Added $spool_size, $log_size, $spool_inodes, $log_inodes.
[exim.git] / doc / doc-txt / ChangeLog
5cb8cbc6 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.29 2004/11/17 14:32:25 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.
126Exim version 4.43
129 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
130 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
131 Now it returns an integer. A number of calls were still expecting a T/F
132 return. Fortuitously, in all cases, the tests worked in OK situations,
133 which is the norm. However, things would have gone wrong on any write
134 failures on the smtp file descriptor. This function is used when sending
135 messages over SMTP and also when doing verify callouts.
137 2. When Exim is called to do synchronous delivery of a locally submitted
138 message (the -odf or -odi options), it no longer closes stderr before doing
139 the delivery.
141 3. Implemented the mua_wrapper option.
143 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
145 5. Implemented the functions header_remove(), header_testname(),
146 header_add_at_position(), and receive_remove_recipient(), and exported them
147 to local_scan().
149 6. If an ACL "warn" statement specified the addition of headers, Exim already
150 inserted X-ACL-Warn: at the start if there was no header name. However, it
151 was not making this test for the second and subsequent header lines if
152 there were newlines in the string. This meant that an invalid header could
153 be inserted if Exim was badly configured.
155 7. Allow an ACL "warn" statement to add header lines at the start or after all
156 the Received: headers, as well as at the end.
158 8. Added the rcpt_4xx retry error code.
160 9. Added postmaster_mailfrom=xxx to callout verification option.
16210. Added mailfrom=xxxx to the callout verification option, for verify=
163 header_sender only.
16511. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
166 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
16812. Inserted some casts to stop certain compilers warning when using pointer
169 differences as field lengths or precisions in printf-type calls (mostly
170 affecting debugging statements).
17213. Added optional readline() support for -be (dynamically loaded).
17414. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
175 same clock tick as a message's arrival, so that its received time was the
176 same as the "first fail" time on the retry record, and that message
177 remained on the queue past the ultimate address timeout, every queue runner
178 would try a delivery (because it was past the ultimate address timeout) but
179 after another failure, the ultimate address timeout, which should have then
180 bounced the address, did not kick in. This was a "< instead of <=" error;
181 in most cases the first failure would have been in the next clock tick
182 after the received time, and all would be well.
18415. The special items beginning with @ in domain lists (e.g. @mx_any) were not
185 being recognized when the domain list was tested by the match_domain
186 condition in an expansion string.
18816. Added the ${str2b64: operator.
19017. Exim was always calling setrlimit() to set a large limit for the number of
191 processes, without checking whether the existing limit was already
192 adequate. (It did check for the limit on file descriptors.) Furthermore,
193 errors from getrlimit() and setrlimit() were being ignored. Now they are
194 logged to the main and panic logs, but Exim does carry on, to try to do its
195 job under whatever limits there are.
19718. Imported PCRE 5.0.
19919. Trivial typo in log message " temporarily refused connection" (the leading
200 space).
20220. If the log selector return_path_on_delivery was set and an address was
203 redirected to /dev/null, the delivery process crashed because it assumed
204 that a return path would always be set for a "successful" delivery. In this
205 case, the whole delivery is bypassed as an optimization, and therefore no
206 return path is set.
20821. Internal re-arrangement: the function for sending a challenge and reading
209 a response while authentication was assuming a zero-terminated challenge
210 string. It's now changed to take a pointer and a length, to allow for
211 binary data in such strings.
21322. Added the cyrus_sasl authenticator (code supplied by MBM).
21523. Exim was not respecting finduser_retries when seeking the login of the
216 uid under which it was called; it was always trying 10 times. (The default
217 setting of finduser_retries is zero.) Also, it was sleeping after the final
218 failure, which is pointless.
22024. Implemented tls_on_connect_ports.
22225. Implemented acl_smtp_predata.
22426. If the domain in control=submission is set empty, Exim assumes that the
225 authenticated id is a complete email address when it generates From: or
226 Sender: header lines.
22827. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
229 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
230 chown and chgrp in /bin and hostname in /usr/bin.
23228. Exim was keeping the "process log" file open after each use, just as it
233 does for the main log. This opens the possibility of it remaining open for
234 long periods when the USR1 signal hits a daemon. Occasional processlog
235 errors were reported, that could have been caused by this. Anyway, it seems
236 much more sensible not to leave this file open at all, so that is what now
237 happens.
23929. The long-running daemon process does not normally write to the log once it
240 has entered its main loop, and it closes the log before doing so. This is
241 so that log files can straightforwardly be renamed and moved. However,
242 there are a couple of unusual error situations where the daemon does write
243 log entries, and I had neglected to close the log afterwards.
24530. The text of an SMTP error response that was received during a remote
246 delivery was being truncated at 512 bytes. This is too short for some of
247 the long messages that one sometimes sees. I've increased the limit to
248 1024.
25031. It is now possible to make retry rules that apply only when a message has a
251 specific sender, in particular, an empty sender.
25332. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
254 it possible to be selective about when SMTP synchronization is enforced.
25633. Added "control = caseful_local_part" and "control = "caselower_local_part".
25832. Implemented hosts_connection_nolog.
26033. Added an ACL for QUIT.
26234. Setting "delay_warning=" to disable warnings was not working; it gave a
263 syntax error.
26535. Added mailbox_size and mailbox_filecount to appendfile.
26736. Added control = no_multiline_responses to ACLs.
26937. There was a bug in the logic of the code that waits for the clock to tick
270 in the case where the clock went backwards by a substantial amount such
271 that the microsecond fraction of "now" was more than the microsecond
272 fraction of "then" (but the whole seconds number was less).
27438. Added support for the libradius Radius client library this is found on
275 FreeBSD (previously only the radiusclient library was supported).
278Exim version 4.42
281 1. When certain lookups returned multiple values in the form name=value, the
282 quoting of the values was not always being done properly. Specifically:
283 (a) If the value started with a double quote, but contained no whitespace,
284 it was not quoted.
285 (b) If the value contained whitespace other than a space character (i.e.
286 tabs or newlines or carriage returns) it was not quoted.
287 This fix has been applied to the mysql and pgsql lookups by writing a
288 separate quoting function and calling it from the lookup code. The fix
289 should probably also be applied to nisplus, ibase and oracle lookups, but
290 since I cannot test any of those, I have not disturbed their existing code.
292 2. A hit in the callout cache for a specific address caused a log line with no
293 reason for rejecting RCPT. Now it says "Previous (cached) callout
294 verification failure".
296 3. There was an off-by-one bug in the queryprogram router. An over-long
297 return line was truncated at 256 instead of 255 characters, thereby
298 overflowing its buffer with the terminating zero. As well as fixing this, I
299 have increased the buffer size to 1024 (and made a note to document this).
301 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
302 when Exim is waiting for an SMTP response from a remote server, Exim
303 restarts its select() call on the socket, thereby resetting its timeout.
304 This is not a problem when such interrupts are rare. Somebody set up a cron
305 job to run exiwhat every 2 minutes, which is less than the normal select()
306 timeout (5 or 10 minutes). This meant that the select() timeout never
307 kicked in because it was always reset. I have fixed this by comparing the
308 time when an interrupt arrives with the time at the start of the first call
309 to select(). If more time than the timeout has elapsed, the interrupt is
310 treated as a timeout.
312 5. Some internal re-factoring in preparation for the addition of Sieve
313 extensions (by MH). In particular, the "personal" test is moved to a
314 separate function, and given an option for scanning Cc: and Bcc: (which is
315 not set for Exim filters).
317 6. When Exim created an email address using the login of the caller as the
318 local part (e.g. when creating a From: or Sender: header line), it was not
319 quoting the local part when it contained special characters such as @.
321 7. Installed new OpenBSD configuration files.
323 8. Reworded some messages for syntax errors in "and" and "or" conditions to
324 try to make them clearer.
326 9. Callout options, other than the timeout value, were being ignored when
327 verifying sender addresses in header lines. For example, when using
329 verify = header_sender/callout=no_cache
331 the cache was (incorrectly) being used.
33310. Added a missing instance of ${EXE} to the exim_install script; this affects
334 only the Cygwin environment.
33611. When return_path_on_delivery was set as a log selector, if different remote
337 addresses in the same message used different return paths and parallel
338 remote delivery occurred, the wrong values would sometimes be logged.
339 (Whenever a remote delivery process finished, the return path value from
340 the most recently started remote delivery process was logged.)
34212. RFC 3848 specifies standard names for the "with" phrase in Received: header
343 lines when AUTH and/or TLS are in use. This is the "received protocol"
344 field. Exim used to use "asmtp" for authenticated SMTP, without any
345 indication (in the protocol name) for TLS use. Now it follows the RFC and
346 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
347 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
348 names appear in log lines as well as in Received: header lines.
35013. Installed MH's patches for Sieve to add the "copy" and "vacation"
351 extensions, and comparison tests, and to fix some bugs.
35314. Changes to the "personal" filter test:
355 (1) The test was buggy in that it was just doing the equivalent of
356 "contains" tests on header lines. For example, if a user's address was
357 anne@some.where, the "personal" test would incorrectly be true for
359 To: susanne@some.where
361 This test is now done by extracting each address from the header in turn,
362 and checking the entire address. Other tests that are part of "personal"
363 are now done using regular expressions (for example, to check local parts
364 of addresses in From: header lines).
366 (2) The list of non-personal local parts in From: addresses has been
367 extended to include "listserv", "majordomo", "*-request", and "owner-*",
368 taken from the Sieve specification recommendations.
370 (3) If the message contains any header line starting with "List-" it is
371 treated as non-personal.
373 (4) The test for "circular" in the Subject: header line has been removed
374 because it now seems ill-conceived.
37615. Minor typos in src/EDITME comments corrected.
37816. Installed latest exipick from John Jetmore.
38017. If headers_add on a router specified a text string that was too long for
381 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
382 of string_sprintf() is now avoided.
