (the -odf or -odi options), it no longer closes stderr before doing
177 the delivery.
179 3. Implemented the mua_wrapper option.
181 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
183 5. Implemented the functions header_remove(), header_testname(),
184 header_add_at_position(), and receive_remove_recipient(), and exported them
185 to local_scan().
187 6. If an ACL "warn" statement specified the addition of headers, Exim already
188 inserted X-ACL-Warn: at the start if there was no header name. However, it
189 was not making this test for the second and subsequent header lines if
190 there were newlines in the string. This meant that an invalid header could
191 be inserted if Exim was badly configured.
193 7. Allow an ACL "warn" statement to add header lines at the start or after all
194 the Received: headers, as well as at the end.
196 8. Added the rcpt_4xx retry error code.
198 9. Added postmaster_mailfrom=xxx to callout verification option.
20010. Added mailfrom=xxxx to the callout verification option, for verify=
201 header_sender only.
20311. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
204 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
20612. Inserted some casts to stop certain compilers warning when using pointer
207 differences as field lengths or precisions in printf-type calls (mostly
208 affecting debugging statements).
21013. Added optional readline() support for -be (dynamically loaded).
21214. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
213 same clock tick as a message's arrival, so that its received time was the
214 same as the "first fail" time on the retry record, and that message
215 remained on the queue past the ultimate address timeout, every queue runner
216 would try a delivery (because it was past the ultimate address timeout) but
217 after another failure, the ultimate address timeout, which should have then
218 bounced the address, did not kick in. This was a "< instead of <=" error;
219 in most cases the first failure would have been in the next clock tick
220 after the received time, and all would be well.
22215. The special items beginning with @ in domain lists (e.g. @mx_any) were not
223 being recognized when the domain list was tested by the match_domain
224 condition in an expansion string.
22616. Added the ${str2b64: operator.
22817. Exim was always calling setrlimit() to set a large limit for the number of
229 processes, without checking whether the existing limit was already
230 adequate. (It did check for the limit on file descriptors.) Furthermore,
231 errors from getrlimit() and setrlimit() were being ignored. Now they are
232 logged to the main and panic logs, but Exim does carry on, to try to do its
233 job under whatever limits there are.
23518. Imported PCRE 5.0.
23719. Trivial typo in log message " temporarily refused connection" (the leading
238 space).
24020. If the log selector return_path_on_delivery was set and an address was
241 redirected to /dev/null, the delivery process crashed because it assumed
242 that a return path would always be set for a "successful" delivery. In this
243 case, the whole delivery is bypassed as an optimization, and therefore no
244 return path is set.
24621. Internal re-arrangement: the function for sending a challenge and reading
247 a response while authentication was assuming a zero-terminated challenge
248 string. It's now changed to take a pointer and a length, to allow for
249 binary data in such strings.
25122. Added the cyrus_sasl authenticator (code supplied by MBM).
25323. Exim was not respecting finduser_retries when seeking the login of the
254 uid under which it was called; it was always trying 10 times. (The default
255 setting of finduser_retries is zero.) Also, it was sleeping after the final
256 failure, which is pointless.
25824. Implemented tls_on_connect_ports.
26025. Implemented acl_smtp_predata.
26226. If the domain in control=submission is set empty, Exim assumes that the
263 authenticated id is a complete email address when it generates From: or
264 Sender: header lines.
26627. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
267 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
268 chown and chgrp in /bin and hostname in /usr/bin.
27028. Exim was keeping the "process log" file open after each use, just as it
271 does for the main log. This opens the possibility of it remaining open for
272 long periods when the USR1 signal hits a daemon. Occasional processlog
273 errors were reported, that could have been caused by this. Anyway, it seems
274 much more sensible not to leave this file open at all, so that is what now
275 happens.
27729. The long-running daemon process does not normally write to the log once it
278 has entered its main loop, and it closes the log before doing so. This is
279 so that log files can straightforwardly be renamed and moved. However,
280 there are a couple of unusual error situations where the daemon does write
281 log entries, and I had neglected to close the log afterwards.
28330. The text of an SMTP error response that was received during a remote
284 delivery was being truncated at 512 bytes. This is too short for some of
285 the long messages that one sometimes sees. I've increased the limit to
286 1024.
28831. It is now possible to make retry rules that apply only when a message has a
289 specific sender, in particular, an empty sender.
29132. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
292 it possible to be selective about when SMTP synchronization is enforced.
29433. Added "control = caseful_local_part" and "control = "caselower_local_part".
29632. Implemented hosts_connection_nolog.
29833. Added an ACL for QUIT.
30034. Setting "delay_warning=" to disable warnings was not working; it gave a
301 syntax error.
30335. Added mailbox_size and mailbox_filecount to appendfile.
30536. Added control = no_multiline_responses to ACLs.
30737. There was a bug in the logic of the code that waits for the clock to tick
308 in the case where the clock went backwards by a substantial amount such
309 that the microsecond fraction of "now" was more than the microsecond
310 fraction of "then" (but the whole seconds number was less).
31238. Added support for the libradius Radius client library this is found on
313 FreeBSD (previously only the radiusclient library was supported).
316Exim version 4.42
319 1. When certain lookups returned multiple values in the form name=value, the
320 quoting of the values was not always being done properly. Specifically:
321 (a) If the value started with a double quote, but contained no whitespace,
322 it was not quoted.
323 (b) If the value contained whitespace other than a space character (i.e.
324 tabs or newlines or carriage returns) it was not quoted.
325 This fix has been applied to the mysql and pgsql lookups by writing a
326 separate quoting function and calling it from the lookup code. The fix
327 should probably also be applied to nisplus, ibase and oracle lookups, but
328 since I cannot test any of those, I have not disturbed their existing code.
330 2. A hit in the callout cache for a specific address caused a log line with no
331 reason for rejecting RCPT. Now it says "Previous (cached) callout
332 verification failure".
334 3. There was an off-by-one bug in the queryprogram router. An over-long
335 return line was truncated at 256 instead of 255 characters, thereby
336 overflowing its buffer with the terminating zero. As well as fixing this, I
337 have increased the buffer size to 1024 (and made a note to document this).
339 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
340 when Exim is waiting for an SMTP response from a remote server, Exim
341 restarts its select() call on the socket, thereby resetting its timeout.
342 This is not a problem when such interrupts are rare. Somebody set up a cron
343 job to run exiwhat every 2 minutes, which is less than the normal select()
344 timeout (5 or 10 minutes). This meant that the select() timeout never
345 kicked in because it was always reset. I have fixed this by comparing the
346 time when an interrupt arrives with the time at the start of the first call
347 to select(). If more time than the timeout has elapsed, the interrupt is
348 treated as a timeout.
350 5. Some internal re-factoring in preparation for the addition of Sieve
351 extensions (by MH). In particular, the "personal" test is moved to a
352 separate function, and given an option for scanning Cc: and Bcc: (which is
353 not set for Exim filters).
355 6. When Exim created an email address using the login of the caller as the
356 local part (e.g. when creating a From: or Sender: header line), it was not
357 quoting the local part when it contained special characters such as @.
359 7. Installed new OpenBSD configuration files.
361 8. Reworded some messages for syntax errors in "and" and "or" conditions to
362 try to make them clearer.
364 9. Callout options, other than the timeout value, were being ignored when
365 verifying sender addresses in header lines. For example, when using
367 verify = header_sender/callout=no_cache
369 the cache was (incorrectly) being used.
37110. Added a missing instance of ${EXE} to the exim_install script; this affects
372 only the Cygwin environment.
37411. When return_path_on_delivery was set as a log selector, if different remote
375 addresses in the same message used different return paths and parallel
376 remote delivery occurred, the wrong values would sometimes be logged.
377 (Whenever a remote delivery process finished, the return path value from
378 the most recently started remote delivery process was logged.)
38012. RFC 3848 specifies standard names for the "with" phrase in Received: header
381 lines when AUTH and/or TLS are in use. This is the "received protocol"
382 field. Exim used to use "asmtp" for authenticated SMTP, without any
383 indication (in the protocol name) for TLS use. Now it follows the RFC and
384 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
385 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
386 names appear in log lines as well as in Received: header lines.
38813. Installed MH's patches for Sieve to add the "copy" and "vacation"
389 extensions, and comparison tests, and to fix some bugs.
39114. Changes to the "personal" filter test:
393 (1) The test was buggy in that it was just doing the equivalent of
394 "contains" tests on header lines. For example, if a user's address was
395 anne@some.where, the "personal" test would incorrectly be true for
397 To: susanne@some.where
399 This test is now done by extracting each address from the header in turn,
400 and checking the entire address. Other tests that are part of "personal"
401 are now done using regular expressions (for example, to check local parts
402 of addresses in From: header lines).
404 (2) The list of non-personal local parts in From: addresses has been
405 extended to include "listserv", "majordomo", "*-request", and "owner-*",
406 taken from the Sieve specification recommendations.
408 (3) If the message contains any header line starting with "List-" it is
409 treated as non-personal.
411 (4) The test for "circular" in the Subject: header line has been removed
412 because it now seems ill-conceived.
41415. Minor typos in src/EDITME comments corrected.
41616. Installed latest exipick from John Jetmore.
41817. If headers_add on a router specified a text string that was too long for
419 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
420 of string_sprintf() is now avoided.
42218. $message_body_size was not set (it was always zero) when running the DATA
423 ACL and the local_scan() function.
42519. For the "mail" command in an Exim filter, no default was being set for
426 the once_repeat time, causing a random time value to be used if "once" was
427 specified. (If the value happened to be <= 0, no repeat happened.) The
428 default is now 0s, meaning "never repeat". The "vacation" command was OK
429 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
430 (I found it when inspecting the code).
