1. Tidies to get rid of compiler warnings from the merged Exiscan files.
[exim.git] / doc / doc-txt / ChangeLog
f7b63901 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.43 2004/12/17 14:52:44 ph10 Exp $
3Change log file for Exim from version 4.21
f7b63901 6Exim version 4.50
9 1. Minor wording change to the doc/README.SIEVE file.
139059f6 11 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 12 computation of the current number of files was incorrect.
495ae4b0 13
14 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
15 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
16 place.
18 4. Give more explanation in the error message when the command for a transport
19 filter fails to execute.
21 5. There are several places where Exim runs a non-Exim command in a
22 subprocess. The SIGUSR1 signal should be disabled for these processes. This
23 was being done only for the command run by the queryprogram router. It is
24 now done for all such subprocesses. The other cases are: ${run, transport
25 filters, and the commands run by the lmtp and pipe transports.
27 6. Added CONFIGURE_GROUP build-time option.
29 7. Some older OS have a limit of 256 on the maximum number of file
30 descriptors. Exim was using setrlimit() to set 1000 as a large value
31 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
32 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 33
34 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
35 was an oversight, and furthermore, ever since the addition of extra
36 controls (e.g. 4.43/32), the checks on when to allow different forms of
37 "control" were broken. There should now be diagnostics for all cases when a
38 control that does not make sense is encountered.
40 9. Added the /retain_sender option to "control=submission".
4210. $recipients is now available in the predata ACL (oversight).
4411. Tidy the search cache before the fork to do a delivery from a message
45 received from the command line. Otherwise the child will trigger a lookup
46 failure and thereby defer the delivery if it tries to use (for example) a
47 cached ldap connection that the parent has called unbind on.
4912. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
50 of $address_data from the recipient verification was clobbered by the
51 sender verification.
5313. The value of address_data from a sender verification is now available in
54 $sender_address_data in subsequent conditions in the ACL statement.
5614. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5815. Added a new option "connect=<time>" to callout options, to set a different
59 connection timeout.
6116. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
62 was its contents. (It was OK if the option was not defined at all.)
6417. A "Completed" log line is now written for messages that are removed from
65 the spool by the -Mrm option.
6718. New variables $sender_verify_failure and $recipient_verify_failure contain
68 information about exactly what failed.
7019. Added -dd to debug only the daemon process.
7220. Incorporated Michael Haardt's patch to ldap.c for improving the way it
73 handles timeouts, both on the server side and network timeouts. Renamed the
74 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
7621. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
77 It is now set to "smtps".
7922. $host_address is now set to the target address during the checking of
80 ignore_target_hosts.
8223. When checking ignore_target_hosts for an ipliteral router, no host name was
83 being passed; this would have caused $sender_host_name to have been used if
84 matching the list had actually called for a host name (not very likely,
85 since this list is usually IP addresses). A host name is now passed as
86 "[x.x.x.x]".
8824. Changed the calls that set up the SIGCHLD handler in the daemon to use the
89 code that specifies a non-restarting handler (typically sigaction() in
90 modern systems) in an attempt to fix a rare and obscure crash bug.
9225. Narrowed the window for a race in the daemon that could cause it to ignore
93 SIGCHLD signals. This is not a major problem, because they are used only to
94 wake it up if nothing else does.
9626. A malformed maildirsize file could cause Exim to calculate negative values
97 for the mailbox size or file count. Odd effects could occur as a result.
98 The maildirsize information is now recalculated if the size or filecount
99 end up negative.
10127. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
102 support for a long time. Removed HAVE_SYS_VFS_H.
10428. Installed the latest version of exipick from John Jetmore.
10629. In an address list, if the pattern was not a regular expression, an empty
107 subject address (from a bounce message) matched only if the pattern was an
108 empty string. Non-empty patterns were not even tested. This was the wrong
109 because it is perfectly reasonable to use an empty address as part of a
110 database query. An empty address is now tested by patterns that are
111 lookups. However, all the other forms of pattern expect the subject to
112 contain a local part and a domain, and therefore, for them, an empty
113 address still always fails if the pattern is not itself empty.
11530. Exim went into a mad DNS loop when attempting to do a callout where the
116 host was specified on an smtp transport, and looking it up yielded more
117 than one IP address.
11931. Re-factored the code for checking spool and log partition space into a
120 function that finds that data and another that does the check. The former
121 is then used to implement four new variables: $spool_space, $log_space,
122 $spool_inodes, and $log_inodes.
12432. The RFC2047 encoding function was originally intended for short strings
125 such as real names; it was not keeping to the 75-character limit for
126 encoded words that the RFC imposes. It now respects the limit, and
127 generates multiple encoded words if necessary. To be on the safe side, I
128 have increased the buffer size for the ${rfc2047: expansion operator from
129 1024 to 2048 bytes.
13133. It is now permitted to omit both strings after an "if" condition; if the
132 condition is true, the result is "true". As before, when the second string
133 is omitted, a false condition yields an empty string. This makes it less
134 cumbersome to write custom ACL and router conditions.
13634. Failure to deliver a bounce message always caused it to be frozen, even if
137 there was an errors_to setting on the router. The errors_to setting is now
138 respected.
14035. If an IPv6 address is given for -bh or -bhc, it is now converted to the
141 canonical form (fully expanded) before being placed in
142 $sender_host_address.
14436. The table in the code that translates DNS record types into text (T_A to
145 "A" for instance) was missing entries for NS and CNAME. It is just possible
146 that this could have caused confusion if both these types were looked up
147 for the same domain, because the text type is used as part of Exim's
148 per-process caching. But the chance of anyone hitting this buglet seems
149 very small.
15137. The dnsdb lookup has been extended in a number of ways.
153 (1) There is a new type, "zns", which walks up the domain tree until it
154 finds some nameserver records. It should be used with care.
156 (2) There is a new type, "mxh", which is like "mx" except that it returns
157 just the host names, not the priorities.
159 (3) It is now possible to give a list of domains (or IP addresses) to be
160 looked up.
ea3bc19b 162 (4) It is now possible to specify the separator character for use when
7bb56e1f 163 multiple records are returned.
33397d19 164
16538. The dnslists ACL condition has been extended: it is now possible to supply
166 a list of IP addresses and/or domains to be looked up in a particular DNS
167 domain.
16939. Added log_selector=+queue_time_overall.
17140. When running the queue in the test harness, wait just a tad after forking a
172 delivery process, to get repeatability of debugging output.