38418. $message_body_size was not set (it was always zero) when running the DATA
385 ACL and the local_scan() function.
38719. For the "mail" command in an Exim filter, no default was being set for
388 the once_repeat time, causing a random time value to be used if "once" was
389 specified. (If the value happened to be <= 0, no repeat happened.) The
390 default is now 0s, meaning "never repeat". The "vacation" command was OK
391 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
392 (I found it when inspecting the code).
39420. There is now an overall timeout for performing a callout verification. It
395 defaults to 4 times the callout timeout, which applies to individual SMTP
396 commands during the callout. The overall timeout applies when there is more
397 than one host that can be tried. The timeout is checked before trying the
398 next host. This prevents very long delays if there are a large number of
399 hosts and all are timing out (e.g. when the network connections are timing
400 out). The value of the overall timeout can be changed by specifying an
401 additional sub-option for "callout", called "maxwait". For example:
403 verify = sender/callout=5s,maxwait=20s
40521. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
406 to the end before writing, but this should make it even safer).
40822. Exim was forgetting that it had advertised PIPELINING for the second and
409 subsequent messages on an SMTP connection. It was also not resetting its
410 memory on STARTTLS and an internal HELO.
41223. When Exim logs an SMTP synchronization error within a session, it now
413 records whether PIPELINING has been advertised or not.
41524. Added 3 instances of "(long int)" casts to time_t variables that were being
416 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
417 rather than long int.
41925. Installed the latest Cygwin configuration files from the Cygwin maintainer.
42126. Added the never_mail option to autoreply.
424Exim version 4.41
427 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
428 crash if the getsockname() call failed; this can happen if a connection is
429 closed very soon after it is established. The problem was simply in the
430 order in which certain operations were done, causing Exim to try to write
431 to the SMTP stream before it had set up the file descriptor. The bug has
432 been fixed by making things happen in the correct order.
435Exim version 4.40
438 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
439 before the connection was closed, thus losing the rejection response.
441 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
442 some early Solaris releases, but causes trouble in current releases where
443 socklen_t is defined.
445 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
446 always exist.
448 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
449 configured.
451 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
452 that releasing the top of it at the end releases what was used for sub-
453 expansions (unless the block got too big). However, discard this block if
454 the first thing is a variable or header, so that we can use its block when
455 it is dynamic (useful for very large $message_headers, for example).
457 6. Lookups now cache *every* query, not just the most recent. A new, separate
458 store pool is used for this. It can be recovered when all lookup caches are
459 flushed. Lookups now release memory at the end of their result strings.
460 This has involved some general refactoring of the lookup sources.
462 7. Some code has been added to the store_xxx() functions to reduce the amount
463 of flapping under certain conditions.
465 8. log_incoming_interface used to affect only the <= reception log lines. Now
466 it causes the local interface and port to be added to several more SMTP log
467 lines, for example "SMTP connection from", and rejection lines.
469 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
47110. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
47311. If $host_data was set by virtue of a hosts lookup in an ACL, its value
474 could be overwritten at the end of the current message (or the start of a
475 new message if it was set in a HELO ACL). The value is now preserved for
476 the duration of the SMTP connection.
47812. If a transport had a headers_rewrite setting, and a matching header line
479 contained an unqualified address, that address was qualified, even if it
480 did not match any rewriting rules. The underlying bug was that the values
481 of the flags that permit the existence of unqualified sender and recipient
482 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
483 non-local messages, and by -bnq for local messages) were not being
484 preserved with the message after it was received.
48613. When Exim was logging an SMTP synchronization error, it could sometimes log
487 "next input=" as part of the text comprising the host identity instead of
488 the correct text. The code was using the same buffer for two different
489 strings. However, depending on which order the printing function evaluated
490 its arguments, the bug did not always show up. Under Linux, for example, my
491 test suite worked just fine.
49314. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
494 doesn't work with some older versions of Perl. It has been changed to "my",
495 which in any case is probably the better facility to use.
49715. A really picky compiler found some instances of statements for creating
498 error messages that either had too many or two few arguments for the format
499 string.
50116. The size of the buffer for calls to the DNS resolver has been increased
502 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
503 for addresses that have a lot of PTR records. This alleviates a problem; it
504 does not fully solve it.
50617. A dnsdb lookup for PTR records that receives more data than will fit in the
507 buffer now truncates the list and logs the incident, which is the same
508 action as happens when Exim is looking up a host name and its aliases.
509 Previously in this situation something unpredictable would happen;
510 sometimes it was "internal error: store_reset failed".
51218. If a server dropped the connection unexpectedly when an Exim client was
513 using GnuTLS and trying to read a response, the client delivery process
514 crashed while trying to generate an error log message.
51619. If a "warn" verb in an ACL added multiple headers to a message in a single
517 string, for example:
519 warn message = H1: something\nH2: something
521 the text was added as a single header line from Exim's point of view
522 though it ended up OK in the delivered message. However, searching for the
523 second and subsequent header lines using $h_h2: did not work. This has been
524 fixed. Similarly, if a system filter added multiple headers in this way,
525 the routers could not see them.
52720. Expanded the error message when iplsearch is called with an invalid key to
528 suggest using net-iplsearch in a host list.
53021. When running tests using -bh, any delays imposed by "delay" modifiers in
531 ACLs are no longer actually imposed (and a message to that effect is
532 output).
53422. If a "gecos" field in a passwd entry contained escaped characters, in
535 particular, if it contained a \" sequence, Exim got it wrong when building
536 a From: or a Sender: header from that name. A second bug also caused
537 incorrect handling when an unquoted " was present following a character
538 that needed quoting.
54023. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
541 was not being matched caselessly.
54324. Arranged for all hyphens in the exim.8 source to be escaped with
544 backslashes.
54625. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
547 properly. Recipient callout cache records were still being keyed to include
548 the sender, even when use_sender was set false. This led to far more
549 callouts that were necessary. The sender is no longer included in the key
550 when use_sender is false.
55226. Added "control = submission" modifier to ACLs.
55427. Added the ${base62d: operator to decode base 62 numbers.
55628. dnsdb lookups can now access SRV records.
55829. CONFIGURE_OWNER can be set at build time to define an alternative owner for
559 the configuration file.
56130. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
562 (-v) mode. This makes the output for a verbose queue run more intelligible.
56431. Added a use_postmaster feature to recipient callouts.
56632. Added the $body_zerocount variable, containing the number of binary zero
567 bytes in the message body.
56933. The time of last modification of the "new" subdirectory is now used as the
570 "mailbox time last read" when there is a quota error for a maildir
571 delivery.
57334. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
57535. Added +ignore_unknown as a special item in host lists.
57736. Code for decoding IPv6 addresses in host lists is now included, even if
578 IPv6 support is not being compiled. This fixes a bug in which an IPv6
579 address was recognized as an IP address, but was then not correctly decoded
580 into binary, causing unexpected and incorrect effects when compared with
581 another IP address.
584Exim version 4.34
587 1. Very minor rewording of debugging text in manualroute to say "list of
588 hosts" instead of "hostlist".
590 2. If verify=header_syntax was set, and a header line with an unqualified
591 address (no domain) and a large number of spaces between the end of the
592 name and the colon was received, the reception process suffered a buffer
593 overflow, and (when I tested it) crashed. This was caused by some obsolete
594 code that should have been removed. The fix is to remove it!
596 3. When running in the test harness, delay a bit after writing a bounce
597 message to get a bit more predictability in the log output.
599 4. Added a call to search_tidyup() just before forking a reception process. In
600 theory, someone could use a lookup in the expansion of smtp_accept_max_
601 per_host which, without the tidyup, could leave open a database connection.
603 5. Added the variables $recipient_data and $sender_data which get set from a
604 lookup success in an ACL "recipients" or "senders" condition, or a router
605 "senders" option, similar to $domain_data and $local_part_data.
607 6. Moved the writing of debug_print from before to after the "senders" test
608 for routers.
610 7. Change 4.31/66 (moving the time when the Received: is generated) caused
611 problems for message scanning, either using a data ACL, or using
612 local_scan() because the Received: header was not generated till after they
613 were called (in order to set the time as the time of reception completion).
614 I have revised the way this works. The header is now generated after the
615 body is received, but before the ACL or local_scan() are called. After they
616 are run, the timestamp in the header is updated.
619Exim version 4.33
622 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
623 before starting a queue runner without re-exec. This happened only when
624 deliver_drop_privilege was set or when the Exim user was set to root. The
625 effect of the bug was that timeouts during subsequent deliveries caused
626 crashes instead of being properly handled. The handler is now left at its
627 default (and expected) setting.
629 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
630 message, again when deliver_drop_privilege is set or Exim is run as root.
631 The bug described in (1) was not present in this case, but the tidying up
632 of the other signals was missing. I have made the two cases consistent.
634 3. The ignore_target_hosts setting on a manualroute router was being ignored
635 for hosts that were looked up using the /MX notation.
637 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
638 in domain lists.
640 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
641 operated on the sender address. After changing the $sender_address to <>
642 for the sender address verify, Exim was re-instated it as the original
643 (before rewriting) address, but remembering that it had rewritten it, so it
644 wasn't rewriting it again. This bug also had the effect of breaking the
645 sender address verification caching when the sender address was rewritten.
647 6. The ignore_target_hosts option was being ignored by the ipliteral router.
648 This has been changed so that if the ip literal address matches
649 ignore_target_hosts, the router declines.
651 7. Added expansion conditions match_domain, match_address, and match_local_
652 part (NOT match_host).
654 8. The placeholder for the Received: header didn't have a length field set.