43220. There is now an overall timeout for performing a callout verification. It
433 defaults to 4 times the callout timeout, which applies to individual SMTP
434 commands during the callout. The overall timeout applies when there is more
435 than one host that can be tried. The timeout is checked before trying the
436 next host. This prevents very long delays if there are a large number of
437 hosts and all are timing out (e.g. when the network connections are timing
438 out). The value of the overall timeout can be changed by specifying an
439 additional sub-option for "callout", called "maxwait". For example:
441 verify = sender/callout=5s,maxwait=20s
44321. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
444 to the end before writing, but this should make it even safer).
44622. Exim was forgetting that it had advertised PIPELINING for the second and
447 subsequent messages on an SMTP connection. It was also not resetting its
448 memory on STARTTLS and an internal HELO.
45023. When Exim logs an SMTP synchronization error within a session, it now
451 records whether PIPELINING has been advertised or not.
45324. Added 3 instances of "(long int)" casts to time_t variables that were being
454 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
455 rather than long int.
45725. Installed the latest Cygwin configuration files from the Cygwin maintainer.
45926. Added the never_mail option to autoreply.
462Exim version 4.41
465 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
466 crash if the getsockname() call failed; this can happen if a connection is
467 closed very soon after it is established. The problem was simply in the
468 order in which certain operations were done, causing Exim to try to write
469 to the SMTP stream before it had set up the file descriptor. The bug has
470 been fixed by making things happen in the correct order.
473Exim version 4.40
476 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
477 before the connection was closed, thus losing the rejection response.
479 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
480 some early Solaris releases, but causes trouble in current releases where
481 socklen_t is defined.
483 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
484 always exist.
486 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
487 configured.
489 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
490 that releasing the top of it at the end releases what was used for sub-
491 expansions (unless the block got too big). However, discard this block if
492 the first thing is a variable or header, so that we can use its block when
493 it is dynamic (useful for very large $message_headers, for example).
495 6. Lookups now cache *every* query, not just the most recent. A new, separate
496 store pool is used for this. It can be recovered when all lookup caches are
497 flushed. Lookups now release memory at the end of their result strings.
498 This has involved some general refactoring of the lookup sources.
500 7. Some code has been added to the store_xxx() functions to reduce the amount
501 of flapping under certain conditions.
503 8. log_incoming_interface used to affect only the <= reception log lines. Now
504 it causes the local interface and port to be added to several more SMTP log
505 lines, for example "SMTP connection from", and rejection lines.
507 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
50910. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
51111. If $host_data was set by virtue of a hosts lookup in an ACL, its value
512 could be overwritten at the end of the current message (or the start of a
513 new message if it was set in a HELO ACL). The value is now preserved for
514 the duration of the SMTP connection.
51612. If a transport had a headers_rewrite setting, and a matching header line
517 contained an unqualified address, that address was qualified, even if it
518 did not match any rewriting rules. The underlying bug was that the values
519 of the flags that permit the existence of unqualified sender and recipient
520 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
521 non-local messages, and by -bnq for local messages) were not being
522 preserved with the message after it was received.
52413. When Exim was logging an SMTP synchronization error, it could sometimes log
525 "next input=" as part of the text comprising the host identity instead of
526 the correct text. The code was using the same buffer for two different
527 strings. However, depending on which order the printing function evaluated
528 its arguments, the bug did not always show up. Under Linux, for example, my
529 test suite worked just fine.
53114. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
532 doesn't work with some older versions of Perl. It has been changed to "my",
533 which in any case is probably the better facility to use.
53515. A really picky compiler found some instances of statements for creating
536 error messages that either had too many or two few arguments for the format
537 string.
53916. The size of the buffer for calls to the DNS resolver has been increased
540 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
541 for addresses that have a lot of PTR records. This alleviates a problem; it
542 does not fully solve it.
54417. A dnsdb lookup for PTR records that receives more data than will fit in the
545 buffer now truncates the list and logs the incident, which is the same
546 action as happens when Exim is looking up a host name and its aliases.
547 Previously in this situation something unpredictable would happen;
548 sometimes it was "internal error: store_reset failed".
55018. If a server dropped the connection unexpectedly when an Exim client was
551 using GnuTLS and trying to read a response, the client delivery process
552 crashed while trying to generate an error log message.
55419. If a "warn" verb in an ACL added multiple headers to a message in a single
555 string, for example:
557 warn message = H1: something\nH2: something
559 the text was added as a single header line from Exim's point of view
560 though it ended up OK in the delivered message. However, searching for the
561 second and subsequent header lines using $h_h2: did not work. This has been
562 fixed. Similarly, if a system filter added multiple headers in this way,
563 the routers could not see them.
56520. Expanded the error message when iplsearch is called with an invalid key to
566 suggest using net-iplsearch in a host list.
56821. When running tests using -bh, any delays imposed by "delay" modifiers in
569 ACLs are no longer actually imposed (and a message to that effect is
570 output).
57222. If a "gecos" field in a passwd entry contained escaped characters, in
573 particular, if it contained a \" sequence, Exim got it wrong when building
574 a From: or a Sender: header from that name. A second bug also caused
575 incorrect handling when an unquoted " was present following a character
576 that needed quoting.
57823. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
579 was not being matched caselessly.
58124. Arranged for all hyphens in the exim.8 source to be escaped with
582 backslashes.
58425. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
585 properly. Recipient callout cache records were still being keyed to include
586 the sender, even when use_sender was set false. This led to far more
587 callouts that were necessary. The sender is no longer included in the key
588 when use_sender is false.
59026. Added "control = submission" modifier to ACLs.
59227. Added the ${base62d: operator to decode base 62 numbers.
59428. dnsdb lookups can now access SRV records.
59629. CONFIGURE_OWNER can be set at build time to define an alternative owner for
597 the configuration file.
59930. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
600 (-v) mode. This makes the output for a verbose queue run more intelligible.
60231. Added a use_postmaster feature to recipient callouts.
60432. Added the $body_zerocount variable, containing the number of binary zero
605 bytes in the message body.
60733. The time of last modification of the "new" subdirectory is now used as the
608 "mailbox time last read" when there is a quota error for a maildir
609 delivery.
61134. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
61335. Added +ignore_unknown as a special item in host lists.
61536. Code for decoding IPv6 addresses in host lists is now included, even if
616 IPv6 support is not being compiled. This fixes a bug in which an IPv6
617 address was recognized as an IP address, but was then not correctly decoded
618 into binary, causing unexpected and incorrect effects when compared with
619 another IP address.
622Exim version 4.34
625 1. Very minor rewording of debugging text in manualroute to say "list of
626 hosts" instead of "hostlist".
628 2. If verify=header_syntax was set, and a header line with an unqualified
629 address (no domain) and a large number of spaces between the end of the
630 name and the colon was received, the reception process suffered a buffer
631 overflow, and (when I tested it) crashed. This was caused by some obsolete
632 code that should have been removed. The fix is to remove it!
634 3. When running in the test harness, delay a bit after writing a bounce
635 message to get a bit more predictability in the log output.
637 4. Added a call to search_tidyup() just before forking a reception process. In
638 theory, someone could use a lookup in the expansion of smtp_accept_max_
639 per_host which, without the tidyup, could leave open a database connection.
641 5. Added the variables $recipient_data and $sender_data which get set from a
642 lookup success in an ACL "recipients" or "senders" condition, or a router
643 "senders" option, similar to $domain_data and $local_part_data.
645 6. Moved the writing of debug_print from before to after the "senders" test
646 for routers.
648 7. Change 4.31/66 (moving the time when the Received: is generated) caused
649 problems for message scanning, either using a data ACL, or using
650 local_scan() because the Received: header was not generated till after they
651 were called (in order to set the time as the time of reception completion).
652 I have revised the way this works. The header is now generated after the
653 body is received, but before the ACL or local_scan() are called. After they
654 are run, the timestamp in the header is updated.
657Exim version 4.33
660 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
661 before starting a queue runner without re-exec. This happened only when
662 deliver_drop_privilege was set or when the Exim user was set to root. The
663 effect of the bug was that timeouts during subsequent deliveries caused
664 crashes instead of being properly handled. The handler is now left at its
665 default (and expected) setting.
667 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
668 message, again when deliver_drop_privilege is set or Exim is run as root.
669 The bug described in (1) was not present in this case, but the tidying up
670 of the other signals was missing. I have made the two cases consistent.
672 3. The ignore_target_hosts setting on a manualroute router was being ignored
673 for hosts that were looked up using the /MX notation.
675 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
676 in domain lists.
678 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
679 operated on the sender address. After changing the $sender_address to <>
680 for the sender address verify, Exim was re-instated it as the original
681 (before rewriting) address, but remembering that it had rewritten it, so it
682 wasn't rewriting it again. This bug also had the effect of breaking the
683 sender address verification caching when the sender address was rewritten.
685 6. The ignore_target_hosts option was being ignored by the ipliteral router.
686 This has been changed so that if the ip literal address matches
687 ignore_target_hosts, the router declines.
689 7. Added expansion conditions match_domain, match_address, and match_local_
690 part (NOT match_host).
692 8. The placeholder for the Received: header didn't have a length field set.
694 9. Added code to Exim itself and to exim_lock to test for a specific race
695 condition that could lead to file corruption when using MBX delivery. The
696 issue is with the lockfile that is created in /tmp. If this file is removed
697 after a process has opened it but before that process has acquired a lock,
698 there is the potential for a second process to recreate the file and also
699 acquire a lock. This could lead to two Exim processes writing to the file
700 at the same time. The added code performs the same test as UW imapd; it
701 checks after acquiring the lock that its file descriptor still refers to
702 the same named file.
70410. The buffer for building added header lines was of fixed size, 8192 bytes.
705 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
706 when Exim is built.