17441. Include certificate and key file names in error message when GnuTLS fails
175 to set them up, because the GnuTLS error message doesn't include the name
176 of the failing file when there is a problem reading it.
17842. Allow both -bf and -bF in the same test run.
18043. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
18244. The "Exiscan patch" is now merged into the mainline Exim source.
18445. Sometimes the final signoff response after QUIT could fail to get
185 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
186 < 0 before doing a fflush(). This bug looks as though it goes back to the
187 introduction of TLS in release 3.20, but "sometimes" must have been rare
188 because the tests only now provoked it.
191Exim version 4.43
194 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
195 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
196 Now it returns an integer. A number of calls were still expecting a T/F
197 return. Fortuitously, in all cases, the tests worked in OK situations,
198 which is the norm. However, things would have gone wrong on any write
199 failures on the smtp file descriptor. This function is used when sending
200 messages over SMTP and also when doing verify callouts.
202 2. When Exim is called to do synchronous delivery of a locally submitted
203 message (the -odf or -odi options), it no longer closes stderr before doing
204 the delivery.
206 3. Implemented the mua_wrapper option.
208 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
210 5. Implemented the functions header_remove(), header_testname(),
211 header_add_at_position(), and receive_remove_recipient(), and exported them
212 to local_scan().
214 6. If an ACL "warn" statement specified the addition of headers, Exim already
215 inserted X-ACL-Warn: at the start if there was no header name. However, it
216 was not making this test for the second and subsequent header lines if
217 there were newlines in the string. This meant that an invalid header could
218 be inserted if Exim was badly configured.
220 7. Allow an ACL "warn" statement to add header lines at the start or after all
221 the Received: headers, as well as at the end.
223 8. Added the rcpt_4xx retry error code.
225 9. Added postmaster_mailfrom=xxx to callout verification option.
22710. Added mailfrom=xxxx to the callout verification option, for verify=
228 header_sender only.
23011. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
231 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
23312. Inserted some casts to stop certain compilers warning when using pointer
234 differences as field lengths or precisions in printf-type calls (mostly
235 affecting debugging statements).
23713. Added optional readline() support for -be (dynamically loaded).
23914. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
240 same clock tick as a message's arrival, so that its received time was the
241 same as the "first fail" time on the retry record, and that message
242 remained on the queue past the ultimate address timeout, every queue runner
243 would try a delivery (because it was past the ultimate address timeout) but
244 after another failure, the ultimate address timeout, which should have then
245 bounced the address, did not kick in. This was a "< instead of <=" error;
246 in most cases the first failure would have been in the next clock tick
247 after the received time, and all would be well.
24915. The special items beginning with @ in domain lists (e.g. @mx_any) were not
250 being recognized when the domain list was tested by the match_domain
251 condition in an expansion string.
25316. Added the ${str2b64: operator.
25517. Exim was always calling setrlimit() to set a large limit for the number of
256 processes, without checking whether the existing limit was already
257 adequate. (It did check for the limit on file descriptors.) Furthermore,
258 errors from getrlimit() and setrlimit() were being ignored. Now they are
259 logged to the main and panic logs, but Exim does carry on, to try to do its
260 job under whatever limits there are.
26218. Imported PCRE 5.0.
26419. Trivial typo in log message " temporarily refused connection" (the leading
265 space).
26720. If the log selector return_path_on_delivery was set and an address was
268 redirected to /dev/null, the delivery process crashed because it assumed
269 that a return path would always be set for a "successful" delivery. In this
270 case, the whole delivery is bypassed as an optimization, and therefore no
271 return path is set.
27321. Internal re-arrangement: the function for sending a challenge and reading
274 a response while authentication was assuming a zero-terminated challenge
275 string. It's now changed to take a pointer and a length, to allow for
276 binary data in such strings.
27822. Added the cyrus_sasl authenticator (code supplied by MBM).
28023. Exim was not respecting finduser_retries when seeking the login of the
281 uid under which it was called; it was always trying 10 times. (The default
282 setting of finduser_retries is zero.) Also, it was sleeping after the final
283 failure, which is pointless.
28524. Implemented tls_on_connect_ports.
28725. Implemented acl_smtp_predata.
28926. If the domain in control=submission is set empty, Exim assumes that the
290 authenticated id is a complete email address when it generates From: or
291 Sender: header lines.
29327. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
294 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
295 chown and chgrp in /bin and hostname in /usr/bin.
29728. Exim was keeping the "process log" file open after each use, just as it
298 does for the main log. This opens the possibility of it remaining open for
299 long periods when the USR1 signal hits a daemon. Occasional processlog
300 errors were reported, that could have been caused by this. Anyway, it seems
301 much more sensible not to leave this file open at all, so that is what now
302 happens.
30429. The long-running daemon process does not normally write to the log once it
305 has entered its main loop, and it closes the log before doing so. This is
306 so that log files can straightforwardly be renamed and moved. However,
307 there are a couple of unusual error situations where the daemon does write
308 log entries, and I had neglected to close the log afterwards.
31030. The text of an SMTP error response that was received during a remote
311 delivery was being truncated at 512 bytes. This is too short for some of
312 the long messages that one sometimes sees. I've increased the limit to
313 1024.
31531. It is now possible to make retry rules that apply only when a message has a
316 specific sender, in particular, an empty sender.
31832. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
319 it possible to be selective about when SMTP synchronization is enforced.
32133. Added "control = caseful_local_part" and "control = "caselower_local_part".
32332. Implemented hosts_connection_nolog.
32533. Added an ACL for QUIT.
32734. Setting "delay_warning=" to disable warnings was not working; it gave a
328 syntax error.
33035. Added mailbox_size and mailbox_filecount to appendfile.
33236. Added control = no_multiline_responses to ACLs.
33437. There was a bug in the logic of the code that waits for the clock to tick
335 in the case where the clock went backwards by a substantial amount such
336 that the microsecond fraction of "now" was more than the microsecond
337 fraction of "then" (but the whole seconds number was less).
33938. Added support for the libradius Radius client library this is found on
340 FreeBSD (previously only the radiusclient library was supported).
343Exim version 4.42
346 1. When certain lookups returned multiple values in the form name=value, the
347 quoting of the values was not always being done properly. Specifically:
348 (a) If the value started with a double quote, but contained no whitespace,
349 it was not quoted.
350 (b) If the value contained whitespace other than a space character (i.e.
351 tabs or newlines or carriage returns) it was not quoted.