656 9. Added code to Exim itself and to exim_lock to test for a specific race
657 condition that could lead to file corruption when using MBX delivery. The
658 issue is with the lockfile that is created in /tmp. If this file is removed
659 after a process has opened it but before that process has acquired a lock,
660 there is the potential for a second process to recreate the file and also
661 acquire a lock. This could lead to two Exim processes writing to the file
662 at the same time. The added code performs the same test as UW imapd; it
663 checks after acquiring the lock that its file descriptor still refers to
664 the same named file.
66610. The buffer for building added header lines was of fixed size, 8192 bytes.
667 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
668 when Exim is built.
67011. Added the smtp_active_hostname option. If used, this will typically be made
671 to depend on the incoming interface address. Because $interface_address is
672 not set up until the daemon has forked a reception process, error responses
673 that can happen earlier (such as "too many connections") no longer contain
674 a host name.
67612. If an expansion in a condition on a "warn" statement fails because a lookup
677 defers, the "warn" statement is abandoned, and the next ACL statement is
678 processed. Previously this caused the whole ACL to be aborted.
68013. Added the iplsearch lookup type.
68214. Added ident_timeout as a log selector.
68415. Added tls_certificate_verified as a log selector.
68616. Added a global option tls_require_ciphers (compare the smtp transport
687 option of the same name). This controls incoming TLS connections.
68917. I finally figured out how to make tls_require_ciphers do a similar thing
690 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
691 before starting the TLS session.
69318. Tabs are now shown as \t in -bP output.
69519. If the log selector return_path_on_delivery was set, Exim crashed when
696 bouncing a message because it had too many Received: header lines.
69820. If two routers both had headers_remove settings, and the first one included
699 a superfluous trailing colon, the final name in the first list and the
700 first name in the second list were incorrectly joined into one item (with a
701 colon in the middle).
704Exim version 4.32
707 1. Added -C and -D options to the exinext utility, mainly to make it easier
708 to include in the automated testing, but these could be helpful when
709 multiple configurations are in use.
711 2. The exinext utility was not formatting the output nicely when there was
712 an alternate port involved in the retry record key, nor when there was a
713 message id as well (for retries that were specific to a specific message
714 and a specific host). It was also confused by IPv6 addresses, because of
715 the additional colons they contain. I have fixed the IPv4 problem, and
716 patched it up to do a reasonable job for IPv6.
718 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
719 delivery, the log line now contains "pipelined" if PIPELINING was used.
721 4. An SMTP transport process used to panic and die if the bind() call to set
722 an explicit outgoing interface failed. This has been changed; it is now
723 treated in the same way as a connect() failure.
725 5. A reference to $sender_host_name in the part of a conditional expansion
726 that was being skipped was still causing a DNS lookup. This no longer
727 occurs.
729 6. The def: expansion condition was not recognizing references to header lines
730 that used bh_ and bheader_.
732 7. Added the _cache feature to named lists.
734 8. The code for checking quota_filecount in the appendfile transport was
735 allowing one more file than it should have been.
737 9. For compatibility with Sendmail, the command line option
739 -prval:sval
741 is equivalent to
743 -oMr rval -oMs sval
745 and sets the incoming protocol and host name (for trusted callers). The
746 host name and its colon can be omitted when only the protocol is to be set.
747 Note the Exim already has two private options, -pd and -ps, that refer to
748 embedded Perl. It is therefore impossible to set a protocol value of "d" or
749 "s", but I don't think that's a major issue.
75110. A number of refactoring changes to the code, none of which should affect
752 Exim's behaviour:
754 (a) The number of logging options was getting close to filling up the
755 32-bit word that was used as a bit map. I have split them into two classes:
756 those that are passed in the argument to log_write(), and those that are
757 only ever tested independently outside of that function. These are now in
758 separate 32-bit words, so there is plenty of room for expansion again.
759 There is no change in the user interface or the logging behaviour.
761 (b) When building, for example, log lines, the code previously used a
762 macro that called string_cat() twice, in order to add two strings. This is
763 not really sufficiently general. Furthermore, there was one instance where
764 it was actually wrong because one of the argument was used twice, and in
765 one call a function was used. (As it happened, calling the function twice
766 did not affect the overall behaviour.) The macro has been replaced by a
767 function that can join an arbitrary number of extra strings onto a growing
768 string.
770 (c) The code for expansion conditions now uses a table and a binary chop
771 instead of a serial search (which was left over from when there were very
772 few conditions). Also, it now recognizes conditions like "pam" even when
773 the relevant support is not compiled in: a suitably worded error message is
774 given if an attempt is made to use such a condition.
77611. Added ${time_interval:xxxxx}.
77812. A bug was causing one of the ddress fields not to be passed back correctly
779 from remote delivery subprocesses. The field in question was not being
780 subsequently used, so this caused to problems in practice.
78213. Added new log selectors queue_time and deliver_time.
78414. Might have fixed a bug in maildirsizefile handling that threw up
785 "unexpected character" debug warnings, and recalculated the data
786 unnecessarily. In any case, I expanded the warning message to give more
787 information.
78915. Added the message "Restricted characters in address" to the statements in
790 the default ACL that block characters like @ and % in local parts.
79216. Change 71 for release 4.31 proved to be much less benign that I imagined.
793 Three changes have been made:
795 (a) There was a serious bug; a negative response to MAIL caused the whole
796 recipient domain to be cached as invalid, thereby blocking all messages
797 to all local parts at the same domain, from all senders. This bug has
798 been fixed. The domain is no longer cached after a negative response to
799 MAIL if the sender used is not empty.
801 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
802 been restored.
804 (c) A new callout option, "use_sender" has been added for people who want
805 the modified behaviour.
808Exim version 4.31
811 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
812 Larry Rosenman.
814 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
815 indeed breaks things for older releases.
817 3. Added additional logging to the case where there is a problem reading data
818 from a filter that is running in a subprocess using a pipe, in order to
819 try to track down a specific problem.
821 4. Testing facility fudge: when running in the test harness and attempting
822 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
823 getting "No route to host". Convert this to a timeout.
825 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
826 warning.
828 6. Some OS don't have socklen_t but use size_t instead. This affects the
829 fifth argument of getsockopt() amongst other things. This is now
830 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
831 can be set for individual OS. I have set it for SunOS5, OSF1, and
832 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
833 some earlier ones do not.
835 7. Change 4.30/15 was not doing the test caselessly.
837 8. The standard form for an IPv6 address literal was being rejected by address
838 parsing in, for example, MAIL and RCPT commands. An example of this kind of
839 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
840 this, as well as the form without the "IPv6" on the front (but only when
841 address literals are enabled, of course).
843 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
84510. Exim crashed if a message with an empty sender address specified by -f
846 encountered a router with an errors_to setting. This could be provoked only
847 by a command such as
849 exim -f "" ...
851 where an empty string was supplied; "<>" did not hit this bug.
85311. Installed PCRE release 4.5.
85512. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
856 remained set. It is now erased.
85813. exiqgrep wasn't working on MacOS X because it didn't correctly compute
859 times from message ids (which are base 36 rather than the normal 62).
86114. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
862 were being counted as actual protocol errors, and logged if the log
863 selector +smtp_protocol_error was set. One cannot be perfect in this test,
864 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
865 and DATA following a set of rejected RCPTs do not count as protocol errors.
866 In other words, Exim assumes they were pipelined, though this may not
867 actually be the case. Of course, in all cases the client gets an
868 appropriate error code.
87015. If a lookup fails in an ACL condition, a message about the failure may
871 be available; it is used if testing the ACL cannot continue, because most
872 such messages specify what the cause of the deferral is. However, some
873 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
874 that caused an old message to be retained and used if a later statement
875 caused a defer, replacing the real cause of the deferral.
87716. If an IP address had so many PTR records that the DNS lookup buffer
878 was not large enough to hold them, Exim could crash while trying to process
879 the truncated data. It now detects and logs this case.
88117. Further to 4.21/58, another change has been made: if (and only if) the
882 first line of a message (the first header line) ends with CRLF, a bare LF
883 in a subsequent header line has a space inserted after it, so as not to
884 terminate the header.
88618. Refactoring: tidied an ugly bit of code in appendfile that copied data
887 unnecessarily, used atoi() instead of strtol(), and didn't check the
888 termination when getting file sizes from file names by regex.
89019. Completely re-implemented the support for maildirsize files, in the light
891 of a number of problems with the previous contributed implementation
892 (4.30/29). In particular:
894 . If the quota is zero, the maildirsize file is maintained, but no quota is
895 imposed.
897 . If the maildir directory does not exist, it is created before any attempt
898 to write a maildirsize file.
900 . The quota value in the file is just a cache; if the quota is changed in
901 the transport, the new value overrides.
903 . A regular expression is available for excluding directories from the
904 count.
90620. The autoreply transport checks the characters in options that define the
907 message's headers; it allows continued headers, but it was checking with
908 isspace() after an embedded newline instead of explicitly looking for a
909 space or a tab.
91121. If all the "regular" hosts to which an address was routed had passed their
912 expiry times, and had not reached their retry times, the address was
913 bounced, even if fallback hosts were defined. Now Exim should go on to try
914 the fallback hosts.
91622. Increased buffer sizes in the callout code from 1024 to 4096 to match the
917 equivalent code in the SMTP transport. Some hosts send humungous responses
918 to HELO/EHLO, more than 1024 it seems.
92023. Refactoring: code in filter.c used (void *) for "any old type" but this
921 gives compiler warnings in some environments. I've now done it "properly",
922 using a union.
92424. The replacement for inet_ntoa() that is used with gcc on IRIX systems
925 (because of problems with the built-in one) was declared to return uschar *
926 instead of char *, causing compiler failure.
92825. Fixed a file descriptor leak when processing alias/forward files.
93026. Fixed a minor format string issue in dbfn.c.
93227. Typo in exim.c: ("dmbnz" for "dbmnz").
93428. If a filter file refered to $h_xxx or $message_headers, and the headers
935 contained RFC 2047 "words", Exim's memory could, under certain conditions,
936 become corrupted.