70811. Added the smtp_active_hostname option. If used, this will typically be made
709 to depend on the incoming interface address. Because $interface_address is
710 not set up until the daemon has forked a reception process, error responses
711 that can happen earlier (such as "too many connections") no longer contain
712 a host name.
71412. If an expansion in a condition on a "warn" statement fails because a lookup
715 defers, the "warn" statement is abandoned, and the next ACL statement is
716 processed. Previously this caused the whole ACL to be aborted.
71813. Added the iplsearch lookup type.
72014. Added ident_timeout as a log selector.
72215. Added tls_certificate_verified as a log selector.
72416. Added a global option tls_require_ciphers (compare the smtp transport
725 option of the same name). This controls incoming TLS connections.
72717. I finally figured out how to make tls_require_ciphers do a similar thing
728 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
729 before starting the TLS session.
73118. Tabs are now shown as \t in -bP output.
73319. If the log selector return_path_on_delivery was set, Exim crashed when
734 bouncing a message because it had too many Received: header lines.
73620. If two routers both had headers_remove settings, and the first one included
737 a superfluous trailing colon, the final name in the first list and the
738 first name in the second list were incorrectly joined into one item (with a
739 colon in the middle).
742Exim version 4.32
745 1. Added -C and -D options to the exinext utility, mainly to make it easier
746 to include in the automated testing, but these could be helpful when
747 multiple configurations are in use.
749 2. The exinext utility was not formatting the output nicely when there was
750 an alternate port involved in the retry record key, nor when there was a
751 message id as well (for retries that were specific to a specific message
752 and a specific host). It was also confused by IPv6 addresses, because of
753 the additional colons they contain. I have fixed the IPv4 problem, and
754 patched it up to do a reasonable job for IPv6.
756 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
757 delivery, the log line now contains "pipelined" if PIPELINING was used.
759 4. An SMTP transport process used to panic and die if the bind() call to set
760 an explicit outgoing interface failed. This has been changed; it is now
761 treated in the same way as a connect() failure.
763 5. A reference to $sender_host_name in the part of a conditional expansion
764 that was being skipped was still causing a DNS lookup. This no longer
765 occurs.
767 6. The def: expansion condition was not recognizing references to header lines
768 that used bh_ and bheader_.
770 7. Added the _cache feature to named lists.
772 8. The code for checking quota_filecount in the appendfile transport was
773 allowing one more file than it should have been.
775 9. For compatibility with Sendmail, the command line option
777 -prval:sval
779 is equivalent to
781 -oMr rval -oMs sval
783 and sets the incoming protocol and host name (for trusted callers). The
784 host name and its colon can be omitted when only the protocol is to be set.
785 Note the Exim already has two private options, -pd and -ps, that refer to
786 embedded Perl. It is therefore impossible to set a protocol value of "d" or
787 "s", but I don't think that's a major issue.
78910. A number of refactoring changes to the code, none of which should affect
790 Exim's behaviour:
792 (a) The number of logging options was getting close to filling up the
793 32-bit word that was used as a bit map. I have split them into two classes:
794 those that are passed in the argument to log_write(), and those that are
795 only ever tested independently outside of that function. These are now in
796 separate 32-bit words, so there is plenty of room for expansion again.
797 There is no change in the user interface or the logging behaviour.
799 (b) When building, for example, log lines, the code previously used a
800 macro that called string_cat() twice, in order to add two strings. This is
801 not really sufficiently general. Furthermore, there was one instance where
802 it was actually wrong because one of the argument was used twice, and in
803 one call a function was used. (As it happened, calling the function twice
804 did not affect the overall behaviour.) The macro has been replaced by a
805 function that can join an arbitrary number of extra strings onto a growing
806 string.
808 (c) The code for expansion conditions now uses a table and a binary chop
809 instead of a serial search (which was left over from when there were very
810 few conditions). Also, it now recognizes conditions like "pam" even when
811 the relevant support is not compiled in: a suitably worded error message is
812 given if an attempt is made to use such a condition.
81411. Added ${time_interval:xxxxx}.
81612. A bug was causing one of the ddress fields not to be passed back correctly
817 from remote delivery subprocesses. The field in question was not being
818 subsequently used, so this caused to problems in practice.
82013. Added new log selectors queue_time and deliver_time.
82214. Might have fixed a bug in maildirsizefile handling that threw up
823 "unexpected character" debug warnings, and recalculated the data
824 unnecessarily. In any case, I expanded the warning message to give more
825 information.
82715. Added the message "Restricted characters in address" to the statements in
828 the default ACL that block characters like @ and % in local parts.
83016. Change 71 for release 4.31 proved to be much less benign that I imagined.
831 Three changes have been made:
833 (a) There was a serious bug; a negative response to MAIL caused the whole
834 recipient domain to be cached as invalid, thereby blocking all messages
835 to all local parts at the same domain, from all senders. This bug has
836 been fixed. The domain is no longer cached after a negative response to
837 MAIL if the sender used is not empty.
839 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
840 been restored.
842 (c) A new callout option, "use_sender" has been added for people who want
843 the modified behaviour.
846Exim version 4.31
849 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
850 Larry Rosenman.
852 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
853 indeed breaks things for older releases.
855 3. Added additional logging to the case where there is a problem reading data
856 from a filter that is running in a subprocess using a pipe, in order to
857 try to track down a specific problem.
859 4. Testing facility fudge: when running in the test harness and attempting
860 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
861 getting "No route to host". Convert this to a timeout.
863 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
864 warning.
866 6. Some OS don't have socklen_t but use size_t instead. This affects the
867 fifth argument of getsockopt() amongst other things. This is now
868 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
869 can be set for individual OS. I have set it for SunOS5, OSF1, and
870 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
871 some earlier ones do not.
873 7. Change 4.30/15 was not doing the test caselessly.
875 8. The standard form for an IPv6 address literal was being rejected by address
876 parsing in, for example, MAIL and RCPT commands. An example of this kind of
877 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
878 this, as well as the form without the "IPv6" on the front (but only when
879 address literals are enabled, of course).
881 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
88310. Exim crashed if a message with an empty sender address specified by -f
884 encountered a router with an errors_to setting. This could be provoked only
885 by a command such as
887 exim -f "" ...
889 where an empty string was supplied; "<>" did not hit this bug.
89111. Installed PCRE release 4.5.
89312. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
894 remained set. It is now erased.
89613. exiqgrep wasn't working on MacOS X because it didn't correctly compute
897 times from message ids (which are base 36 rather than the normal 62).
89914. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
900 were being counted as actual protocol errors, and logged if the log
901 selector +smtp_protocol_error was set. One cannot be perfect in this test,
902 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
903 and DATA following a set of rejected RCPTs do not count as protocol errors.
904 In other words, Exim assumes they were pipelined, though this may not
905 actually be the case. Of course, in all cases the client gets an
906 appropriate error code.
90815. If a lookup fails in an ACL condition, a message about the failure may
909 be available; it is used if testing the ACL cannot continue, because most
910 such messages specify what the cause of the deferral is. However, some
911 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
912 that caused an old message to be retained and used if a later statement
913 caused a defer, replacing the real cause of the deferral.
91516. If an IP address had so many PTR records that the DNS lookup buffer
916 was not large enough to hold them, Exim could crash while trying to process
917 the truncated data. It now detects and logs this case.
91917. Further to 4.21/58, another change has been made: if (and only if) the
920 first line of a message (the first header line) ends with CRLF, a bare LF
921 in a subsequent header line has a space inserted after it, so as not to
922 terminate the header.
92418. Refactoring: tidied an ugly bit of code in appendfile that copied data
925 unnecessarily, used atoi() instead of strtol(), and didn't check the
926 termination when getting file sizes from file names by regex.
92819. Completely re-implemented the support for maildirsize files, in the light
929 of a number of problems with the previous contributed implementation
930 (4.30/29). In particular:
932 . If the quota is zero, the maildirsize file is maintained, but no quota is
933 imposed.
935 . If the maildir directory does not exist, it is created before any attempt
936 to write a maildirsize file.
938 . The quota value in the file is just a cache; if the quota is changed in
939 the transport, the new value overrides.
941 . A regular expression is available for excluding directories from the
942 count.
94420. The autoreply transport checks the characters in options that define the
945 message's headers; it allows continued headers, but it was checking with
946 isspace() after an embedded newline instead of explicitly looking for a
947 space or a tab.
94921. If all the "regular" hosts to which an address was routed had passed their
950 expiry times, and had not reached their retry times, the address was
951 bounced, even if fallback hosts were defined. Now Exim should go on to try
952 the fallback hosts.
95422. Increased buffer sizes in the callout code from 1024 to 4096 to match the
955 equivalent code in the SMTP transport. Some hosts send humungous responses
956 to HELO/EHLO, more than 1024 it seems.
95823. Refactoring: code in filter.c used (void *) for "any old type" but this
959 gives compiler warnings in some environments. I've now done it "properly",
960 using a union.
96224. The replacement for inet_ntoa() that is used with gcc on IRIX systems
963 (because of problems with the built-in one) was declared to return uschar *
964 instead of char *, causing compiler failure.
96625. Fixed a file descriptor leak when processing alias/forward files.
96826. Fixed a minor format string issue in dbfn.c.
97027. Typo in exim.c: ("dmbnz" for "dbmnz").
97228. If a filter file refered to $h_xxx or $message_headers, and the headers
973 contained RFC 2047 "words", Exim's memory could, under certain conditions,
974 become corrupted.
97629. When a sender address is verified, it is cached, to save repeating the test
977 when there is more than one recipient in a message. However, when the
978 verification involves a callout, it is possible for different callout
979 options to be set for different recipients. It is too complicated to keep
980 track of this in the cache, so now Exim always runs a verification when a
981 callout is required, relying on the callout cache for the optimization.