352 This fix has been applied to the mysql and pgsql lookups by writing a
353 separate quoting function and calling it from the lookup code. The fix
354 should probably also be applied to nisplus, ibase and oracle lookups, but
355 since I cannot test any of those, I have not disturbed their existing code.
357 2. A hit in the callout cache for a specific address caused a log line with no
358 reason for rejecting RCPT. Now it says "Previous (cached) callout
359 verification failure".
361 3. There was an off-by-one bug in the queryprogram router. An over-long
362 return line was truncated at 256 instead of 255 characters, thereby
363 overflowing its buffer with the terminating zero. As well as fixing this, I
364 have increased the buffer size to 1024 (and made a note to document this).
366 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
367 when Exim is waiting for an SMTP response from a remote server, Exim
368 restarts its select() call on the socket, thereby resetting its timeout.
369 This is not a problem when such interrupts are rare. Somebody set up a cron
370 job to run exiwhat every 2 minutes, which is less than the normal select()
371 timeout (5 or 10 minutes). This meant that the select() timeout never
372 kicked in because it was always reset. I have fixed this by comparing the
373 time when an interrupt arrives with the time at the start of the first call
374 to select(). If more time than the timeout has elapsed, the interrupt is
375 treated as a timeout.
377 5. Some internal re-factoring in preparation for the addition of Sieve
378 extensions (by MH). In particular, the "personal" test is moved to a
379 separate function, and given an option for scanning Cc: and Bcc: (which is
380 not set for Exim filters).
382 6. When Exim created an email address using the login of the caller as the
383 local part (e.g. when creating a From: or Sender: header line), it was not
384 quoting the local part when it contained special characters such as @.
386 7. Installed new OpenBSD configuration files.
388 8. Reworded some messages for syntax errors in "and" and "or" conditions to
389 try to make them clearer.
391 9. Callout options, other than the timeout value, were being ignored when
392 verifying sender addresses in header lines. For example, when using
394 verify = header_sender/callout=no_cache
396 the cache was (incorrectly) being used.
39810. Added a missing instance of ${EXE} to the exim_install script; this affects
399 only the Cygwin environment.
40111. When return_path_on_delivery was set as a log selector, if different remote
402 addresses in the same message used different return paths and parallel
403 remote delivery occurred, the wrong values would sometimes be logged.
404 (Whenever a remote delivery process finished, the return path value from
405 the most recently started remote delivery process was logged.)
40712. RFC 3848 specifies standard names for the "with" phrase in Received: header
408 lines when AUTH and/or TLS are in use. This is the "received protocol"
409 field. Exim used to use "asmtp" for authenticated SMTP, without any
410 indication (in the protocol name) for TLS use. Now it follows the RFC and
411 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
412 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
413 names appear in log lines as well as in Received: header lines.
41513. Installed MH's patches for Sieve to add the "copy" and "vacation"
416 extensions, and comparison tests, and to fix some bugs.
41814. Changes to the "personal" filter test:
420 (1) The test was buggy in that it was just doing the equivalent of
421 "contains" tests on header lines. For example, if a user's address was
422 anne@some.where, the "personal" test would incorrectly be true for
424 To: susanne@some.where
426 This test is now done by extracting each address from the header in turn,
427 and checking the entire address. Other tests that are part of "personal"
428 are now done using regular expressions (for example, to check local parts
429 of addresses in From: header lines).
431 (2) The list of non-personal local parts in From: addresses has been
432 extended to include "listserv", "majordomo", "*-request", and "owner-*",
433 taken from the Sieve specification recommendations.
435 (3) If the message contains any header line starting with "List-" it is
436 treated as non-personal.
438 (4) The test for "circular" in the Subject: header line has been removed
439 because it now seems ill-conceived.
44115. Minor typos in src/EDITME comments corrected.
44316. Installed latest exipick from John Jetmore.
44517. If headers_add on a router specified a text string that was too long for
446 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
447 of string_sprintf() is now avoided.
44918. $message_body_size was not set (it was always zero) when running the DATA
450 ACL and the local_scan() function.
45219. For the "mail" command in an Exim filter, no default was being set for
453 the once_repeat time, causing a random time value to be used if "once" was
454 specified. (If the value happened to be <= 0, no repeat happened.) The
455 default is now 0s, meaning "never repeat". The "vacation" command was OK
456 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
457 (I found it when inspecting the code).
45920. There is now an overall timeout for performing a callout verification. It
460 defaults to 4 times the callout timeout, which applies to individual SMTP
461 commands during the callout. The overall timeout applies when there is more
462 than one host that can be tried. The timeout is checked before trying the
463 next host. This prevents very long delays if there are a large number of
464 hosts and all are timing out (e.g. when the network connections are timing
465 out). The value of the overall timeout can be changed by specifying an
466 additional sub-option for "callout", called "maxwait". For example:
468 verify = sender/callout=5s,maxwait=20s
47021. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
471 to the end before writing, but this should make it even safer).
47322. Exim was forgetting that it had advertised PIPELINING for the second and
474 subsequent messages on an SMTP connection. It was also not resetting its
475 memory on STARTTLS and an internal HELO.
47723. When Exim logs an SMTP synchronization error within a session, it now
478 records whether PIPELINING has been advertised or not.
48024. Added 3 instances of "(long int)" casts to time_t variables that were being
481 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
482 rather than long int.
48425. Installed the latest Cygwin configuration files from the Cygwin maintainer.
48626. Added the never_mail option to autoreply.
489Exim version 4.41
492 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
493 crash if the getsockname() call failed; this can happen if a connection is
494 closed very soon after it is established. The problem was simply in the
495 order in which certain operations were done, causing Exim to try to write
496 to the SMTP stream before it had set up the file descriptor. The bug has
497 been fixed by making things happen in the correct order.
500Exim version 4.40
503 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
504 before the connection was closed, thus losing the rejection response.
506 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
507 some early Solaris releases, but causes trouble in current releases where
508 socklen_t is defined.
510 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
511 always exist.
513 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
514 configured.
516 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
517 that releasing the top of it at the end releases what was used for sub-
518 expansions (unless the block got too big). However, discard this block if
519 the first thing is a variable or header, so that we can use its block when
520 it is dynamic (useful for very large $message_headers, for example).
522 6. Lookups now cache *every* query, not just the most recent. A new, separate
523 store pool is used for this. It can be recovered when all lookup caches are
524 flushed. Lookups now release memory at the end of their result strings.