93829. When a sender address is verified, it is cached, to save repeating the test
939 when there is more than one recipient in a message. However, when the
940 verification involves a callout, it is possible for different callout
941 options to be set for different recipients. It is too complicated to keep
942 track of this in the cache, so now Exim always runs a verification when a
943 callout is required, relying on the callout cache for the optimization.
944 The overhead is duplication of the address routing, but this should not be
945 too great.
94730. Fixed a bug in callout caching. If a RCPT command caused the sender address
948 to be verified with callout=postmaster, and the main callout worked but the
949 postmaster check failed, the verification correctly failed. However, if a
950 subsequent RCPT command asked for sender verification *without* the
951 postmaster check, incorrect caching caused this verification also to fail,
952 incorrectly.
95431. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
955 it was not caching the DNS options (qualify_single, search_parents) that
956 were used when the lookup failed. A subsequent lookup with different
957 options therefore always gave the same answer, though there were cases
958 where it should not have. (Example: a "domains = !$mx_any" option on a
959 dnslookup router: the "domains" option is always processed without any
960 widening, but the router might have qualify_single set.) Now Exim uses the
961 cached value only when the same options are set.
96332. Added John Jetmore's "exipick" utility to the distribution.
96533. GnuTLS: When an attempt to start a TLS session fails for any reason other
966 than a timeout (e.g. a certificate is required, and is not provided), an
967 Exim server now closes the connection immediately. Previously it waited for
968 the client to close - but if the client is SSL, it seems that they each
969 wait for each other, leading to a delay before one of them times out.
97134: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
972 maintained 0.8.x compatibility because I don't think many are using it, and
973 it is clearly obsolete.
97535. Added TLS support for CRLs: a tls_crl global option and one for the smtp
976 transport.
97836. OpenSSL: $tls_certificate_verified was being set to 1 even if the
979 client certificate was expired. A simple patch fixes this, though I don't
980 understand the full logic of why the verify callback is called multiple
981 times.
98337. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
984 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
985 which causes problems with some clients (such as the Certicom SSL Plus
986 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
987 disables the coutermeasure allowing Eudora to connect."
98938. Exim was not checking that a write() to a log file succeeded. This could
990 lead to Bad Things if a log got too big, in particular if it hit a file
991 size limit. Exim now panics and dies if it cannot write to a log file, just
992 as it does if it cannot open a log file.
99439. Modified OS/Makefile-Linux so that it now contains
998 The two -D definitions ensure that Exim is compiled with large file
999 support, which makes it possible to handle log files that are bigger than
1000 2^31.
100240. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
1003 instance) a domain was checked against a named list that involved a lookup,
1004 causing $domain_data to be set, then another domain was checked against the
1005 same list, then the first domain was re-checked, the value of $domain_data
1006 after the final check could be wrong. In particular, if the second check
1007 failed, it could be set empty. This bug probably also applied to
1008 $localpart_data.
101041. The strip_trailing_dot option was not being applied to the address given
1011 with the -f command-line option.
101342. The code for reading a message's header from the spool was incrementing
1014 $received_count, but never initializing it. This meant that the value was
1015 incorrect (doubled) while delivering a message in the same process in which
1016 it was received. In the most common configuration of Exim, this never
1017 happens - a fresh exec is done - but it can happen when
1018 deliver_drop_privilege is set.
102043. When Exim logs an SMTP synchronization error - client data sent too soon -
1021 it now includes up to 150 characters of the unexpected data in the log
1022 line.
102444. The exim_dbmbuild utility uses fixed size buffers for reading input lines
1025 and building data strings. The size of both of these buffers was 10 000
1026 bytes - far larger than anybody would *ever* want, thought I. Needless to
1027 say, somebody hit the limit. I have increased the maximum line length to
1028 20 000 and the maximum data length of concatenated lines to 100 000. I have
1029 also fixed two bugs, because there was no checking on these buffers. Tsk,
1030 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1031 buffer is too small.
103345. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1034 lsearch lookups. Now it does.
103646. When parsing a route_list item in a manualroute router, a fixed-length
1037 buffer was used for the list of hosts. I made this 1024 bytes long,
1038 thinking that nobody would ever have a list of hosts that long. Wrong.
1039 Somebody had a whole pile of complicated expansion conditions, and the
1040 string was silently truncated, leading to an expansion error. It turns out
1041 that it is easier to change to an unlimited length (owing to other changes
1042 that have happened since this code was originally written) than to build
1043 structure for giving a limitation error. The length of the item that
1044 expands into the list of hosts is now unlimited.
104647. The lsearch lookup could not handle data where the length of text line was
1047 more than 4095 characters. Such lines were truncated, leading to shortened
1048 data being returned. It should now handle lines of any length.
105048. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
1051 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
1052 ACL").
105449. Cosmetic tidy to scripts like exicyclog that are generated by globally
1055 replacing strings such as BIN_DIRECTORY in a source file: the replacement
1056 no longer happens in comment lines. A list of replacements is now placed
1057 at the head of all of the source files, except those whose only change is
1058 to replace PERL_COMMAND in the very first #! line.
106050. Replaced the slow insertion sort in queue.c, for sorting the list of
1061 messages on the queue, with a bottom-up merge sort, using code contributed
1062 by Michael Haardt. This should make operations like -bp somewhat faster on
1063 large queues. It won't affect queue runners, except when queue_run_in_order
1064 is set.
106651. Installed eximstats 1.31 in the distribution.
106852. Added support for SRV lookups to the dnslookup router.
107053. If an ACL referred to $message_body or $message_body_end, the value was not
1071 reset for any messages that followed in the same SMTP session.
107354. The store-handling optimization for building very long strings was not
1074 differentiating between the different store pools. I don't think this
1075 actually made any difference in practice, but I've tidied it.
107755. While running the routers to verify a sender address, $sender_address
1078 was still set to the sender address. This is wrong, because when routing to
1079 send a bounce to the sender, it would be empty. Therefore, I have changed
1080 it so that, while verifying a sender address, $sender_address is set to <>.
1081 (There is no change to what happens when verifying a recipient address.)
108356. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
1084 target A or AAAA records (if not already returned) without resetting the
1085 qualify_single or search_parents options of the DNS resolver. These are
1086 inappropriate in this case because the targets of MX and SRV records must
1087 be FQDNs. A broken DNS record could cause trouble if it happened to have a
1088 target that, when qualified, matched something in the local domain. These
1089 two options are now turned off when doing these lookups.
109157. It seems that at least some releases of Reiserfs (which does not have the
1092 concept of a fixed number of inodes) returns zero and not -1 for the
1093 number of available inodes. This interacted badly with check_spool_inodes,
1094 which assumed that -1 was the "no such thing" setting. What I have done is
1095 to check that the total number of inodes is greater than zero before doing
1096 the test of how many are available.
109858. When a "warn" ACL statement has a log_message modifier, the message is
1099 remembered, and not repeated. This is to avoid a lot of repetition when a
1100 message has many recipients that cause the same warning to be written.
1101 Howewer, Exim was preserving the list of already written lines for an
1102 entire SMTP session, which doesn't seem right. The memory is now reset if a
1103 new message is started.
110559. The "rewrite" debugging flag was not showing the result of rewriting in the
1106 debugging output unless log_rewrite was also set.
110860. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
1109 of (int)(handle) when we know that handle contains (void *)(-1).
111161. The Exim daemon panic-logs an error return when it closes the incoming
1112 connection. However "connection reset by peer" seems to be common, and
1113 isn't really an error worthy of noting specially, so that particular error
1114 is no long logged.
111662. When Exim is trying to find all the local interfaces, it used to panic and
1117 die if the ioctl to get the interface flags failed. However, it seems that
1118 on at least one OS (Solaris 9) it is possible to have an interface that is
1119 included in the list of interfaces, but for which you get a failure error
1120 for this call. This happens when the interface is not "plumbed" into a
1121 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
1122 failure of the "get flags" call assumes that the interface is down.
112463. Added a ${eval10: operator, which assumes all numbers are decimal. This
1125 makes life easier for people who are doing arithmetic on fields extracted
1126 from dates, where you often get leading zeros that should not be
1127 interpreted as octal.
112964. Added qualify_domain to the redirect router, to override the global
1130 setting.
113265. If a pathologically long header line contained very many addresses (the
1133 report of this problem mentioned 10 000) and each of them was rewritten,
1134 Exim could use up a very large amount of memory. (It kept on making new
1135 copies of the header line as it rewrote, and never released the old ones.)
1136 At the expense of a bit more processing, the header rewriting function has
1137 been changed so that it no longer eats memory in this way.
113966. The generation of the Received: header has been moved from the time that a
1140 message starts to be received, to the time that it finishes. The timestamp
1141 in the Received: header should now be very close to that of the <= log
1142 line. There are two side-effects of this change:
1144 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
1145 logged header lines no longer include the local Received: line, because
1146 it has not yet been created. The same applies to a copy of the message
1147 that is returned to a non-SMTP sender when a message is rejected.
1149 (b) When a filter file is tested using -bf, no additional Received: header
1150 is added to the test message. After some thought, I decided that this
1151 is a bug fix.
1153 This change does not affect the value of $received_for. It is still set
1154 after address rewriting, but before local_scan() is called.
115667. Installed the latest Cygwin-specific files from the Cygwin maintainer.
115868. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
1159 gave an unhelpful panic error message, and a defer error. I have managed to
1160 change this behaviour so that it now rejects any supplied certificate,
1161 which seems right, as the list of acceptable certificates is empty.
116369. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
1164 gave an unhelpful defer error. I have not managed to make this reject any
1165 supplied certificates, but the error message it gives is "no certificate
1166 supplied", which is not helpful.