982 The overhead is duplication of the address routing, but this should not be
983 too great.
98530. Fixed a bug in callout caching. If a RCPT command caused the sender address
986 to be verified with callout=postmaster, and the main callout worked but the
987 postmaster check failed, the verification correctly failed. However, if a
988 subsequent RCPT command asked for sender verification *without* the
989 postmaster check, incorrect caching caused this verification also to fail,
990 incorrectly.
99231. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
993 it was not caching the DNS options (qualify_single, search_parents) that
994 were used when the lookup failed. A subsequent lookup with different
995 options therefore always gave the same answer, though there were cases
996 where it should not have. (Example: a "domains = !$mx_any" option on a
997 dnslookup router: the "domains" option is always processed without any
998 widening, but the router might have qualify_single set.) Now Exim uses the
999 cached value only when the same options are set.
100132. Added John Jetmore's "exipick" utility to the distribution.
100333. GnuTLS: When an attempt to start a TLS session fails for any reason other
1004 than a timeout (e.g. a certificate is required, and is not provided), an
1005 Exim server now closes the connection immediately. Previously it waited for
1006 the client to close - but if the client is SSL, it seems that they each
1007 wait for each other, leading to a delay before one of them times out.
100934: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
1010 maintained 0.8.x compatibility because I don't think many are using it, and
1011 it is clearly obsolete.
101335. Added TLS support for CRLs: a tls_crl global option and one for the smtp
1014 transport.
101636. OpenSSL: $tls_certificate_verified was being set to 1 even if the
1017 client certificate was expired. A simple patch fixes this, though I don't
1018 understand the full logic of why the verify callback is called multiple
1019 times.
102137. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
1022 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
1023 which causes problems with some clients (such as the Certicom SSL Plus
1024 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
1025 disables the coutermeasure allowing Eudora to connect."
102738. Exim was not checking that a write() to a log file succeeded. This could
1028 lead to Bad Things if a log got too big, in particular if it hit a file
1029 size limit. Exim now panics and dies if it cannot write to a log file, just
1030 as it does if it cannot open a log file.
103239. Modified OS/Makefile-Linux so that it now contains
1036 The two -D definitions ensure that Exim is compiled with large file
1037 support, which makes it possible to handle log files that are bigger than
1038 2^31.
104040. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
1041 instance) a domain was checked against a named list that involved a lookup,
1042 causing $domain_data to be set, then another domain was checked against the
1043 same list, then the first domain was re-checked, the value of $domain_data
1044 after the final check could be wrong. In particular, if the second check
1045 failed, it could be set empty. This bug probably also applied to
1046 $localpart_data.
104841. The strip_trailing_dot option was not being applied to the address given
1049 with the -f command-line option.
105142. The code for reading a message's header from the spool was incrementing
1052 $received_count, but never initializing it. This meant that the value was
1053 incorrect (doubled) while delivering a message in the same process in which
1054 it was received. In the most common configuration of Exim, this never
1055 happens - a fresh exec is done - but it can happen when
1056 deliver_drop_privilege is set.
105843. When Exim logs an SMTP synchronization error - client data sent too soon -
1059 it now includes up to 150 characters of the unexpected data in the log
1060 line.
106244. The exim_dbmbuild utility uses fixed size buffers for reading input lines
1063 and building data strings. The size of both of these buffers was 10 000
1064 bytes - far larger than anybody would *ever* want, thought I. Needless to
1065 say, somebody hit the limit. I have increased the maximum line length to
1066 20 000 and the maximum data length of concatenated lines to 100 000. I have
1067 also fixed two bugs, because there was no checking on these buffers. Tsk,
1068 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1069 buffer is too small.
107145. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1072 lsearch lookups. Now it does.
107446. When parsing a route_list item in a manualroute router, a fixed-length
1075 buffer was used for the list of hosts. I made this 1024 bytes long,
1076 thinking that nobody would ever have a list of hosts that long. Wrong.
1077 Somebody had a whole pile of complicated expansion conditions, and the
1078 string was silently truncated, leading to an expansion error. It turns out
1079 that it is easier to change to an unlimited length (owing to other changes
1080 that have happened since this code was originally written) than to build
1081 structure for giving a limitation error. The length of the item that
1082 expands into the list of hosts is now unlimited.
108447. The lsearch lookup could not handle data where the length of text line was
1085 more than 4095 characters. Such lines were truncated, leading to shortened
1086 data being returned. It should now handle lines of any length.
108848. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
1089 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
1090 ACL").
109249. Cosmetic tidy to scripts like exicyclog that are generated by globally
1093 replacing strings such as BIN_DIRECTORY in a source file: the replacement
1094 no longer happens in comment lines. A list of replacements is now placed
1095 at the head of all of the source files, except those whose only change is
1096 to replace PERL_COMMAND in the very first #! line.
109850. Replaced the slow insertion sort in queue.c, for sorting the list of
1099 messages on the queue, with a bottom-up merge sort, using code contributed
1100 by Michael Haardt. This should make operations like -bp somewhat faster on
1101 large queues. It won't affect queue runners, except when queue_run_in_order
1102 is set.
110451. Installed eximstats 1.31 in the distribution.
110652. Added support for SRV lookups to the dnslookup router.
110853. If an ACL referred to $message_body or $message_body_end, the value was not
1109 reset for any messages that followed in the same SMTP session.
111154. The store-handling optimization for building very long strings was not
1112 differentiating between the different store pools. I don't think this
1113 actually made any difference in practice, but I've tidied it.
111555. While running the routers to verify a sender address, $sender_address
1116 was still set to the sender address. This is wrong, because when routing to
1117 send a bounce to the sender, it would be empty. Therefore, I have changed
1118 it so that, while verifying a sender address, $sender_address is set to <>.
1119 (There is no change to what happens when verifying a recipient address.)
112156. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
1122 target A or AAAA records (if not already returned) without resetting the
1123 qualify_single or search_parents options of the DNS resolver. These are
1124 inappropriate in this case because the targets of MX and SRV records must
1125 be FQDNs. A broken DNS record could cause trouble if it happened to have a
1126 target that, when qualified, matched something in the local domain. These
1127 two options are now turned off when doing these lookups.
112957. It seems that at least some releases of Reiserfs (which does not have the
1130 concept of a fixed number of inodes) returns zero and not -1 for the
1131 number of available inodes. This interacted badly with check_spool_inodes,
1132 which assumed that -1 was the "no such thing" setting. What I have done is
1133 to check that the total number of inodes is greater than zero before doing
1134 the test of how many are available.
113658. When a "warn" ACL statement has a log_message modifier, the message is
1137 remembered, and not repeated. This is to avoid a lot of repetition when a
1138 message has many recipients that cause the same warning to be written.
1139 Howewer, Exim was preserving the list of already written lines for an
1140 entire SMTP session, which doesn't seem right. The memory is now reset if a
1141 new message is started.
114359. The "rewrite" debugging flag was not showing the result of rewriting in the
1144 debugging output unless log_rewrite was also set.
114660. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
1147 of (int)(handle) when we know that handle contains (void *)(-1).
114961. The Exim daemon panic-logs an error return when it closes the incoming
1150 connection. However "connection reset by peer" seems to be common, and
1151 isn't really an error worthy of noting specially, so that particular error
1152 is no long logged.
115462. When Exim is trying to find all the local interfaces, it used to panic and
1155 die if the ioctl to get the interface flags failed. However, it seems that
1156 on at least one OS (Solaris 9) it is possible to have an interface that is
1157 included in the list of interfaces, but for which you get a failure error
1158 for this call. This happens when the interface is not "plumbed" into a
1159 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
1160 failure of the "get flags" call assumes that the interface is down.
116263. Added a ${eval10: operator, which assumes all numbers are decimal. This
1163 makes life easier for people who are doing arithmetic on fields extracted
1164 from dates, where you often get leading zeros that should not be
1165 interpreted as octal.
116764. Added qualify_domain to the redirect router, to override the global
1168 setting.
117065. If a pathologically long header line contained very many addresses (the
1171 report of this problem mentioned 10 000) and each of them was rewritten,
1172 Exim could use up a very large amount of memory. (It kept on making new
1173 copies of the header line as it rewrote, and never released the old ones.)
1174 At the expense of a bit more processing, the header rewriting function has
1175 been changed so that it no longer eats memory in this way.
117766. The generation of the Received: header has been moved from the time that a
1178 message starts to be received, to the time that it finishes. The timestamp
1179 in the Received: header should now be very close to that of the <= log
1180 line. There are two side-effects of this change:
1182 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
1183 logged header lines no longer include the local Received: line, because
1184 it has not yet been created. The same applies to a copy of the message
1185 that is returned to a non-SMTP sender when a message is rejected.
1187 (b) When a filter file is tested using -bf, no additional Received: header
1188 is added to the test message. After some thought, I decided that this
1189 is a bug fix.
1191 This change does not affect the value of $received_for. It is still set
1192 after address rewriting, but before local_scan() is called.
119467. Installed the latest Cygwin-specific files from the Cygwin maintainer.
119668. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
1197 gave an unhelpful panic error message, and a defer error. I have managed to
1198 change this behaviour so that it now rejects any supplied certificate,
1199 which seems right, as the list of acceptable certificates is empty.
120169. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
1202 gave an unhelpful defer error. I have not managed to make this reject any
1203 supplied certificates, but the error message it gives is "no certificate
1204 supplied", which is not helpful.
120670. exigrep's output now also includes lines that are not associated with any
1207 message, but which match the given pattern. Implemented by a patch from
1208 Martin Sluka, which also tidied up the Perl a bit.
121071. Recipient callout verification, like sender verification, was using <> in
1211 the MAIL FROM command. This isn't really the right thing, since the actual
1212 sender may affect whether the remote host accepts the recipient or not. I
1213 have changed it to use the actual sender in the callout; this means that
1214 the cache record is now keyed on a recipient/sender pair, not just the
1215 recipient address. There doesn't seem to be a real danger of callout loops,
1216 since a callout by the remote host to check the sender would use <>.