525 This has involved some general refactoring of the lookup sources.
527 7. Some code has been added to the store_xxx() functions to reduce the amount
528 of flapping under certain conditions.
530 8. log_incoming_interface used to affect only the <= reception log lines. Now
531 it causes the local interface and port to be added to several more SMTP log
532 lines, for example "SMTP connection from", and rejection lines.
534 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
53610. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
53811. If $host_data was set by virtue of a hosts lookup in an ACL, its value
539 could be overwritten at the end of the current message (or the start of a
540 new message if it was set in a HELO ACL). The value is now preserved for
541 the duration of the SMTP connection.
54312. If a transport had a headers_rewrite setting, and a matching header line
544 contained an unqualified address, that address was qualified, even if it
545 did not match any rewriting rules. The underlying bug was that the values
546 of the flags that permit the existence of unqualified sender and recipient
547 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
548 non-local messages, and by -bnq for local messages) were not being
549 preserved with the message after it was received.
55113. When Exim was logging an SMTP synchronization error, it could sometimes log
552 "next input=" as part of the text comprising the host identity instead of
553 the correct text. The code was using the same buffer for two different
554 strings. However, depending on which order the printing function evaluated
555 its arguments, the bug did not always show up. Under Linux, for example, my
556 test suite worked just fine.
55814. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
559 doesn't work with some older versions of Perl. It has been changed to "my",
560 which in any case is probably the better facility to use.
56215. A really picky compiler found some instances of statements for creating
563 error messages that either had too many or two few arguments for the format
564 string.
56616. The size of the buffer for calls to the DNS resolver has been increased
567 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
568 for addresses that have a lot of PTR records. This alleviates a problem; it
569 does not fully solve it.
57117. A dnsdb lookup for PTR records that receives more data than will fit in the
572 buffer now truncates the list and logs the incident, which is the same
573 action as happens when Exim is looking up a host name and its aliases.
574 Previously in this situation something unpredictable would happen;
575 sometimes it was "internal error: store_reset failed".
57718. If a server dropped the connection unexpectedly when an Exim client was
578 using GnuTLS and trying to read a response, the client delivery process
579 crashed while trying to generate an error log message.
58119. If a "warn" verb in an ACL added multiple headers to a message in a single
582 string, for example:
584 warn message = H1: something\nH2: something
586 the text was added as a single header line from Exim's point of view
587 though it ended up OK in the delivered message. However, searching for the
588 second and subsequent header lines using $h_h2: did not work. This has been
589 fixed. Similarly, if a system filter added multiple headers in this way,
590 the routers could not see them.
59220. Expanded the error message when iplsearch is called with an invalid key to
593 suggest using net-iplsearch in a host list.
59521. When running tests using -bh, any delays imposed by "delay" modifiers in
596 ACLs are no longer actually imposed (and a message to that effect is
597 output).
59922. If a "gecos" field in a passwd entry contained escaped characters, in
600 particular, if it contained a \" sequence, Exim got it wrong when building
601 a From: or a Sender: header from that name. A second bug also caused
602 incorrect handling when an unquoted " was present following a character
603 that needed quoting.
60523. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
606 was not being matched caselessly.
60824. Arranged for all hyphens in the exim.8 source to be escaped with
609 backslashes.
61125. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
612 properly. Recipient callout cache records were still being keyed to include
613 the sender, even when use_sender was set false. This led to far more
614 callouts that were necessary. The sender is no longer included in the key
615 when use_sender is false.
61726. Added "control = submission" modifier to ACLs.
61927. Added the ${base62d: operator to decode base 62 numbers.
62128. dnsdb lookups can now access SRV records.
62329. CONFIGURE_OWNER can be set at build time to define an alternative owner for
624 the configuration file.
62630. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
627 (-v) mode. This makes the output for a verbose queue run more intelligible.
62931. Added a use_postmaster feature to recipient callouts.
63132. Added the $body_zerocount variable, containing the number of binary zero
632 bytes in the message body.
63433. The time of last modification of the "new" subdirectory is now used as the
635 "mailbox time last read" when there is a quota error for a maildir
636 delivery.
63834. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
64035. Added +ignore_unknown as a special item in host lists.
64236. Code for decoding IPv6 addresses in host lists is now included, even if
643 IPv6 support is not being compiled. This fixes a bug in which an IPv6
644 address was recognized as an IP address, but was then not correctly decoded
645 into binary, causing unexpected and incorrect effects when compared with
646 another IP address.
649Exim version 4.34
652 1. Very minor rewording of debugging text in manualroute to say "list of
653 hosts" instead of "hostlist".
655 2. If verify=header_syntax was set, and a header line with an unqualified
656 address (no domain) and a large number of spaces between the end of the
657 name and the colon was received, the reception process suffered a buffer
658 overflow, and (when I tested it) crashed. This was caused by some obsolete
659 code that should have been removed. The fix is to remove it!
661 3. When running in the test harness, delay a bit after writing a bounce
662 message to get a bit more predictability in the log output.
664 4. Added a call to search_tidyup() just before forking a reception process. In
665 theory, someone could use a lookup in the expansion of smtp_accept_max_
666 per_host which, without the tidyup, could leave open a database connection.
668 5. Added the variables $recipient_data and $sender_data which get set from a
669 lookup success in an ACL "recipients" or "senders" condition, or a router
670 "senders" option, similar to $domain_data and $local_part_data.
672 6. Moved the writing of debug_print from before to after the "senders" test
673 for routers.
675 7. Change 4.31/66 (moving the time when the Received: is generated) caused
676 problems for message scanning, either using a data ACL, or using
677 local_scan() because the Received: header was not generated till after they
678 were called (in order to set the time as the time of reception completion).
679 I have revised the way this works. The header is now generated after the
680 body is received, but before the ACL or local_scan() are called. After they
681 are run, the timestamp in the header is updated.
684Exim version 4.33
687 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
688 before starting a queue runner without re-exec. This happened only when
689 deliver_drop_privilege was set or when the Exim user was set to root. The
690 effect of the bug was that timeouts during subsequent deliveries caused
691 crashes instead of being properly handled. The handler is now left at its
692 default (and expected) setting.
694 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
695 message, again when deliver_drop_privilege is set or Exim is run as root.
696 The bug described in (1) was not present in this case, but the tidying up
697 of the other signals was missing. I have made the two cases consistent.
699 3. The ignore_target_hosts setting on a manualroute router was being ignored
700 for hosts that were looked up using the /MX notation.