116870. exigrep's output now also includes lines that are not associated with any
1169 message, but which match the given pattern. Implemented by a patch from
1170 Martin Sluka, which also tidied up the Perl a bit.
117271. Recipient callout verification, like sender verification, was using <> in
1173 the MAIL FROM command. This isn't really the right thing, since the actual
1174 sender may affect whether the remote host accepts the recipient or not. I
1175 have changed it to use the actual sender in the callout; this means that
1176 the cache record is now keyed on a recipient/sender pair, not just the
1177 recipient address. There doesn't seem to be a real danger of callout loops,
1178 since a callout by the remote host to check the sender would use <>.
1179 [SEE ABOVE: changed after hitting problems.]
118172. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
1182 temporary errors. However, in the case of such a code being given after
1183 the end of a data transmission (i.e. after ".") Exim was failing to write
1184 a retry record for the message. (Yes, there was some broken host that was
1185 actually sending 8xx at this point.)
118773. An unknown lookup type in a host list could cause Exim to panic-die when
1188 the list was checked. (An example that provoked this was putting <; in the
1189 middle of a list instead of at the start.) If this happened during a DATA
1190 ACL check, a -D file could be left lying around. This kind of configuration
1191 error no longer causes Exim to die; instead it causes a defer errror. The
1192 incident is still logged to the main and panic logs.
119474. Buglet left over from Exim 3 conversion. The message "too many messages
1195 in one connection" was written to the rejectlog but not the mainlog, except
1196 when address rewriting (yes!) was being logged.
119875. Added write_rejectlog option.
120076. When a system filter was run not as root (that is, when system_filter_user
1201 was set), the values of the $n variables were not being returned to the
1202 main process; thus, they were not subsequently available in the $sn
1203 variables.
120577. Added +return_path_on_delivery log selector.
120778. A connection timeout was being treated differently from recipients deferred
1208 when testing hosts_max_try with a message that was older than the host's
1209 retry timeout. (The host should not be counted, thus allowing all hosts to
1210 be tried at least once before bouncing.) This may have been the cause of an
1211 occasionally reported bug whereby a message would remain on the queue
1212 longer than the retry timeout, but would be bounced if a delivery was
1213 forced. I say "may" because I never totally pinned down the problem;
1214 setting up timeout/retry tests is difficult. See also the next item.
121679. The ultimate address timeout was not being applied to errors that involved
1217 a combination of host plus message (for example, a timeout on a MAIL
1218 command). When an address resolved to a number of possible hosts, and they
1219 were not all tried for each delivery (e.g. because of hosts_max_try), a
1220 message could remain on the queue longer than the retry timeout.
122280. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
1223 Haardt.
122581. Fixed an obscure SMTP outgoing bug which required at least the following
1226 conditions: (a) there was another message waiting for the same server;
1227 (b) the server returned 5xx to all RCPT commands in the first message so
1228 that the message was not completed; (c) the server dropped the connection
1229 or gave a negative response to the RSET that Exim sends to abort the
1230 transaction. The observed case was a dropped connection after DATA that had
1231 been sent in pipelining mode. That is, the server had advertised PIPELINING
1232 but was not implementing it correctly. The effect of the bug was incorrect
1233 behaviour, such as trying another host, and this could lead to a crash.
1236Exim version 4.30
1239 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
1240 and daemon.c were passed as pointers to ints; they should have been
1241 pointers to socklen_t variables (which are typically unsigned ints).
1243 2. Some signed/unsigned type warnings in the os.c file for Linux have been
1244 fixed.
1246 3. Fixed a really odd bug that affected only the testing scheme; patching a
1247 certain fixed string in the binary changed the value of another string that
1248 happened to be identical to the end of the original first string.
1250 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
1251 name", it returns that address as the IP address. On some operating
1252 systems (e.g. Solaris), it also passes back the IP address string as the
1253 "host name". However, on others (e.g. Linux), it passes back an empty
1254 string. Exim wasn't checking for this, and was changing the host name to an
1255 empty string, assuming it had been canonicized.
1257 5. Although rare, it is permitted to have more than one PTR record for a given
1258 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
1259 all the names associated with an address, because they do in Solaris.
1260 However, it seems that they do not in Linux for data that comes from the
1261 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
1262 I found this out when I moved to a new Linux workstation and tried to run
1263 the Exim test suite.
1265 To get round this problem I have changed the code so that it now does its
1266 own call to the DNS to look up PTR records when searching for a host name.
1267 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
1268 addresses that are only in /etc/hosts are still found.
1270 This behaviour is, however, controlled by an option called host_lookup_
1271 order, which defaults to "bydns:byaddr". If people want to use the other
1272 order, or indeed, just use one or the other means of lookup, they can
1273 specify it in this variable.
1275 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
1276 some operating systems, this comes back from gethostbyaddr() as an empty
1277 string, and this is what Exim used to test for. However, it seems that in
1278 other systems, "." is yielded. Exim now tests for this case too.
1280 7. The values of check_spool_space and check_log_space are now held internally
1281 as a number of kilobytes instead of an absolute number of bytes. If a
1282 numbers is specified without 'K' or 'M', it is rounded up to the nearest
1283 kilobyte. This means that much larger values can be stored.
1285 8. Exim monitor: an attempt to get the action menu when not actually pointing
1286 at a message produces an empty menu entitled "No message selected". This
1287 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
1288 no entries in it ("Shell widget menu has zero width and/or height"). So I
1289 have added a single, blank menu entry in this case.
1291 9. Added ${quote_local_part.
129310. MIME decoding is now applied to the contents of Subject: header lines when
1294 they are logged.
129611. Now that a reference to $sender_host_address automatically causes a reverse
1297 lookup to occur if necessary (4.13/18), there is no need to arrange for a
1298 host lookup before query-style lookups in lists that might use this
1299 variable. This has therefore been abolished, and the "net-" prefix is no
1300 longer necessary for query-style lookups.
130212. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
1303 have been a typo for LFLAGS, so it has been changed.
130513. The install script calls Exim with "-C /dev/null" in order to find the
1306 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
1307 to be output. Howeve, since Exim outputs its version number before the
1308 error, it didn't break the script. It just looked ugly. I fixed this by
1309 always allowing "-C /dev/null" if the caller is root.
131114. Ignore overlarge ACL variable number when reading spool file - insurance
1312 against a later release with more variables having written the file.
131415. The standard form for an IPv6 address literal was being rejected by EHLO.
1315 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1316 this, as well as the form without the "IPv6" on the front.
131816. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
1319 OS/Makefile-Darwin file.
132117. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
1322 only to LDAP libraries that do not have LDAP_OPT_DEREF.
132418. After change 4.21/52, "%ld" was used to format the contents of the $inode
1325 variable. However, some OS use ints for inodes. I've added cast to long int
1326 to get rid of the compiler warning.
132819. I had forgotten to lock out "/../" in configuration file names when
1329 ALT_CONFIG_PREFIX was set.
133120. Routers used for verification do not need to specify transports. However,
1332 if such a router generated a host list, and callout was configured, Exim
1333 crashed, because it could not find a port number from the (non-existent)
1334 transport. It now assumes port 25 in this circumstance.
133621. Added the -t option to exigrep.
133822. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
1339 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
1340 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
1341 them is set, LOOKUP_LSEARCH is forced.
134323. "exim -bV" now outputs a list of lookups that are included in the binary.
134524. Added sender and host information to the "rejected by local_scan()" log
1346 line; previously there was no indication of these.
134825. Added .include_if_exists.
135026. Change 3.952/11 added an explicit directory sync on top of a file sync for
1351 Linux. It turns out that not all file systems support this. Apparently some
1352 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
1353 do it.) To cope with this, the error EINVAL, which means that sync-ing is
1354 not supported on the file descriptor, is now ignored when Exim is trying to
1355 sync a directory. This applies only to Linux.
135727. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
135928. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
1360 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
1361 this.
136329. Added support for maildirsize files from supplied patch (modified a bit).
136530. The use of :fail: followed by an empty string could lead Exim to respond to
1366 sender verification failures with (e.g.):
1368 550 Verification failed for <xxx>
1369 550 Sender verify failed
1371 where the first response line was missing the '-' that indicates it is not
1372 the final line of the response.
137431. The loop for finding the name of the user that called Exim had a hardwired
1375 limit of 10; it now uses the value of finduser_retries, which is used for
1376 all other user lookups.
137832. Added $received_count variable, available in data and not_smtp ACLs, and at
1379 delivery time.
138133. Exim was neglecting to zero errno before one call of strtol() when
1382 expanding a string and expecting an integer value. On some systems this
1383 resulted in spurious "integer overflow" errors. Also, it was casting the
1384 result into an int without checking.
138634. Testing for a connection timeout using "timeout_connect" in the retry rules
1387 did not work. The code looks as if it has *never* worked, though it appears
1388 to have been documented since at least releast 1.62. I have made it work.
139035. The "timeout_DNS" error in retry rules, also documented since at least
1391 1.62, also never worked. As it isn't clear exactly what this means, and
1392 clearly it isn't a major issue, I have abolished the feature by treating it
1393 as "timeout", and writing a warning to the main and panic logs.
139536. The display of retry rules for -brt wasn't always showing the error code
1396 correctly.
139837. Added new error conditions to retry rules: timeout_A, timeout_MX,
1399 timeout_connect_A, timeout_connect_MX.
140138. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
1402 to the empty sender.
140439. The daemon was not analysing the content of -oX till after it had closed
1405 stderr and disconnected from the controlling terminal. This meant that any
1406 syntax errors were only noted on the panic log, and the return code from
1407 the command was 0. By re-arranging the code a little, I've made the
1408 decoding happen first, so such errors now appear on stderr, and the return
1409 code is 1. However, the actual setting up of the sockets still happens in
1410 the disconnected process, so errors there are still only recorded on the
1411 panic log.