1217 [SEE ABOVE: changed after hitting problems.]
121972. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
1220 temporary errors. However, in the case of such a code being given after
1221 the end of a data transmission (i.e. after ".") Exim was failing to write
1222 a retry record for the message. (Yes, there was some broken host that was
1223 actually sending 8xx at this point.)
122573. An unknown lookup type in a host list could cause Exim to panic-die when
1226 the list was checked. (An example that provoked this was putting <; in the
1227 middle of a list instead of at the start.) If this happened during a DATA
1228 ACL check, a -D file could be left lying around. This kind of configuration
1229 error no longer causes Exim to die; instead it causes a defer errror. The
1230 incident is still logged to the main and panic logs.
123274. Buglet left over from Exim 3 conversion. The message "too many messages
1233 in one connection" was written to the rejectlog but not the mainlog, except
1234 when address rewriting (yes!) was being logged.
123675. Added write_rejectlog option.
123876. When a system filter was run not as root (that is, when system_filter_user
1239 was set), the values of the $n variables were not being returned to the
1240 main process; thus, they were not subsequently available in the $sn
1241 variables.
124377. Added +return_path_on_delivery log selector.
124578. A connection timeout was being treated differently from recipients deferred
1246 when testing hosts_max_try with a message that was older than the host's
1247 retry timeout. (The host should not be counted, thus allowing all hosts to
1248 be tried at least once before bouncing.) This may have been the cause of an
1249 occasionally reported bug whereby a message would remain on the queue
1250 longer than the retry timeout, but would be bounced if a delivery was
1251 forced. I say "may" because I never totally pinned down the problem;
1252 setting up timeout/retry tests is difficult. See also the next item.
125479. The ultimate address timeout was not being applied to errors that involved
1255 a combination of host plus message (for example, a timeout on a MAIL
1256 command). When an address resolved to a number of possible hosts, and they
1257 were not all tried for each delivery (e.g. because of hosts_max_try), a
1258 message could remain on the queue longer than the retry timeout.
126080. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
1261 Haardt.
126381. Fixed an obscure SMTP outgoing bug which required at least the following
1264 conditions: (a) there was another message waiting for the same server;
1265 (b) the server returned 5xx to all RCPT commands in the first message so
1266 that the message was not completed; (c) the server dropped the connection
1267 or gave a negative response to the RSET that Exim sends to abort the
1268 transaction. The observed case was a dropped connection after DATA that had
1269 been sent in pipelining mode. That is, the server had advertised PIPELINING
1270 but was not implementing it correctly. The effect of the bug was incorrect
1271 behaviour, such as trying another host, and this could lead to a crash.
1274Exim version 4.30
1277 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
1278 and daemon.c were passed as pointers to ints; they should have been
1279 pointers to socklen_t variables (which are typically unsigned ints).
1281 2. Some signed/unsigned type warnings in the os.c file for Linux have been
1282 fixed.
1284 3. Fixed a really odd bug that affected only the testing scheme; patching a
1285 certain fixed string in the binary changed the value of another string that
1286 happened to be identical to the end of the original first string.
1288 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
1289 name", it returns that address as the IP address. On some operating
1290 systems (e.g. Solaris), it also passes back the IP address string as the
1291 "host name". However, on others (e.g. Linux), it passes back an empty
1292 string. Exim wasn't checking for this, and was changing the host name to an
1293 empty string, assuming it had been canonicized.
1295 5. Although rare, it is permitted to have more than one PTR record for a given
1296 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
1297 all the names associated with an address, because they do in Solaris.
1298 However, it seems that they do not in Linux for data that comes from the
1299 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
1300 I found this out when I moved to a new Linux workstation and tried to run
1301 the Exim test suite.
1303 To get round this problem I have changed the code so that it now does its
1304 own call to the DNS to look up PTR records when searching for a host name.
1305 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
1306 addresses that are only in /etc/hosts are still found.
1308 This behaviour is, however, controlled by an option called host_lookup_
1309 order, which defaults to "bydns:byaddr". If people want to use the other
1310 order, or indeed, just use one or the other means of lookup, they can
1311 specify it in this variable.
1313 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
1314 some operating systems, this comes back from gethostbyaddr() as an empty
1315 string, and this is what Exim used to test for. However, it seems that in
1316 other systems, "." is yielded. Exim now tests for this case too.
1318 7. The values of check_spool_space and check_log_space are now held internally
1319 as a number of kilobytes instead of an absolute number of bytes. If a
1320 numbers is specified without 'K' or 'M', it is rounded up to the nearest
1321 kilobyte. This means that much larger values can be stored.
1323 8. Exim monitor: an attempt to get the action menu when not actually pointing
1324 at a message produces an empty menu entitled "No message selected". This
1325 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
1326 no entries in it ("Shell widget menu has zero width and/or height"). So I
1327 have added a single, blank menu entry in this case.
1329 9. Added ${quote_local_part.
133110. MIME decoding is now applied to the contents of Subject: header lines when
1332 they are logged.
133411. Now that a reference to $sender_host_address automatically causes a reverse
1335 lookup to occur if necessary (4.13/18), there is no need to arrange for a
1336 host lookup before query-style lookups in lists that might use this
1337 variable. This has therefore been abolished, and the "net-" prefix is no
1338 longer necessary for query-style lookups.
134012. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
1341 have been a typo for LFLAGS, so it has been changed.
134313. The install script calls Exim with "-C /dev/null" in order to find the
1344 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
1345 to be output. Howeve, since Exim outputs its version number before the
1346 error, it didn't break the script. It just looked ugly. I fixed this by
1347 always allowing "-C /dev/null" if the caller is root.
134914. Ignore overlarge ACL variable number when reading spool file - insurance
1350 against a later release with more variables having written the file.
135215. The standard form for an IPv6 address literal was being rejected by EHLO.
1353 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1354 this, as well as the form without the "IPv6" on the front.
135616. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
1357 OS/Makefile-Darwin file.
135917. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
1360 only to LDAP libraries that do not have LDAP_OPT_DEREF.
136218. After change 4.21/52, "%ld" was used to format the contents of the $inode
1363 variable. However, some OS use ints for inodes. I've added cast to long int
1364 to get rid of the compiler warning.
136619. I had forgotten to lock out "/../" in configuration file names when
1367 ALT_CONFIG_PREFIX was set.
136920. Routers used for verification do not need to specify transports. However,
1370 if such a router generated a host list, and callout was configured, Exim
1371 crashed, because it could not find a port number from the (non-existent)
1372 transport. It now assumes port 25 in this circumstance.
137421. Added the -t option to exigrep.
137622. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
1377 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
1378 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
1379 them is set, LOOKUP_LSEARCH is forced.
138123. "exim -bV" now outputs a list of lookups that are included in the binary.
138324. Added sender and host information to the "rejected by local_scan()" log
1384 line; previously there was no indication of these.
138625. Added .include_if_exists.
138826. Change 3.952/11 added an explicit directory sync on top of a file sync for
1389 Linux. It turns out that not all file systems support this. Apparently some
1390 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
1391 do it.) To cope with this, the error EINVAL, which means that sync-ing is
1392 not supported on the file descriptor, is now ignored when Exim is trying to
1393 sync a directory. This applies only to Linux.
139527. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
139728. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
1398 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
1399 this.
140129. Added support for maildirsize files from supplied patch (modified a bit).
140330. The use of :fail: followed by an empty string could lead Exim to respond to
1404 sender verification failures with (e.g.):
1406 550 Verification failed for <xxx>
1407 550 Sender verify failed
1409 where the first response line was missing the '-' that indicates it is not
1410 the final line of the response.
141231. The loop for finding the name of the user that called Exim had a hardwired
1413 limit of 10; it now uses the value of finduser_retries, which is used for
1414 all other user lookups.
141632. Added $received_count variable, available in data and not_smtp ACLs, and at
1417 delivery time.
141933. Exim was neglecting to zero errno before one call of strtol() when
1420 expanding a string and expecting an integer value. On some systems this
1421 resulted in spurious "integer overflow" errors. Also, it was casting the
1422 result into an int without checking.
142434. Testing for a connection timeout using "timeout_connect" in the retry rules
1425 did not work. The code looks as if it has *never* worked, though it appears
1426 to have been documented since at least releast 1.62. I have made it work.
142835. The "timeout_DNS" error in retry rules, also documented since at least
1429 1.62, also never worked. As it isn't clear exactly what this means, and
1430 clearly it isn't a major issue, I have abolished the feature by treating it
1431 as "timeout", and writing a warning to the main and panic logs.
143336. The display of retry rules for -brt wasn't always showing the error code
1434 correctly.
143637. Added new error conditions to retry rules: timeout_A, timeout_MX,
1437 timeout_connect_A, timeout_connect_MX.
143938. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
1440 to the empty sender.
144239. The daemon was not analysing the content of -oX till after it had closed
1443 stderr and disconnected from the controlling terminal. This meant that any
1444 syntax errors were only noted on the panic log, and the return code from
1445 the command was 0. By re-arranging the code a little, I've made the
1446 decoding happen first, so such errors now appear on stderr, and the return
1447 code is 1. However, the actual setting up of the sockets still happens in
1448 the disconnected process, so errors there are still only recorded on the
1449 panic log.
145140. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
1452 connections (as happens on some IP stacks) was logged at start up time as
1453 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
1454 it from "IPv6 and IPv4", which means that two separate sockets are being
1455 used.
145741. The debug output for gethostbyname2() or getipnodebyname() failures now
1458 says whether AF_INET or AF_INET6 was passed as an argument.