702 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
703 in domain lists.
705 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
706 operated on the sender address. After changing the $sender_address to <>
707 for the sender address verify, Exim was re-instated it as the original
708 (before rewriting) address, but remembering that it had rewritten it, so it
709 wasn't rewriting it again. This bug also had the effect of breaking the
710 sender address verification caching when the sender address was rewritten.
712 6. The ignore_target_hosts option was being ignored by the ipliteral router.
713 This has been changed so that if the ip literal address matches
714 ignore_target_hosts, the router declines.
716 7. Added expansion conditions match_domain, match_address, and match_local_
717 part (NOT match_host).
719 8. The placeholder for the Received: header didn't have a length field set.
721 9. Added code to Exim itself and to exim_lock to test for a specific race
722 condition that could lead to file corruption when using MBX delivery. The
723 issue is with the lockfile that is created in /tmp. If this file is removed
724 after a process has opened it but before that process has acquired a lock,
725 there is the potential for a second process to recreate the file and also
726 acquire a lock. This could lead to two Exim processes writing to the file
727 at the same time. The added code performs the same test as UW imapd; it
728 checks after acquiring the lock that its file descriptor still refers to
729 the same named file.
73110. The buffer for building added header lines was of fixed size, 8192 bytes.
732 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
733 when Exim is built.
73511. Added the smtp_active_hostname option. If used, this will typically be made
736 to depend on the incoming interface address. Because $interface_address is
737 not set up until the daemon has forked a reception process, error responses
738 that can happen earlier (such as "too many connections") no longer contain
739 a host name.
74112. If an expansion in a condition on a "warn" statement fails because a lookup
742 defers, the "warn" statement is abandoned, and the next ACL statement is
743 processed. Previously this caused the whole ACL to be aborted.
74513. Added the iplsearch lookup type.
74714. Added ident_timeout as a log selector.
74915. Added tls_certificate_verified as a log selector.
75116. Added a global option tls_require_ciphers (compare the smtp transport
752 option of the same name). This controls incoming TLS connections.
75417. I finally figured out how to make tls_require_ciphers do a similar thing
755 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
756 before starting the TLS session.
75818. Tabs are now shown as \t in -bP output.
76019. If the log selector return_path_on_delivery was set, Exim crashed when
761 bouncing a message because it had too many Received: header lines.
76320. If two routers both had headers_remove settings, and the first one included
764 a superfluous trailing colon, the final name in the first list and the
765 first name in the second list were incorrectly joined into one item (with a
766 colon in the middle).
769Exim version 4.32
772 1. Added -C and -D options to the exinext utility, mainly to make it easier
773 to include in the automated testing, but these could be helpful when
774 multiple configurations are in use.
776 2. The exinext utility was not formatting the output nicely when there was
777 an alternate port involved in the retry record key, nor when there was a
778 message id as well (for retries that were specific to a specific message
779 and a specific host). It was also confused by IPv6 addresses, because of
780 the additional colons they contain. I have fixed the IPv4 problem, and
781 patched it up to do a reasonable job for IPv6.
783 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
784 delivery, the log line now contains "pipelined" if PIPELINING was used.
786 4. An SMTP transport process used to panic and die if the bind() call to set
787 an explicit outgoing interface failed. This has been changed; it is now
788 treated in the same way as a connect() failure.
790 5. A reference to $sender_host_name in the part of a conditional expansion
791 that was being skipped was still causing a DNS lookup. This no longer
792 occurs.
794 6. The def: expansion condition was not recognizing references to header lines
795 that used bh_ and bheader_.
797 7. Added the _cache feature to named lists.
799 8. The code for checking quota_filecount in the appendfile transport was
800 allowing one more file than it should have been.
802 9. For compatibility with Sendmail, the command line option
804 -prval:sval
806 is equivalent to
808 -oMr rval -oMs sval
810 and sets the incoming protocol and host name (for trusted callers). The
811 host name and its colon can be omitted when only the protocol is to be set.
812 Note the Exim already has two private options, -pd and -ps, that refer to
813 embedded Perl. It is therefore impossible to set a protocol value of "d" or
814 "s", but I don't think that's a major issue.
81610. A number of refactoring changes to the code, none of which should affect
817 Exim's behaviour:
819 (a) The number of logging options was getting close to filling up the
820 32-bit word that was used as a bit map. I have split them into two classes:
821 those that are passed in the argument to log_write(), and those that are
822 only ever tested independently outside of that function. These are now in
823 separate 32-bit words, so there is plenty of room for expansion again.
824 There is no change in the user interface or the logging behaviour.
826 (b) When building, for example, log lines, the code previously used a
827 macro that called string_cat() twice, in order to add two strings. This is
828 not really sufficiently general. Furthermore, there was one instance where
829 it was actually wrong because one of the argument was used twice, and in
830 one call a function was used. (As it happened, calling the function twice
831 did not affect the overall behaviour.) The macro has been replaced by a
832 function that can join an arbitrary number of extra strings onto a growing
833 string.
835 (c) The code for expansion conditions now uses a table and a binary chop
836 instead of a serial search (which was left over from when there were very
837 few conditions). Also, it now recognizes conditions like "pam" even when
838 the relevant support is not compiled in: a suitably worded error message is
839 given if an attempt is made to use such a condition.
84111. Added ${time_interval:xxxxx}.
84312. A bug was causing one of the ddress fields not to be passed back correctly
844 from remote delivery subprocesses. The field in question was not being
845 subsequently used, so this caused to problems in practice.
84713. Added new log selectors queue_time and deliver_time.
84914. Might have fixed a bug in maildirsizefile handling that threw up
850 "unexpected character" debug warnings, and recalculated the data
851 unnecessarily. In any case, I expanded the warning message to give more
852 information.
85415. Added the message "Restricted characters in address" to the statements in
855 the default ACL that block characters like @ and % in local parts.
85716. Change 71 for release 4.31 proved to be much less benign that I imagined.
858 Three changes have been made:
860 (a) There was a serious bug; a negative response to MAIL caused the whole
861 recipient domain to be cached as invalid, thereby blocking all messages
862 to all local parts at the same domain, from all senders. This bug has
863 been fixed. The domain is no longer cached after a negative response to
864 MAIL if the sender used is not empty.
866 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
867 been restored.
869 (c) A new callout option, "use_sender" has been added for people who want
870 the modified behaviour.