141340. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
1414 connections (as happens on some IP stacks) was logged at start up time as
1415 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
1416 it from "IPv6 and IPv4", which means that two separate sockets are being
1417 used.
141941. The debug output for gethostbyname2() or getipnodebyname() failures now
1420 says whether AF_INET or AF_INET6 was passed as an argument.
142242. Exiwhat output was messed up when time zones were included in log
1423 timestamps.
142543. Exiwhat now gives more information about the daemon's listening ports,
1426 and whether -tls-on-connect was used.
142844. The "port" option of the smtp transport is now expanded.
143045. A "message" modifier in a "warn" statement in a non-message ACL was being
1431 silently ignored. Now an error message is written to the main and panic
1432 logs.
143446. There's a new ACL modifier called "logwrite" which writes to a log file
1435 as soon as it is encountered.
143747. Added $local_user_uid and $local_user_gid at routing time.
143948. Exim crashed when trying to verify a sender address that was being
1440 rewritten to "<>".
144249. Exim was recognizing only a space character after ".include". It now also
1443 recognizes a tab character.
144550. Fixed several bugs in the Perl script that creates the exim.8 man page by
1446 extracting the relevant information from the specification. The man page no
1447 longer contains scrambled data for the -d option, and I've added a section
1448 at the front about calling Exim under different names.
145051. Added "extra_headers" argument to the "mail" command in filter files.
145252. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
1453 crash.
145553. Installed eximstats 1.29.
145754. Added transport_filter_timeout as a generic transport option.
145955. Exim no longer adds an empty Bcc: header to messages that have no To: or
1460 Cc: header lines. This was required by RFC 822, but it not required by RFC
1461 2822.
146356. Exim used to add From:, Date:, and Message-Id: header lines to any
1464 incoming messages that did not have them. Now it does so only if the
1465 message originates locally, that is, if there is no associated remote host
1466 address. When Resent- header lines are present, this applies to the Resent-
1467 lines rather than the non-Resent- lines.
146957. Drop incoming SMTP connection after too many syntax or protocol errors. The
1470 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
147258. Messages for configuration errors now include the name of the main
1473 configuration file - useful now that there may be more than one file in a
1474 list (.included file names were always shown).
147659. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
1477 for those rare installations that do not start the daemon as root or run it
1478 setuid root. I've cut out the call to initgroups() if the daemon is not
1479 root at that time.
148160. The Exim user and group can now be bound into the binary as text strings
1482 that are looked up at the start of Exim's processing.
148461. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
148662. Added $mailstore_basename variable.
148863. Installed patch to sieve.c from Michael Haardt.
149064. When Exim failed to open the panic log after failing to open the main log,
1491 the original message it was trying to log was written to stderr and debug
1492 output, but if they were not available (the usual case in production), it
1493 was lost. Now it is written to syslog before the two lines that record the
1494 failures to open the logs.
149665. Users' Exim filters run in subprocesses under the user's uid. It is
1497 possible for a "deliver" command or an alias in a "personal" command to
1498 provoke an address rewrite. If logging of address rewriting is configured,
1499 this fails because the process is not running as root or exim. There may be
1500 a better way of dealing with this, but for the moment (because 4.30 needs
1501 to be released), I have disabled address rewrite logging when running a
1502 filter in a non-root, non-exim process.
1505Exim version 4.24
1508 1. The buildconfig auxiliary program wasn't quoting the value set for
1509 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
1510 not defined. This bug was masked in 4.22 by the effect that was fixed in
1511 change 4.23/1.
1513 2. Some messages that were rejected after a message id was allocated were
1514 shown as "incomplete" by exigrep. It no longer does this for messages that
1515 are rejected by local_scan() or the DATA or non-SMTP ACLs.
1517 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
1518 have allow_domain_literals set, the ID did not get logged in the <= line.
1519 Domain literals are now always recognized in Message-ID: header lines.
1521 4. The first argument for a ${extract expansion item is the key name or field
1522 number. Leading and trailing spaces in this item were not being ignored,
1523 causing some misleading effects.
1525 5. When deliver_drop_privilege was set, single queue runner processes started
1526 manually (i.e. by the command "exim -q") or by the daemon (which uses the
1527 same command in the process it spins off) were not dropping privilege.
1529 6. When the daemon running as "exim" started a queue runner, it always
1530 re-executed Exim in the spun-off process. This is a waste of effort when
1531 deliver_drop_privilege is set. The new process now just calls the
1532 queue-runner function directly.
1535Exim version 4.23
1538 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
1541 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
1542 ignored. Though the use of -oP was forcing the writing of a pid file, it
1543 was always written to the default place.
1545 3. If the message "no IP address found for host xxxx" is generated during
1546 incoming verification, it is now followed by identification of the incoming
1547 connection (so you can more easily find what provoked it).
1549 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
1551 5. Added some features to "harden" Exim a bit more against certain attacks:
1553 (a) There is now a build-time option called FIXED_NEVER_USERS that can
1554 be put in Local/Makefile. This is like the never_users runtime option,
1555 but it cannot be overridden. The default setting is "root".
1557 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
1558 prefix string with which any file named in a -C command line option
1559 must start.
1561 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
1562 is retained for -C and -D only if the caller of Exim is root. Without
1563 it, the exim user may also use -C and -D and retain privilege.
1565 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
1566 command line option is disabled.
1568 6. Macro names set by the -D option must start with an upper case letter, just
1569 like macro names defined in the configuration file.
1571 7. Added "dereference=" facility to LDAP.
1573 8. Two instances of the typo "uknown" in the source files are fixed.
1575 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
1576 the Configure-Makefile script screwed up while processing it.
157810. Incorporated PCRE 4.4.
158011. The SMTP synchronization check was not operating right at the start of an
1581 SMTP session. For example, it could not catch a HELO sent before the client
1582 waited for the greeting. There is now a check for outstanding input at the
1583 point when the greeting is written. Because of the duplex, asynchronous
1584 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
1585 way, but not yet received, when the check is performed.
158712. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
1588 on TCP/IP sockets, because this apparently causes some broken clients to
1589 timeout.
159113. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
1592 unchanged) from the Cygwin maintainer.
159414. The code for -bV that shows what is in the binary showed "mbx" when maildir
1595 was supported instead of testing for mbx. Effectively a typo.
159715. The spa authenticator server code was not checking that the input it
1598 received was valid base64.
160016. The debug output line for the "set" modifier in ACLs was not showing the
1601 name of the variable that was being set.
160317. Code tidy: the variable type "vtype_string" was never used. Removed it.
160518. Previously, a reference to $sender_host_name did not cause a DNS reverse
1606 lookup on its own. Something else was needed to trigger the lookup. For
1607 example, a match in host_lookup or the need for a host name in a host list.
1608 Now, if $sender_host_name is referenced and the host name has not yet been
1609 looked up, a lookup is performed. If the lookup fails, the variable remains
1610 empty, and $host_lookup_failed is set to "1".
161219. Added "eqi" as a case-independent comparison operator.
161420. The saslauthd authentication condition could segfault if neither service
1615 nor realm was specified.
161721. If an overflowing value such as "2048M" was set for message_size_limit, the
1618 error message that was logged was misleading, and incoming SMTP
1619 connections were dropped. The message is now more accurate, and temporary
1620 errors are given to SMTP connections.
162222. In some error situations (such as 21 above) Exim rejects all SMTP commands
1623 (except RSET) with a 421 error, until QUIT is received. However, it was
1624 failing to send a response to QUIT.
162623. The HELO ACL was being run before the code for helo_try_verify_hosts,
1627 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
1628 ACL is now run after the helo_try_verify_hosts code.
163024. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
1631 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
1632 case-independent, so other case variants are also recognized). Apparently
1633 some systems use these upper case variants.
163525. If more than two messages were waiting for the same host, and a transport
1636 filter was specified for the transport, Exim sent two messages over the
1637 same TCP/IP connection, and then failed with "socket operation on non-
1638 socket" when it tried to send the third.
164026. Added Exim::debug_write and Exim::log_write for embedded Perl use.
164227. The extern definition of crypt16() in expand.c was not being excluded when
1643 the OS had its own crypt16() function.
164528. Added bounce_return_body as a new option, and bounce_return_size_limit
1646 as a preferred synonym for return_size_limit, both as an option and as an
1647 expansion variable.
164929. Added LIBS=-liconv to OS/Makefile-OSF1.
165130. Changed the default configuration ACL to relax the local part checking rule
1652 for addresses that are not in any local domains. For these addresses,
1653 slashes and pipe symbols are allowed within local parts, but the sequence
1654 /../ is explicitly forbidden.
165631. SPA server authentication was not clearing the challenge buffer before
1657 using it.
165932. log_message in a "warn" ACL statement was writing to the reject log as
1660 well as to the main log, which contradicts the documentation and doesn't
1661 seem right (because no rejection is happening). So I have stopped it.
166333. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
1664 However, I am unable to do any testing of this.
166634. Fixed an infelicity in the appendfile transport. When checking directories
1667 for a mailbox, to see if any needed to be created, it was accidentally
1668 using path names with one or more superfluous leading slashes; tracing
1669 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
167135. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
1672 contents are added to the log line that is written for every discarded
1673 recipient. (Previously a log_message setting was ignored.)
167536. The ${quote: operator now quotes the string if it is empty.
167737. The install script runs exim in order to find its version number. If for
1678 some reason other than non-existence or emptiness, which it checks, it
1679 could not run './exim', it was installing it with an empty version number,
1680 i.e. as "exim-". This error state is now caught, and the installation is
1681 aborted.
168338. An argument was missing from the function that creates an error message
1684 when Exim fails to connect to the socket for saslauthd authentication.
1685 This could cause Exim to crash, or give a corrupted message.
168739. Added isip, isip4, and isip6 to ${if conditions.
168940. The ACL variables $acl_xx are now saved with the message, and can be
1690 accessed later in routers, transports, and filters.