146042. Exiwhat output was messed up when time zones were included in log
1461 timestamps.
146343. Exiwhat now gives more information about the daemon's listening ports,
1464 and whether -tls-on-connect was used.
146644. The "port" option of the smtp transport is now expanded.
146845. A "message" modifier in a "warn" statement in a non-message ACL was being
1469 silently ignored. Now an error message is written to the main and panic
1470 logs.
147246. There's a new ACL modifier called "logwrite" which writes to a log file
1473 as soon as it is encountered.
147547. Added $local_user_uid and $local_user_gid at routing time.
147748. Exim crashed when trying to verify a sender address that was being
1478 rewritten to "<>".
148049. Exim was recognizing only a space character after ".include". It now also
1481 recognizes a tab character.
148350. Fixed several bugs in the Perl script that creates the exim.8 man page by
1484 extracting the relevant information from the specification. The man page no
1485 longer contains scrambled data for the -d option, and I've added a section
1486 at the front about calling Exim under different names.
148851. Added "extra_headers" argument to the "mail" command in filter files.
149052. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
1491 crash.
149353. Installed eximstats 1.29.
149554. Added transport_filter_timeout as a generic transport option.
149755. Exim no longer adds an empty Bcc: header to messages that have no To: or
1498 Cc: header lines. This was required by RFC 822, but it not required by RFC
1499 2822.
150156. Exim used to add From:, Date:, and Message-Id: header lines to any
1502 incoming messages that did not have them. Now it does so only if the
1503 message originates locally, that is, if there is no associated remote host
1504 address. When Resent- header lines are present, this applies to the Resent-
1505 lines rather than the non-Resent- lines.
150757. Drop incoming SMTP connection after too many syntax or protocol errors. The
1508 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
151058. Messages for configuration errors now include the name of the main
1511 configuration file - useful now that there may be more than one file in a
1512 list (.included file names were always shown).
151459. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
1515 for those rare installations that do not start the daemon as root or run it
1516 setuid root. I've cut out the call to initgroups() if the daemon is not
1517 root at that time.
151960. The Exim user and group can now be bound into the binary as text strings
1520 that are looked up at the start of Exim's processing.
152261. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
152462. Added $mailstore_basename variable.
152663. Installed patch to sieve.c from Michael Haardt.
152864. When Exim failed to open the panic log after failing to open the main log,
1529 the original message it was trying to log was written to stderr and debug
1530 output, but if they were not available (the usual case in production), it
1531 was lost. Now it is written to syslog before the two lines that record the
1532 failures to open the logs.
153465. Users' Exim filters run in subprocesses under the user's uid. It is
1535 possible for a "deliver" command or an alias in a "personal" command to
1536 provoke an address rewrite. If logging of address rewriting is configured,
1537 this fails because the process is not running as root or exim. There may be
1538 a better way of dealing with this, but for the moment (because 4.30 needs
1539 to be released), I have disabled address rewrite logging when running a
1540 filter in a non-root, non-exim process.
1543Exim version 4.24
1546 1. The buildconfig auxiliary program wasn't quoting the value set for
1547 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
1548 not defined. This bug was masked in 4.22 by the effect that was fixed in
1549 change 4.23/1.
1551 2. Some messages that were rejected after a message id was allocated were
1552 shown as "incomplete" by exigrep. It no longer does this for messages that
1553 are rejected by local_scan() or the DATA or non-SMTP ACLs.
1555 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
1556 have allow_domain_literals set, the ID did not get logged in the <= line.
1557 Domain literals are now always recognized in Message-ID: header lines.
1559 4. The first argument for a ${extract expansion item is the key name or field
1560 number. Leading and trailing spaces in this item were not being ignored,
1561 causing some misleading effects.
1563 5. When deliver_drop_privilege was set, single queue runner processes started
1564 manually (i.e. by the command "exim -q") or by the daemon (which uses the
1565 same command in the process it spins off) were not dropping privilege.
1567 6. When the daemon running as "exim" started a queue runner, it always
1568 re-executed Exim in the spun-off process. This is a waste of effort when
1569 deliver_drop_privilege is set. The new process now just calls the
1570 queue-runner function directly.
1573Exim version 4.23
1576 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
1579 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
1580 ignored. Though the use of -oP was forcing the writing of a pid file, it
1581 was always written to the default place.
1583 3. If the message "no IP address found for host xxxx" is generated during
1584 incoming verification, it is now followed by identification of the incoming
1585 connection (so you can more easily find what provoked it).
1587 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
1589 5. Added some features to "harden" Exim a bit more against certain attacks:
1591 (a) There is now a build-time option called FIXED_NEVER_USERS that can
1592 be put in Local/Makefile. This is like the never_users runtime option,
1593 but it cannot be overridden. The default setting is "root".
1595 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
1596 prefix string with which any file named in a -C command line option
1597 must start.
1599 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
1600 is retained for -C and -D only if the caller of Exim is root. Without
1601 it, the exim user may also use -C and -D and retain privilege.
1603 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
1604 command line option is disabled.
1606 6. Macro names set by the -D option must start with an upper case letter, just
1607 like macro names defined in the configuration file.
1609 7. Added "dereference=" facility to LDAP.
1611 8. Two instances of the typo "uknown" in the source files are fixed.
1613 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
1614 the Configure-Makefile script screwed up while processing it.
161610. Incorporated PCRE 4.4.
161811. The SMTP synchronization check was not operating right at the start of an
1619 SMTP session. For example, it could not catch a HELO sent before the client
1620 waited for the greeting. There is now a check for outstanding input at the
1621 point when the greeting is written. Because of the duplex, asynchronous
1622 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
1623 way, but not yet received, when the check is performed.
162512. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
1626 on TCP/IP sockets, because this apparently causes some broken clients to
1627 timeout.
162913. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
1630 unchanged) from the Cygwin maintainer.
163214. The code for -bV that shows what is in the binary showed "mbx" when maildir
1633 was supported instead of testing for mbx. Effectively a typo.
163515. The spa authenticator server code was not checking that the input it
1636 received was valid base64.
163816. The debug output line for the "set" modifier in ACLs was not showing the
1639 name of the variable that was being set.
164117. Code tidy: the variable type "vtype_string" was never used. Removed it.
164318. Previously, a reference to $sender_host_name did not cause a DNS reverse
1644 lookup on its own. Something else was needed to trigger the lookup. For
1645 example, a match in host_lookup or the need for a host name in a host list.
1646 Now, if $sender_host_name is referenced and the host name has not yet been
1647 looked up, a lookup is performed. If the lookup fails, the variable remains
1648 empty, and $host_lookup_failed is set to "1".
165019. Added "eqi" as a case-independent comparison operator.
165220. The saslauthd authentication condition could segfault if neither service
1653 nor realm was specified.
165521. If an overflowing value such as "2048M" was set for message_size_limit, the
1656 error message that was logged was misleading, and incoming SMTP
1657 connections were dropped. The message is now more accurate, and temporary
1658 errors are given to SMTP connections.
166022. In some error situations (such as 21 above) Exim rejects all SMTP commands
1661 (except RSET) with a 421 error, until QUIT is received. However, it was
1662 failing to send a response to QUIT.
166423. The HELO ACL was being run before the code for helo_try_verify_hosts,
1665 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
1666 ACL is now run after the helo_try_verify_hosts code.
166824. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
1669 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
1670 case-independent, so other case variants are also recognized). Apparently
1671 some systems use these upper case variants.
167325. If more than two messages were waiting for the same host, and a transport
1674 filter was specified for the transport, Exim sent two messages over the
1675 same TCP/IP connection, and then failed with "socket operation on non-
1676 socket" when it tried to send the third.
167826. Added Exim::debug_write and Exim::log_write for embedded Perl use.
168027. The extern definition of crypt16() in expand.c was not being excluded when
1681 the OS had its own crypt16() function.
168328. Added bounce_return_body as a new option, and bounce_return_size_limit
1684 as a preferred synonym for return_size_limit, both as an option and as an
1685 expansion variable.
168729. Added LIBS=-liconv to OS/Makefile-OSF1.
168930. Changed the default configuration ACL to relax the local part checking rule
1690 for addresses that are not in any local domains. For these addresses,
1691 slashes and pipe symbols are allowed within local parts, but the sequence
1692 /../ is explicitly forbidden.
169431. SPA server authentication was not clearing the challenge buffer before
1695 using it.
169732. log_message in a "warn" ACL statement was writing to the reject log as
1698 well as to the main log, which contradicts the documentation and doesn't
1699 seem right (because no rejection is happening). So I have stopped it.
170133. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
1702 However, I am unable to do any testing of this.
170434. Fixed an infelicity in the appendfile transport. When checking directories
1705 for a mailbox, to see if any needed to be created, it was accidentally
1706 using path names with one or more superfluous leading slashes; tracing
1707 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
170935. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
1710 contents are added to the log line that is written for every discarded
1711 recipient. (Previously a log_message setting was ignored.)
171336. The ${quote: operator now quotes the string if it is empty.
171537. The install script runs exim in order to find its version number. If for
1716 some reason other than non-existence or emptiness, which it checks, it
1717 could not run './exim', it was installing it with an empty version number,
1718 i.e. as "exim-". This error state is now caught, and the installation is
1719 aborted.
172138. An argument was missing from the function that creates an error message
1722 when Exim fails to connect to the socket for saslauthd authentication.
1723 This could cause Exim to crash, or give a corrupted message.
172539. Added isip, isip4, and isip6 to ${if conditions.
172740. The ACL variables $acl_xx are now saved with the message, and can be
1728 accessed later in routers, transports, and filters.
173041. The new lookup type nwildlsearch is like wildlsearch, except that the key
1731 strings in the file are not string-expanded.