873Exim version 4.31
876 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
877 Larry Rosenman.
879 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
880 indeed breaks things for older releases.
882 3. Added additional logging to the case where there is a problem reading data
883 from a filter that is running in a subprocess using a pipe, in order to
884 try to track down a specific problem.
886 4. Testing facility fudge: when running in the test harness and attempting
887 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
888 getting "No route to host". Convert this to a timeout.
890 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
891 warning.
893 6. Some OS don't have socklen_t but use size_t instead. This affects the
894 fifth argument of getsockopt() amongst other things. This is now
895 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
896 can be set for individual OS. I have set it for SunOS5, OSF1, and
897 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
898 some earlier ones do not.
900 7. Change 4.30/15 was not doing the test caselessly.
902 8. The standard form for an IPv6 address literal was being rejected by address
903 parsing in, for example, MAIL and RCPT commands. An example of this kind of
904 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
905 this, as well as the form without the "IPv6" on the front (but only when
906 address literals are enabled, of course).
908 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
91010. Exim crashed if a message with an empty sender address specified by -f
911 encountered a router with an errors_to setting. This could be provoked only
912 by a command such as
914 exim -f "" ...
916 where an empty string was supplied; "<>" did not hit this bug.
91811. Installed PCRE release 4.5.
92012. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
921 remained set. It is now erased.
92313. exiqgrep wasn't working on MacOS X because it didn't correctly compute
924 times from message ids (which are base 36 rather than the normal 62).
92614. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
927 were being counted as actual protocol errors, and logged if the log
928 selector +smtp_protocol_error was set. One cannot be perfect in this test,
929 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
930 and DATA following a set of rejected RCPTs do not count as protocol errors.
931 In other words, Exim assumes they were pipelined, though this may not
932 actually be the case. Of course, in all cases the client gets an
933 appropriate error code.
93515. If a lookup fails in an ACL condition, a message about the failure may
936 be available; it is used if testing the ACL cannot continue, because most
937 such messages specify what the cause of the deferral is. However, some
938 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
939 that caused an old message to be retained and used if a later statement
940 caused a defer, replacing the real cause of the deferral.
94216. If an IP address had so many PTR records that the DNS lookup buffer
943 was not large enough to hold them, Exim could crash while trying to process
944 the truncated data. It now detects and logs this case.
94617. Further to 4.21/58, another change has been made: if (and only if) the
947 first line of a message (the first header line) ends with CRLF, a bare LF
948 in a subsequent header line has a space inserted after it, so as not to
949 terminate the header.
95118. Refactoring: tidied an ugly bit of code in appendfile that copied data
952 unnecessarily, used atoi() instead of strtol(), and didn't check the
953 termination when getting file sizes from file names by regex.
95519. Completely re-implemented the support for maildirsize files, in the light
956 of a number of problems with the previous contributed implementation
957 (4.30/29). In particular:
959 . If the quota is zero, the maildirsize file is maintained, but no quota is
960 imposed.
962 . If the maildir directory does not exist, it is created before any attempt
963 to write a maildirsize file.
965 . The quota value in the file is just a cache; if the quota is changed in
966 the transport, the new value overrides.
968 . A regular expression is available for excluding directories from the
969 count.
97120. The autoreply transport checks the characters in options that define the
972 message's headers; it allows continued headers, but it was checking with
973 isspace() after an embedded newline instead of explicitly looking for a
974 space or a tab.
97621. If all the "regular" hosts to which an address was routed had passed their
977 expiry times, and had not reached their retry times, the address was
978 bounced, even if fallback hosts were defined. Now Exim should go on to try
979 the fallback hosts.
98122. Increased buffer sizes in the callout code from 1024 to 4096 to match the
982 equivalent code in the SMTP transport. Some hosts send humungous responses
983 to HELO/EHLO, more than 1024 it seems.
98523. Refactoring: code in filter.c used (void *) for "any old type" but this
986 gives compiler warnings in some environments. I've now done it "properly",
987 using a union.
98924. The replacement for inet_ntoa() that is used with gcc on IRIX systems
990 (because of problems with the built-in one) was declared to return uschar *
991 instead of char *, causing compiler failure.
99325. Fixed a file descriptor leak when processing alias/forward files.
99526. Fixed a minor format string issue in dbfn.c.
99727. Typo in exim.c: ("dmbnz" for "dbmnz").
99928. If a filter file refered to $h_xxx or $message_headers, and the headers
1000 contained RFC 2047 "words", Exim's memory could, under certain conditions,
1001 become corrupted.
100329. When a sender address is verified, it is cached, to save repeating the test
1004 when there is more than one recipient in a message. However, when the
1005 verification involves a callout, it is possible for different callout
1006 options to be set for different recipients. It is too complicated to keep
1007 track of this in the cache, so now Exim always runs a verification when a
1008 callout is required, relying on the callout cache for the optimization.
1009 The overhead is duplication of the address routing, but this should not be
1010 too great.
101230. Fixed a bug in callout caching. If a RCPT command caused the sender address
1013 to be verified with callout=postmaster, and the main callout worked but the
1014 postmaster check failed, the verification correctly failed. However, if a
1015 subsequent RCPT command asked for sender verification *without* the
1016 postmaster check, incorrect caching caused this verification also to fail,
1017 incorrectly.
101931. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
1020 it was not caching the DNS options (qualify_single, search_parents) that
1021 were used when the lookup failed. A subsequent lookup with different
1022 options therefore always gave the same answer, though there were cases
1023 where it should not have. (Example: a "domains = !$mx_any" option on a
1024 dnslookup router: the "domains" option is always processed without any
1025 widening, but the router might have qualify_single set.) Now Exim uses the
1026 cached value only when the same options are set.
102832. Added John Jetmore's "exipick" utility to the distribution.
103033. GnuTLS: When an attempt to start a TLS session fails for any reason other
1031 than a timeout (e.g. a certificate is required, and is not provided), an
1032 Exim server now closes the connection immediately. Previously it waited for
1033 the client to close - but if the client is SSL, it seems that they each
1034 wait for each other, leading to a delay before one of them times out.
103634: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
1037 maintained 0.8.x compatibility because I don't think many are using it, and
1038 it is clearly obsolete.
104035. Added TLS support for CRLs: a tls_crl global option and one for the smtp
1041 transport.
104336. OpenSSL: $tls_certificate_verified was being set to 1 even if the
1044 client certificate was expired. A simple patch fixes this, though I don't
1045 understand the full logic of why the verify callback is called multiple
1046 times.