169241. The new lookup type nwildlsearch is like wildlsearch, except that the key
1693 strings in the file are not string-expanded.
169542. If a MAIL command specified a SIZE value that was too large to fit into an
1696 int variable, the check against message_size_limit failed. Such values are
1697 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
1698 day this will have to be increased, but I don't think I want to be around
1699 when emails are that large.
1703Exim version 4.22
1706 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
1707 iconv() is not standard in FreeBSD.
1709 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
1710 IPv6 enabled. The observed symptom was a segmentation fault on return from
1711 the function os_common_find_running_interfaces() in src/os.c.
1713 3. In the check_special_case() function in daemon.c I had used "errno" as an
1714 argument name, which causes warnings on some systems. This was basically a
1715 typo, since it was named "eno" in the comments!
1717 4. The code that waits for the clock to tick (at a resolution of some fraction
1718 of a second) so as to ensure message-id uniqueness was always waiting for
1719 at least one whole tick, when it could have waited for less. [This is
1720 almost certainly not relevant at current processor speeds, where it is
1721 unlikely to ever wait at all. But we try to future-proof.]
1723 5. The function that sleeps for a time interval that includes fractions of a
1724 second contained a race. It did not block SIGALRM between setting the
1725 timer, and suspending (a couple of lines later). If the interval was short
1726 and the sigsuspend() was delayed until after it had expired, the suspension
1727 never ended. On busy systems this could lead to processes getting stuck for
1728 ever.
1730 6. Some uncommon configurations may cause a lookup to happen in a queue runner
1731 process, before it forks any delivery processes. The open lookup caching
1732 mechanism meant that the open file or database connection was passed into
1733 the delivery process. The problem was that delivery processes always tidy
1734 up cached lookup data. This could cause a problem for the next delivery
1735 process started by the queue runner, because the external queue runner
1736 process does not know about the closure. So the next delivery process
1737 still has data in the lookup cache. In the case of a file lookup, there was
1738 no problem because closing a file descriptor in a subprocess doesn't affect
1739 the parent. However, if the lookup was caching a connection to a database,
1740 the connection was closed, and the second delivery process was likely to
1741 see errors such as "PGSQL: query failed: server closed the connection
1742 unexpectedly". The problem has been fixed by closing all cached lookups
1743 in a queue runner before running a delivery process.
1745 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
1746 seem to have the "const" qualifier which it has on other OS. I've
1747 parameterised it.
1749 8. Change 4.21/2 was too strict. It is only if there are two authenticators
1750 *of the same type* (client or server) with the same public name that an
1751 error should be diagnosed.
1753 9. When Exim looked up a host name for an IP address, but failed to find the
1754 original IP address when looking up the host name (a safety check), it
1755 output the message "<ip address> does not match any IP for NULL", which was
1756 confusing, to say the least. The bug was that the host name should have
1757 appeared instead of "NULL".
175910. Since release 3.03, if Exim is called by a uid other than root or the Exim
1760 user that is built into the binary, and the -C or -D options is used, root
1761 privilege is dropped before the configuration file is read. In addition,
1762 logging is switched to stderr instead of the normal log files. If the
1763 configuration then re-defines the Exim user, the unprivileged environment
1764 is probably not what is expected, so Exim logs a panic warning message (but
1765 proceeds).
1767 However, if deliver_drop_privilege is set, the unprivileged state may well
1768 be exactly what is intended, so the warning has been cut out in that case,
1769 and Exim is allowed to try to write to its normal log files.
1772Exim version 4.21
1775 1. smtp_return_error_details was not giving details for temporary sender
1776 or receiver verification errors.
1778 2. Diagnose a configuration error if two authenticators have the same public
1779 name.
1781 3. Exim used not to create the message log file for a message until the first
1782 delivery attempt. This could be confusing when incoming messages were held
1783 for policy or load reasons. The message log file is now created at the time
1784 the message is received, and an initial "Received" line is written to it.
1786 4. The automatically generated man page for command line options had a minor
1787 bug that caused no ill effects; however, a more serious problem was that
1788 the procedure for building the man page automatically didn't always
1789 operate. Consequently, release 4.20 contains an out-of-date version. This
1790 shouldn't happen again.
1792 5. When building Exim with embedded Perl support, the script that builds the
1793 Makefile was calling 'perl' to find its compile-time parameters, ignoring
1794 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
1796 6. The freeze_tell option was not being used for messages that were frozen on
1797 arrival, either by an ACL or by local_scan().
1799 7. Added the smtp_incomplete_transaction log selector.
1801 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
1802 was accepting AUTH without a new EHLO.
1804 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
1805 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
1806 RFC.
180810. Logging of TCP/IP connections (when configured) now happens in the main
1809 daemon process instead of the child process, so that the TCP/IP connection
1810 count is more accurate (but it can never be perfect).
181211. The use of "drop" in a nested ACL was not being handled correctly in the
1813 outer ACL. Now, if condition failure induced by the nested "drop" causes
1814 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
1815 or "require"), the connection is dropped.
181712. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
1818 that yield "discard" can now be used with an "accept" or a "discard" verb,
1819 but an error is generated for any others (because I can't see a useful way
1820 to define what should happen).
182213. When an ACL is read dynamically from a file (or anywhere else), the lines
1823 are now processed in the same way as lines in the Exim configuration file.
1824 In particular, continuation lines are supported.
182614. Added the "dnslists = a.b.c!=n.n.n.n" feature.
182815. Added -ti meaning -t -i.
183016. Check for letters, digits, hyphens, and dots in the names of dnslist
1831 domains, and warn by logging if others are found.
183317. At least on BSD, alignment is not guarenteed for the array of ifreq's
1834 returned from GIFCONF when Exim is trying to find the list of interfaces on
1835 a host. The code in os.c has been modified to copy each ifreq to an aligned
1836 structure in all cases.
1838 Also, in some cases, the returned ifreq's were being copied to a 'struct
1839 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
1840 means the last couple of bytes of an IPv6 address could be chopped if the
1841 ifreq contained only a normal sockaddr (14 bytes storage).
184318. Named domain lists were not supported in the hosts_treat_as_local option.
1844 An entry such as +xxxx was not recognized, and was treated as a literal
1845 domain name.
184719. Ensure that header lines added by a DATA ACL are included in the reject log
1848 if the ACL subsequently rejects the message.
185020. Upgrade the utility script to use Digest::MD5 instead of just
1851 MD5 (which is deprecated).
185321. When testing a filter file using -bf, Exim was writing a message when it
1854 took the sender from a "From " line in the message, but it was not doing so
1855 when it took $return_path from a Return-Path: header line. It now does.
185722. If the contents of a "message" modifier for a "warn" ACL verb do not begin
1858 with a valid header line field name (a series of printing characters
1859 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
186123. Changed "disc" in the source to "disk" to conform to the documentation and
1862 the book and for uniformity.
186424. Ignore Sendmail's -Ooption=value command line item.
186625. When execve() failed while trying to run a command in a pipe transport,
1867 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
1868 could be confused with a return value of 69 from the command itself. This
1869 has been changed to 127, the value the shell returns if it is asked to run
1870 a non-existent command. The wording for the related log line suggests a
1871 non-existent command as the problem.
187326. If received_header_text expands to an empty string, do not add a Received:
1874 header line to the message. (Well, it adds a token one on the spool, but
1875 marks it "old" so that it doesn't get used or transmitted.)
187727. Installed eximstats 1.28 (addition of -nt option).
187928. There was no check for failure on the call to getsockname() in the daemon
1880 code. This can fail if there is a shortage of resources on the system, with
1881 ENOMEM, for example. A temporary error is now given on failure.
188329. Contrary to the C standard, it seems that in some environments, the
1884 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
1885 program. Exim now does this explicitly; it affects the formatting of
1886 timestamps using strftime().
188830. If exiqsumm was given junk data, it threw up some uninitialized variable
1889 complaints. I've now initialized all the variables, to avoid this.
189132. Header lines added by a system filter were not being "seen" during
1892 transport-time rewrites.
189433. The info_callback() function passed to OpenSSL is set up with type void
1895 (*)(SSL *, int, int), as described somewhere. However, when calling the
1896 function (actually a macro) that sets it up, the type void(*)() is
1897 expected. I've put in a cast to prevent warnings from picky compilers.
189934. If a DNS black list lookup found a CNAME record, but there were no A
1900 records associated with the domain it pointed at, Exim crashed.
190235. If a DNS black list lookup returned more than one A record, Exim ignored
1903 all but the first. It now scans all returned addresses if a particular IP
1904 value is being sought. In this situation, the contents of the
1905 $dnslist_value variable are a list of all the addresses, separated by a
1906 comma and a space.
190836. Tightened up the rules for host name lookups using reverse DNS. Exim used
1909 to accept a host name and all its aliases if the forward lookup for any of
1910 them yielded the IP address of the incoming connection. Now it accepts only
1911 those names whose forward lookup yields the correct IP address. Any other
1912 names are discarded. This closes a loophole whereby a rogue DNS
1913 administrator could create reverse DNS records to break through a
1914 wildcarded host restriction in an ACL.
191637. If a user filter or a system filter that ran in a subprocess used any of
1917 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
1918 the wrong values were passed to the pipe command ($thisaddress had the
1919 value of $0, $0 had the value of $1, etc). This bug was introduced by
1920 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
192238. Improved the line breaking for long SMTP error messages from ACLs.
1923 Previously, if there was no break point between 40 and 75 characters, Exim
1924 left the rest of the message alone. Two changes have been made: (a) I've
1925 reduced the minimum length to 35 characters; (b) if it can't find a break
1926 point between 35 and 75 characters, it looks ahead and uses the first one
1927 that it finds. This may give the occasional overlong line, but at least the
1928 remaining text gets split now.
193039. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
1931 file descriptors might be low, and that setting 1000 would always raise it.