173342. If a MAIL command specified a SIZE value that was too large to fit into an
1734 int variable, the check against message_size_limit failed. Such values are
1735 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
1736 day this will have to be increased, but I don't think I want to be around
1737 when emails are that large.
1741Exim version 4.22
1744 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
1745 iconv() is not standard in FreeBSD.
1747 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
1748 IPv6 enabled. The observed symptom was a segmentation fault on return from
1749 the function os_common_find_running_interfaces() in src/os.c.
1751 3. In the check_special_case() function in daemon.c I had used "errno" as an
1752 argument name, which causes warnings on some systems. This was basically a
1753 typo, since it was named "eno" in the comments!
1755 4. The code that waits for the clock to tick (at a resolution of some fraction
1756 of a second) so as to ensure message-id uniqueness was always waiting for
1757 at least one whole tick, when it could have waited for less. [This is
1758 almost certainly not relevant at current processor speeds, where it is
1759 unlikely to ever wait at all. But we try to future-proof.]
1761 5. The function that sleeps for a time interval that includes fractions of a
1762 second contained a race. It did not block SIGALRM between setting the
1763 timer, and suspending (a couple of lines later). If the interval was short
1764 and the sigsuspend() was delayed until after it had expired, the suspension
1765 never ended. On busy systems this could lead to processes getting stuck for
1766 ever.
1768 6. Some uncommon configurations may cause a lookup to happen in a queue runner
1769 process, before it forks any delivery processes. The open lookup caching
1770 mechanism meant that the open file or database connection was passed into
1771 the delivery process. The problem was that delivery processes always tidy
1772 up cached lookup data. This could cause a problem for the next delivery
1773 process started by the queue runner, because the external queue runner
1774 process does not know about the closure. So the next delivery process
1775 still has data in the lookup cache. In the case of a file lookup, there was
1776 no problem because closing a file descriptor in a subprocess doesn't affect
1777 the parent. However, if the lookup was caching a connection to a database,
1778 the connection was closed, and the second delivery process was likely to
1779 see errors such as "PGSQL: query failed: server closed the connection
1780 unexpectedly". The problem has been fixed by closing all cached lookups
1781 in a queue runner before running a delivery process.
1783 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
1784 seem to have the "const" qualifier which it has on other OS. I've
1785 parameterised it.
1787 8. Change 4.21/2 was too strict. It is only if there are two authenticators
1788 *of the same type* (client or server) with the same public name that an
1789 error should be diagnosed.
1791 9. When Exim looked up a host name for an IP address, but failed to find the
1792 original IP address when looking up the host name (a safety check), it
1793 output the message "<ip address> does not match any IP for NULL", which was
1794 confusing, to say the least. The bug was that the host name should have
1795 appeared instead of "NULL".
179710. Since release 3.03, if Exim is called by a uid other than root or the Exim
1798 user that is built into the binary, and the -C or -D options is used, root
1799 privilege is dropped before the configuration file is read. In addition,
1800 logging is switched to stderr instead of the normal log files. If the
1801 configuration then re-defines the Exim user, the unprivileged environment
1802 is probably not what is expected, so Exim logs a panic warning message (but
1803 proceeds).
1805 However, if deliver_drop_privilege is set, the unprivileged state may well
1806 be exactly what is intended, so the warning has been cut out in that case,
1807 and Exim is allowed to try to write to its normal log files.
1810Exim version 4.21
1813 1. smtp_return_error_details was not giving details for temporary sender
1814 or receiver verification errors.
1816 2. Diagnose a configuration error if two authenticators have the same public
1817 name.
1819 3. Exim used not to create the message log file for a message until the first
1820 delivery attempt. This could be confusing when incoming messages were held
1821 for policy or load reasons. The message log file is now created at the time
1822 the message is received, and an initial "Received" line is written to it.
1824 4. The automatically generated man page for command line options had a minor
1825 bug that caused no ill effects; however, a more serious problem was that
1826 the procedure for building the man page automatically didn't always
1827 operate. Consequently, release 4.20 contains an out-of-date version. This
1828 shouldn't happen again.
1830 5. When building Exim with embedded Perl support, the script that builds the
1831 Makefile was calling 'perl' to find its compile-time parameters, ignoring
1832 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
1834 6. The freeze_tell option was not being used for messages that were frozen on
1835 arrival, either by an ACL or by local_scan().
1837 7. Added the smtp_incomplete_transaction log selector.
1839 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
1840 was accepting AUTH without a new EHLO.
1842 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
1843 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
1844 RFC.
184610. Logging of TCP/IP connections (when configured) now happens in the main
1847 daemon process instead of the child process, so that the TCP/IP connection
1848 count is more accurate (but it can never be perfect).
185011. The use of "drop" in a nested ACL was not being handled correctly in the
1851 outer ACL. Now, if condition failure induced by the nested "drop" causes
1852 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
1853 or "require"), the connection is dropped.
185512. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
1856 that yield "discard" can now be used with an "accept" or a "discard" verb,
1857 but an error is generated for any others (because I can't see a useful way
1858 to define what should happen).
186013. When an ACL is read dynamically from a file (or anywhere else), the lines
1861 are now processed in the same way as lines in the Exim configuration file.
1862 In particular, continuation lines are supported.
186414. Added the "dnslists = a.b.c!=n.n.n.n" feature.
186615. Added -ti meaning -t -i.
186816. Check for letters, digits, hyphens, and dots in the names of dnslist
1869 domains, and warn by logging if others are found.
187117. At least on BSD, alignment is not guarenteed for the array of ifreq's
1872 returned from GIFCONF when Exim is trying to find the list of interfaces on
1873 a host. The code in os.c has been modified to copy each ifreq to an aligned
1874 structure in all cases.
1876 Also, in some cases, the returned ifreq's were being copied to a 'struct
1877 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
1878 means the last couple of bytes of an IPv6 address could be chopped if the
1879 ifreq contained only a normal sockaddr (14 bytes storage).
188118. Named domain lists were not supported in the hosts_treat_as_local option.
1882 An entry such as +xxxx was not recognized, and was treated as a literal
1883 domain name.
188519. Ensure that header lines added by a DATA ACL are included in the reject log
1886 if the ACL subsequently rejects the message.
188820. Upgrade the utility script to use Digest::MD5 instead of just
1889 MD5 (which is deprecated).
189121. When testing a filter file using -bf, Exim was writing a message when it
1892 took the sender from a "From " line in the message, but it was not doing so
1893 when it took $return_path from a Return-Path: header line. It now does.
189522. If the contents of a "message" modifier for a "warn" ACL verb do not begin
1896 with a valid header line field name (a series of printing characters
1897 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
189923. Changed "disc" in the source to "disk" to conform to the documentation and
1900 the book and for uniformity.
190224. Ignore Sendmail's -Ooption=value command line item.
190425. When execve() failed while trying to run a command in a pipe transport,
1905 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
1906 could be confused with a return value of 69 from the command itself. This
1907 has been changed to 127, the value the shell returns if it is asked to run
1908 a non-existent command. The wording for the related log line suggests a
1909 non-existent command as the problem.
191126. If received_header_text expands to an empty string, do not add a Received:
1912 header line to the message. (Well, it adds a token one on the spool, but
1913 marks it "old" so that it doesn't get used or transmitted.)
191527. Installed eximstats 1.28 (addition of -nt option).
191728. There was no check for failure on the call to getsockname() in the daemon
1918 code. This can fail if there is a shortage of resources on the system, with
1919 ENOMEM, for example. A temporary error is now given on failure.
192129. Contrary to the C standard, it seems that in some environments, the
1922 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
1923 program. Exim now does this explicitly; it affects the formatting of
1924 timestamps using strftime().
192630. If exiqsumm was given junk data, it threw up some uninitialized variable
1927 complaints. I've now initialized all the variables, to avoid this.
192932. Header lines added by a system filter were not being "seen" during
1930 transport-time rewrites.
193233. The info_callback() function passed to OpenSSL is set up with type void
1933 (*)(SSL *, int, int), as described somewhere. However, when calling the
1934 function (actually a macro) that sets it up, the type void(*)() is
1935 expected. I've put in a cast to prevent warnings from picky compilers.
193734. If a DNS black list lookup found a CNAME record, but there were no A
1938 records associated with the domain it pointed at, Exim crashed.
194035. If a DNS black list lookup returned more than one A record, Exim ignored
1941 all but the first. It now scans all returned addresses if a particular IP
1942 value is being sought. In this situation, the contents of the
1943 $dnslist_value variable are a list of all the addresses, separated by a
1944 comma and a space.
194636. Tightened up the rules for host name lookups using reverse DNS. Exim used
1947 to accept a host name and all its aliases if the forward lookup for any of
1948 them yielded the IP address of the incoming connection. Now it accepts only
1949 those names whose forward lookup yields the correct IP address. Any other
1950 names are discarded. This closes a loophole whereby a rogue DNS
1951 administrator could create reverse DNS records to break through a
1952 wildcarded host restriction in an ACL.
195437. If a user filter or a system filter that ran in a subprocess used any of
1955 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
1956 the wrong values were passed to the pipe command ($thisaddress had the
1957 value of $0, $0 had the value of $1, etc). This bug was introduced by
1958 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
196038. Improved the line breaking for long SMTP error messages from ACLs.
1961 Previously, if there was no break point between 40 and 75 characters, Exim
1962 left the rest of the message alone. Two changes have been made: (a) I've
1963 reduced the minimum length to 35 characters; (b) if it can't find a break
1964 point between 35 and 75 characters, it looks ahead and uses the first one
1965 that it finds. This may give the occasional overlong line, but at least the
1966 remaining text gets split now.
196839. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
1969 file descriptors might be low, and that setting 1000 would always raise it.