104837. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
1049 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
1050 which causes problems with some clients (such as the Certicom SSL Plus
1051 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
1052 disables the coutermeasure allowing Eudora to connect."
105438. Exim was not checking that a write() to a log file succeeded. This could
1055 lead to Bad Things if a log got too big, in particular if it hit a file
1056 size limit. Exim now panics and dies if it cannot write to a log file, just
1057 as it does if it cannot open a log file.
105939. Modified OS/Makefile-Linux so that it now contains
1063 The two -D definitions ensure that Exim is compiled with large file
1064 support, which makes it possible to handle log files that are bigger than
1065 2^31.
106740. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
1068 instance) a domain was checked against a named list that involved a lookup,
1069 causing $domain_data to be set, then another domain was checked against the
1070 same list, then the first domain was re-checked, the value of $domain_data
1071 after the final check could be wrong. In particular, if the second check
1072 failed, it could be set empty. This bug probably also applied to
1073 $localpart_data.
107541. The strip_trailing_dot option was not being applied to the address given
1076 with the -f command-line option.
107842. The code for reading a message's header from the spool was incrementing
1079 $received_count, but never initializing it. This meant that the value was
1080 incorrect (doubled) while delivering a message in the same process in which
1081 it was received. In the most common configuration of Exim, this never
1082 happens - a fresh exec is done - but it can happen when
1083 deliver_drop_privilege is set.
108543. When Exim logs an SMTP synchronization error - client data sent too soon -
1086 it now includes up to 150 characters of the unexpected data in the log
1087 line.
108944. The exim_dbmbuild utility uses fixed size buffers for reading input lines
1090 and building data strings. The size of both of these buffers was 10 000
1091 bytes - far larger than anybody would *ever* want, thought I. Needless to
1092 say, somebody hit the limit. I have increased the maximum line length to
1093 20 000 and the maximum data length of concatenated lines to 100 000. I have
1094 also fixed two bugs, because there was no checking on these buffers. Tsk,
1095 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1096 buffer is too small.
109845. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1099 lsearch lookups. Now it does.
110146. When parsing a route_list item in a manualroute router, a fixed-length
1102 buffer was used for the list of hosts. I made this 1024 bytes long,
1103 thinking that nobody would ever have a list of hosts that long. Wrong.
1104 Somebody had a whole pile of complicated expansion conditions, and the
1105 string was silently truncated, leading to an expansion error. It turns out
1106 that it is easier to change to an unlimited length (owing to other changes
1107 that have happened since this code was originally written) than to build
1108 structure for giving a limitation error. The length of the item that
1109 expands into the list of hosts is now unlimited.
111147. The lsearch lookup could not handle data where the length of text line was
1112 more than 4095 characters. Such lines were truncated, leading to shortened
1113 data being returned. It should now handle lines of any length.
111548. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
1116 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
1117 ACL").
111949. Cosmetic tidy to scripts like exicyclog that are generated by globally
1120 replacing strings such as BIN_DIRECTORY in a source file: the replacement
1121 no longer happens in comment lines. A list of replacements is now placed
1122 at the head of all of the source files, except those whose only change is
1123 to replace PERL_COMMAND in the very first #! line.
112550. Replaced the slow insertion sort in queue.c, for sorting the list of
1126 messages on the queue, with a bottom-up merge sort, using code contributed
1127 by Michael Haardt. This should make operations like -bp somewhat faster on
1128 large queues. It won't affect queue runners, except when queue_run_in_order
1129 is set.
113151. Installed eximstats 1.31 in the distribution.
113352. Added support for SRV lookups to the dnslookup router.
113553. If an ACL referred to $message_body or $message_body_end, the value was not
1136 reset for any messages that followed in the same SMTP session.
113854. The store-handling optimization for building very long strings was not
1139 differentiating between the different store pools. I don't think this
1140 actually made any difference in practice, but I've tidied it.
114255. While running the routers to verify a sender address, $sender_address
1143 was still set to the sender address. This is wrong, because when routing to
1144 send a bounce to the sender, it would be empty. Therefore, I have changed
1145 it so that, while verifying a sender address, $sender_address is set to <>.
1146 (There is no change to what happens when verifying a recipient address.)
114856. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
1149 target A or AAAA records (if not already returned) without resetting the
1150 qualify_single or search_parents options of the DNS resolver. These are
1151 inappropriate in this case because the targets of MX and SRV records must
1152 be FQDNs. A broken DNS record could cause trouble if it happened to have a
1153 target that, when qualified, matched something in the local domain. These
1154 two options are now turned off when doing these lookups.
115657. It seems that at least some releases of Reiserfs (which does not have the
1157 concept of a fixed number of inodes) returns zero and not -1 for the
1158 number of available inodes. This interacted badly with check_spool_inodes,
1159 which assumed that -1 was the "no such thing" setting. What I have done is
1160 to check that the total number of inodes is greater than zero before doing
1161 the test of how many are available.
116358. When a "warn" ACL statement has a log_message modifier, the message is
1164 remembered, and not repeated. This is to avoid a lot of repetition when a
1165 message has many recipients that cause the same warning to be written.
1166 Howewer, Exim was preserving the list of already written lines for an
1167 entire SMTP session, which doesn't seem right. The memory is now reset if a
1168 new message is started.
117059. The "rewrite" debugging flag was not showing the result of rewriting in the
1171 debugging output unless log_rewrite was also set.
117360. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
1174 of (int)(handle) when we know that handle contains (void *)(-1).
117661. The Exim daemon panic-logs an error return when it closes the incoming
1177 connection. However "connection reset by peer" seems to be common, and
1178 isn't really an error worthy of noting specially, so that particular error
1179 is no long logged.
118162. When Exim is trying to find all the local interfaces, it used to panic and
1182 die if the ioctl to get the interface flags failed. However, it seems that
1183 on at least one OS (Solaris 9) it is possible to have an interface that is
1184 included in the list of interfaces, but for which you get a failure error
1185 for this call. This happens when the interface is not "plumbed" into a
1186 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
1187 failure of the "get flags" call assumes that the interface is down.
118963. Added a ${eval10: operator, which assumes all numbers are decimal. This
1190 makes life easier for people who are doing arithmetic on fields extracted
1191 from dates, where you often get leading zeros that should not be
1192 interpreted as octal.
119464. Added qualify_domain to the redirect router, to override the global
1195 setting.