1932 It turns out that in some environments, the limit is already over 1000 and
1933 that lowering it causes trouble. So now Exim takes care not to decrease it.
193540. When delivering a message, the value of $return_path is set to $sender_
1936 address at the start of routing (routers may change the value). By an
1937 oversight, this default was not being set up when an address was tested by
1938 -bt or -bv, which affected the outcome if any router or filter referred to
1939 $return_path.
194141. The idea of the "warn" ACL verb is that it adds a header or writes to the
1942 log only when "message" or "log_message" are set. However, if one of the
1943 conditions was an address verification, or a call to a nested ACL, the
1944 messages generated by the underlying test were being passed through. This
1945 no longer happens. The underlying message is available in $acl_verify_
1946 message for both "message" and "log_message" expansions, so it can be
1947 passed through if needed.
194942. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
1950 new expansion $bh_ to give the encoded byte string without charset
1951 translation. Translation happens only if iconv() is available; HAVE_ICONV
1952 indicates this at build time. HEADERS_CHARSET gives the charset to
1953 translate to; headers_charset can change it in the configuration, and
1954 "headers charset" can change it in an individual filter file.
195643. Now that we have a default RFC 2047 charset (see above), the code in Exim
1957 that creates RFC 2047 encoded "words" labels them as that charset instead
1958 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
1959 expansion operator; (ii) when Exim creates a From: line for a local
1960 message; (iii) when a header line is rewritten to include a "phrase" part.
196244. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
1963 buggy, causing it to skip the first lines of messages whose message ID
1964 ended in 'D'. This would not have bitten before Exim release 4.14, because
1965 message IDs were unlikely to end in 'D' before then. The effect was to have
1966 incorrect size information for certain domains.
196845. #include "config.h" was missing at the start of the crypt16.c module. This
1969 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
1970 noticed.
197246. If there was a timeout during a "random" callout check, Exim treated it as
1973 a failure of the random address, and carried on sending RSET and the real
1974 address. If the delay was just some slowness somewhere, the response to the
1975 original RCPT would be taken as a response to RSET and so on, causing
1976 mayhem of various kinds.
197847. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
1979 when I implemented it. It didn't allow for the fact that some option values
1980 may legitimatetly be negative (e.g. size_addition), and it didn't even do
1981 the right test for positive values.
198348. Domain names in DNS records are case-independent. Exim always looks them up
1984 in lower case. Some resolvers return domain names in exactly the case they
1985 appear in the zone file, that is, they may contain uppercase letters. Not
1986 all resolvers do this - some return always lower case. Exim was treating a
1987 change of case by a resolver as a change of domain, similar to a widening
1988 of a domain abbreviation. This triggered its re-routing code and so it was
1989 trying to route what was effectively the same domain again. This normally
1990 caused routing to fail (because the router wouldn't handle the domain
1991 twice). Now Exim checks for this case specially, and just changes the
1992 casing of the domain that it ultimately uses when it transmits the message
1993 envelope.
199549. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
1996 module.
199850. If a filter generated a file delivery with a non-absolute name (possible if
1999 no home directory exists for the router), the forbid_file option was not
2000 forbidding it.
200251. Added '&' feature to dnslists, to provide bit mask matching in addition to
2003 the existing equality matching.
200552. Exim was using ints instead of ino_t variables in some places where it was
2006 dealing with inode numbers.
200853. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
2009 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
2010 called TMPDIR, and if it finds it is different, it changes its value.
201254. The smtp_printf() function is now made available to local_scan() so
2013 additional output lines can be written before returning. There is also an
2014 smtp_fflush() function to enable the detection of a dropped connection.
2015 The variables smtp_input and smtp_batched_input are exported to
2016 local_scan().
201855. Changed the default runtime configuration: the message "Unknown user"
2019 has been removed from the ACL, and instead placed on the localuser router,
2020 using the cannot_route_message feature. This means that any verification
2021 failures that generate their own messages won't get overridden. Similarly,
2022 the "Unrouteable address" message that was in the ACL for unverifiable
2023 relay addresses has also been removed.
202556. Added hosts_avoid_esmtp to the smtp transport.
202757. The exicyclog script was not checking for the esoteric option
2028 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2029 will work only if exicyclog is run under the appropriate euid.
203158. Following a discussion on the list, the rules by which Exim recognises line
2032 endings on incoming messages have been changed. The -dropcr and drop_cr
2033 options are now no-ops, retained only for backwards compatibility. The
2034 following line terminators are recognized: LF CRLF CR. However, special
2035 processing applies to CR:
2037 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2038 nor a local message in the state where . is a terminator.
2040 (ii) If a bare CR is encountered in a header line, an extra space is added
2041 after the line terminator so as not to end the header. The reasoning
2042 behind this is that bare CRs in header lines are most likely either
2043 to be mistakes, or people trying to play silly games.
204559. The size of a message, as listed by "-bp" or in the Exim monitor window,
2046 was being incorrectly given as 18 bytes larger than it should have been.
2047 This is a VOB (very old bug).
204960. This may never have affected anything current, but just in case it has:
2050 When the local host is found other than at the start of a list of hosts,
2051 the local host, those with the same MX, and any that follow, are discarded.
2052 When the list in question was part of a longer list of hosts, the following
2053 hosts (not currently being processed) were also being discarded. This no
2054 longer happens. I'm not sure if this situation could ever has previously
2055 arisen.
205761. Added the "/MX" feature to lists of hosts in the manualroute and query
2058 program routers.
206062. Whenever Exim generates a new message, it now adds an Auto-Submitted:
2061 header. This is something that is recommended in a new Internet Draft, and
2062 is something that is documented as being done by Sendmail. There are two
2063 possible values. For messages generated by the autoreply transport, Exim
2064 adds:
2066 Auto-Submitted: auto-replied
2068 whereas for all other generated messages (e.g. bounces) it adds
2070 Auto-Submitted: auto-generated
207263. The "personal" condition in filters now includes a test for the
2073 Auto-Submitted: header. If it contains the string "auto-" the message it
2074 not considered personal.
207664. Added rcpt_include_affixes as a generic transport option.
207865. Added queue_only_override (default true).
208066. Added the syslog_duplication option.
208267. If what should have been the first header line of a message consisted of
2083 a space followed by a colon, Exim was mis-interpreting it as a header line.
2084 It isn't of course - it is syntactically invalid and should therefore be
2085 treated as the start of the message body. The misbehaviour could have
2086 caused a number of strange effects, including loss of data in subsequent
2087 header lines, and spool format errors.
208968. Formerly, the AUTH parameter on a MAIL command was trusted only if the
2090 client host had authenticated. This control can now be exercised by an ACL
2091 for more flexibility.
209369. By default, callouts do not happen when testing with -bh. There is now a
2094 variant, -bhc, which does actually run the callout code, including
2095 consulting and updating the callout cache.
209770. Added support for saslauthd authentication, courtesy of Alexander
2098 Sabourenkov.
210071. If statvfs() failed on the spool or log directories while checking their
2101 size for availability, Exim confusingly gave the error "space shortage".
2102 Furthermore, in debugging mode it crashed with a floating point exception.
2103 These checks are done if check_{spool,log}_{space,inodes} are set, and when
2104 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
2105 serious problem, Exim now writes to the main and panic logs when this
2106 happens, with details of the failure. It then refuses to accept the
2107 incoming message, giving the message "spool directory problem" or "log
2108 directory problem" with a 421 code for SMTP messages.
211072. When Exim is about to re-exec itself, it ensures that the file descriptors
2111 0, 1, and 2 exist, because some OS complain for execs without them (see
2112 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
2113 descriptors. However, the code omitted to check that the open succeeded,
2114 causing mysterious errors if for some reason the permissions on /dev/null
2115 got screwed. Now Exim writes a message to the main and panic logs, and
2116 bombs out if it can't open /dev/null.
211873. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
2119 interact so that it is all more flexible. It is supposed to remain
2120 backwards compatible. Also added extra_local_interfaces.
212274. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
2123 to bomb out with an assertion failure - to the client this appears as a
2124 connection drop. This problem occurs in the part of the code that was taken
2125 from the Samba project. Fortunately, the assertion is in a very simple
2126 function, so I have fixed this by reproducing the function inline in the
2127 one place where it is called, and arranging for authentication to fail
2128 instead of killing the process with assert().
213075. The SPA client code was not working when the server requested OEM rather
2131 than Unicode encoding.
213376. Added code to make require_files with a specific uid setting more usable in
2134 the case where statting the file as root fails - usually a non-root-mounted
2135 NFS file system. When this happens and the failure is EACCES, Exim now
2136 forks a subprocess and does the per-uid checking as the relevant uid.
213877. Added process_log_path.
214078. If log_file_path was not explicitly set, a setting of check_log_space or
2141 check_log_inodes was ignored.
214379. If a space check for the spool or log partitions fails, the incident is now
2144 logged. Of course, in the latter case the data may get lost...
214680. Added the %p formatting code to string_format() so that it can be used to
2147 print addresses in debug_print(). Adjusted all the address printing in the
2148 debugging in store.c to use %p rather than %d.
215081. There was a concern that a line of code in smtp_in.c could overflow a
2151 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
2152 initially expressed, the concern was not well-founded, because trailing
2153 spaces are removed early. However, if the trailing spaces were followed by
2154 a NULL, they did not get removed, so the overflow was possible. Two fixes
2155 were applied:
2157 (a) I re-wrote the offending code in a cleaner fashion.
2158 (b) If an incoming SMTP command contains a NULL character, it is rejected
2159 as invalid.
216182. When Exim changes uid/gid to the Exim user at daemon start time, it now
2162 runs initgroups(), so that if the Exim user is in any additional groups,
2163 they will be used during message reception.
2166Exim version 4.20
2169The change log for 4.20 and earlier releases has been archived.