1970 It turns out that in some environments, the limit is already over 1000 and
1971 that lowering it causes trouble. So now Exim takes care not to decrease it.
197340. When delivering a message, the value of $return_path is set to $sender_
1974 address at the start of routing (routers may change the value). By an
1975 oversight, this default was not being set up when an address was tested by
1976 -bt or -bv, which affected the outcome if any router or filter referred to
1977 $return_path.
197941. The idea of the "warn" ACL verb is that it adds a header or writes to the
1980 log only when "message" or "log_message" are set. However, if one of the
1981 conditions was an address verification, or a call to a nested ACL, the
1982 messages generated by the underlying test were being passed through. This
1983 no longer happens. The underlying message is available in $acl_verify_
1984 message for both "message" and "log_message" expansions, so it can be
1985 passed through if needed.
198742. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
1988 new expansion $bh_ to give the encoded byte string without charset
1989 translation. Translation happens only if iconv() is available; HAVE_ICONV
1990 indicates this at build time. HEADERS_CHARSET gives the charset to
1991 translate to; headers_charset can change it in the configuration, and
1992 "headers charset" can change it in an individual filter file.
199443. Now that we have a default RFC 2047 charset (see above), the code in Exim
1995 that creates RFC 2047 encoded "words" labels them as that charset instead
1996 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
1997 expansion operator; (ii) when Exim creates a From: line for a local
1998 message; (iii) when a header line is rewritten to include a "phrase" part.
200044. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
2001 buggy, causing it to skip the first lines of messages whose message ID
2002 ended in 'D'. This would not have bitten before Exim release 4.14, because
2003 message IDs were unlikely to end in 'D' before then. The effect was to have
2004 incorrect size information for certain domains.
200645. #include "config.h" was missing at the start of the crypt16.c module. This
2007 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
2008 noticed.
201046. If there was a timeout during a "random" callout check, Exim treated it as
2011 a failure of the random address, and carried on sending RSET and the real
2012 address. If the delay was just some slowness somewhere, the response to the
2013 original RCPT would be taken as a response to RSET and so on, causing
2014 mayhem of various kinds.
201647. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
2017 when I implemented it. It didn't allow for the fact that some option values
2018 may legitimatetly be negative (e.g. size_addition), and it didn't even do
2019 the right test for positive values.
202148. Domain names in DNS records are case-independent. Exim always looks them up
2022 in lower case. Some resolvers return domain names in exactly the case they
2023 appear in the zone file, that is, they may contain uppercase letters. Not
2024 all resolvers do this - some return always lower case. Exim was treating a
2025 change of case by a resolver as a change of domain, similar to a widening
2026 of a domain abbreviation. This triggered its re-routing code and so it was
2027 trying to route what was effectively the same domain again. This normally
2028 caused routing to fail (because the router wouldn't handle the domain
2029 twice). Now Exim checks for this case specially, and just changes the
2030 casing of the domain that it ultimately uses when it transmits the message
2031 envelope.
203349. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
2034 module.
203650. If a filter generated a file delivery with a non-absolute name (possible if
2037 no home directory exists for the router), the forbid_file option was not
2038 forbidding it.
204051. Added '&' feature to dnslists, to provide bit mask matching in addition to
2041 the existing equality matching.
204352. Exim was using ints instead of ino_t variables in some places where it was
2044 dealing with inode numbers.
204653. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
2047 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
2048 called TMPDIR, and if it finds it is different, it changes its value.
205054. The smtp_printf() function is now made available to local_scan() so
2051 additional output lines can be written before returning. There is also an
2052 smtp_fflush() function to enable the detection of a dropped connection.
2053 The variables smtp_input and smtp_batched_input are exported to
2054 local_scan().
205655. Changed the default runtime configuration: the message "Unknown user"
2057 has been removed from the ACL, and instead placed on the localuser router,
2058 using the cannot_route_message feature. This means that any verification
2059 failures that generate their own messages won't get overridden. Similarly,
2060 the "Unrouteable address" message that was in the ACL for unverifiable
2061 relay addresses has also been removed.
206356. Added hosts_avoid_esmtp to the smtp transport.
206557. The exicyclog script was not checking for the esoteric option
2066 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2067 will work only if exicyclog is run under the appropriate euid.
206958. Following a discussion on the list, the rules by which Exim recognises line
2070 endings on incoming messages have been changed. The -dropcr and drop_cr
2071 options are now no-ops, retained only for backwards compatibility. The
2072 following line terminators are recognized: LF CRLF CR. However, special
2073 processing applies to CR:
2075 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2076 nor a local message in the state where . is a terminator.
2078 (ii) If a bare CR is encountered in a header line, an extra space is added
2079 after the line terminator so as not to end the header. The reasoning
2080 behind this is that bare CRs in header lines are most likely either
2081 to be mistakes, or people trying to play silly games.
208359. The size of a message, as listed by "-bp" or in the Exim monitor window,
2084 was being incorrectly given as 18 bytes larger than it should have been.
2085 This is a VOB (very old bug).
208760. This may never have affected anything current, but just in case it has:
2088 When the local host is found other than at the start of a list of hosts,
2089 the local host, those with the same MX, and any that follow, are discarded.
2090 When the list in question was part of a longer list of hosts, the following
2091 hosts (not currently being processed) were also being discarded. This no
2092 longer happens. I'm not sure if this situation could ever has previously
2093 arisen.
209561. Added the "/MX" feature to lists of hosts in the manualroute and query
2096 program routers.
209862. Whenever Exim generates a new message, it now adds an Auto-Submitted:
2099 header. This is something that is recommended in a new Internet Draft, and
2100 is something that is documented as being done by Sendmail. There are two
2101 possible values. For messages generated by the autoreply transport, Exim
2102 adds:
2104 Auto-Submitted: auto-replied
2106 whereas for all other generated messages (e.g. bounces) it adds
2108 Auto-Submitted: auto-generated
211063. The "personal" condition in filters now includes a test for the
2111 Auto-Submitted: header. If it contains the string "auto-" the message it
2112 not considered personal.
211464. Added rcpt_include_affixes as a generic transport option.
211665. Added queue_only_override (default true).
211866. Added the syslog_duplication option.
212067. If what should have been the first header line of a message consisted of
2121 a space followed by a colon, Exim was mis-interpreting it as a header line.
2122 It isn't of course - it is syntactically invalid and should therefore be
2123 treated as the start of the message body. The misbehaviour could have
2124 caused a number of strange effects, including loss of data in subsequent
2125 header lines, and spool format errors.
212768. Formerly, the AUTH parameter on a MAIL command was trusted only if the
2128 client host had authenticated. This control can now be exercised by an ACL
2129 for more flexibility.
213169. By default, callouts do not happen when testing with -bh. There is now a
2132 variant, -bhc, which does actually run the callout code, including
2133 consulting and updating the callout cache.
213570. Added support for saslauthd authentication, courtesy of Alexander
2136 Sabourenkov.
213871. If statvfs() failed on the spool or log directories while checking their
2139 size for availability, Exim confusingly gave the error "space shortage".
2140 Furthermore, in debugging mode it crashed with a floating point exception.
2141 These checks are done if check_{spool,log}_{space,inodes} are set, and when
2142 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
2143 serious problem, Exim now writes to the main and panic logs when this
2144 happens, with details of the failure. It then refuses to accept the
2145 incoming message, giving the message "spool directory problem" or "log
2146 directory problem" with a 421 code for SMTP messages.
214872. When Exim is about to re-exec itself, it ensures that the file descriptors
2149 0, 1, and 2 exist, because some OS complain for execs without them (see
2150 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
2151 descriptors. However, the code omitted to check that the open succeeded,
2152 causing mysterious errors if for some reason the permissions on /dev/null
2153 got screwed. Now Exim writes a message to the main and panic logs, and
2154 bombs out if it can't open /dev/null.
215673. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
2157 interact so that it is all more flexible. It is supposed to remain
2158 backwards compatible. Also added extra_local_interfaces.
216074. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
2161 to bomb out with an assertion failure - to the client this appears as a
2162 connection drop. This problem occurs in the part of the code that was taken
2163 from the Samba project. Fortunately, the assertion is in a very simple
2164 function, so I have fixed this by reproducing the function inline in the
2165 one place where it is called, and arranging for authentication to fail
2166 instead of killing the process with assert().
216875. The SPA client code was not working when the server requested OEM rather
2169 than Unicode encoding.
217176. Added code to make require_files with a specific uid setting more usable in
2172 the case where statting the file as root fails - usually a non-root-mounted
2173 NFS file system. When this happens and the failure is EACCES, Exim now
2174 forks a subprocess and does the per-uid checking as the relevant uid.
217677. Added process_log_path.
217878. If log_file_path was not explicitly set, a setting of check_log_space or
2179 check_log_inodes was ignored.
218179. If a space check for the spool or log partitions fails, the incident is now
2182 logged. Of course, in the latter case the data may get lost...
218480. Added the %p formatting code to string_format() so that it can be used to
2185 print addresses in debug_print(). Adjusted all the address printing in the
2186 debugging in store.c to use %p rather than %d.
218881. There was a concern that a line of code in smtp_in.c could overflow a
2189 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
2190 initially expressed, the concern was not well-founded, because trailing
2191 spaces are removed early. However, if the trailing spaces were followed by
2192 a NULL, they did not get removed, so the overflow was possible. Two fixes
2193 were applied:
2195 (a) I re-wrote the offending code in a cleaner fashion.
2196 (b) If an incoming SMTP command contains a NULL character, it is rejected
2197 as invalid.
219982. When Exim changes uid/gid to the Exim user at daemon start time, it now
2200 runs initgroups(), so that if the Exim user is in any additional groups,
2201 they will be used during message reception.
2204Exim version 4.20
2207The change log for 4.20 and earlier releases has been archived.