119765. If a pathologically long header line contained very many addresses (the
1198 report of this problem mentioned 10 000) and each of them was rewritten,
1199 Exim could use up a very large amount of memory. (It kept on making new
1200 copies of the header line as it rewrote, and never released the old ones.)
1201 At the expense of a bit more processing, the header rewriting function has
1202 been changed so that it no longer eats memory in this way.
120466. The generation of the Received: header has been moved from the time that a
1205 message starts to be received, to the time that it finishes. The timestamp
1206 in the Received: header should now be very close to that of the <= log
1207 line. There are two side-effects of this change:
1209 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
1210 logged header lines no longer include the local Received: line, because
1211 it has not yet been created. The same applies to a copy of the message
1212 that is returned to a non-SMTP sender when a message is rejected.
1214 (b) When a filter file is tested using -bf, no additional Received: header
1215 is added to the test message. After some thought, I decided that this
1216 is a bug fix.
1218 This change does not affect the value of $received_for. It is still set
1219 after address rewriting, but before local_scan() is called.
122167. Installed the latest Cygwin-specific files from the Cygwin maintainer.
122368. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
1224 gave an unhelpful panic error message, and a defer error. I have managed to
1225 change this behaviour so that it now rejects any supplied certificate,
1226 which seems right, as the list of acceptable certificates is empty.
122869. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
1229 gave an unhelpful defer error. I have not managed to make this reject any
1230 supplied certificates, but the error message it gives is "no certificate
1231 supplied", which is not helpful.
123370. exigrep's output now also includes lines that are not associated with any
1234 message, but which match the given pattern. Implemented by a patch from
1235 Martin Sluka, which also tidied up the Perl a bit.
123771. Recipient callout verification, like sender verification, was using <> in
1238 the MAIL FROM command. This isn't really the right thing, since the actual
1239 sender may affect whether the remote host accepts the recipient or not. I
1240 have changed it to use the actual sender in the callout; this means that
1241 the cache record is now keyed on a recipient/sender pair, not just the
1242 recipient address. There doesn't seem to be a real danger of callout loops,
1243 since a callout by the remote host to check the sender would use <>.
1244 [SEE ABOVE: changed after hitting problems.]
124672. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
1247 temporary errors. However, in the case of such a code being given after
1248 the end of a data transmission (i.e. after ".") Exim was failing to write
1249 a retry record for the message. (Yes, there was some broken host that was
1250 actually sending 8xx at this point.)
125273. An unknown lookup type in a host list could cause Exim to panic-die when
1253 the list was checked. (An example that provoked this was putting <; in the
1254 middle of a list instead of at the start.) If this happened during a DATA
1255 ACL check, a -D file could be left lying around. This kind of configuration
1256 error no longer causes Exim to die; instead it causes a defer errror. The
1257 incident is still logged to the main and panic logs.
125974. Buglet left over from Exim 3 conversion. The message "too many messages
1260 in one connection" was written to the rejectlog but not the mainlog, except
1261 when address rewriting (yes!) was being logged.
126375. Added write_rejectlog option.
126576. When a system filter was run not as root (that is, when system_filter_user
1266 was set), the values of the $n variables were not being returned to the
1267 main process; thus, they were not subsequently available in the $sn
1268 variables.
127077. Added +return_path_on_delivery log selector.
127278. A connection timeout was being treated differently from recipients deferred
1273 when testing hosts_max_try with a message that was older than the host's
1274 retry timeout. (The host should not be counted, thus allowing all hosts to
1275 be tried at least once before bouncing.) This may have been the cause of an
1276 occasionally reported bug whereby a message would remain on the queue
1277 longer than the retry timeout, but would be bounced if a delivery was
1278 forced. I say "may" because I never totally pinned down the problem;
1279 setting up timeout/retry tests is difficult. See also the next item.
128179. The ultimate address timeout was not being applied to errors that involved
1282 a combination of host plus message (for example, a timeout on a MAIL
1283 command). When an address resolved to a number of possible hosts, and they
1284 were not all tried for each delivery (e.g. because of hosts_max_try), a
1285 message could remain on the queue longer than the retry timeout.
128780. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
1288 Haardt.
129081. Fixed an obscure SMTP outgoing bug which required at least the following
1291 conditions: (a) there was another message waiting for the same server;
1292 (b) the server returned 5xx to all RCPT commands in the first message so
1293 that the message was not completed; (c) the server dropped the connection
1294 or gave a negative response to the RSET that Exim sends to abort the
1295 transaction. The observed case was a dropped connection after DATA that had
1296 been sent in pipelining mode. That is, the server had advertised PIPELINING
1297 but was not implementing it correctly. The effect of the bug was incorrect
1298 behaviour, such as trying another host, and this could lead to a crash.
1301Exim version 4.30
1304 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
1305 and daemon.c were passed as pointers to ints; they should have been
1306 pointers to socklen_t variables (which are typically unsigned ints).
1308 2. Some signed/unsigned type warnings in the os.c file for Linux have been
1309 fixed.
1311 3. Fixed a really odd bug that affected only the testing scheme; patching a
1312 certain fixed string in the binary changed the value of another string that
1313 happened to be identical to the end of the original first string.
1315 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
1316 name", it returns that address as the IP address. On some operating
1317 systems (e.g. Solaris), it also passes back the IP address string as the
1318 "host name". However, on others (e.g. Linux), it passes back an empty
1319 string. Exim wasn't checking for this, and was changing the host name to an
1320 empty string, assuming it had been canonicized.
1322 5. Although rare, it is permitted to have more than one PTR record for a given
1323 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
1324 all the names associated with an address, because they do in Solaris.
1325 However, it seems that they do not in Linux for data that comes from the
1326 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
1327 I found this out when I moved to a new Linux workstation and tried to run
1328 the Exim test suite.
1330 To get round this problem I have changed the code so that it now does its
1331 own call to the DNS to look up PTR records when searching for a host name.
1332 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
1333 addresses that are only in /etc/hosts are still found.
1335 This behaviour is, however, controlled by an option called host_lookup_
1336 order, which defaults to "bydns:byaddr". If people want to use the other
1337 order, or indeed, just use one or the other means of lookup, they can
1338 specify it in this variable.
1340 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
1341 some operating systems, this comes back from gethostbyaddr() as an empty
1342 string, and this is what Exim used to test for. However, it seems that in
1343 other systems, "." is yielded. Exim now tests for this case too.
1345 7. The values of check_spool_space and check_log_space are now held internally
1346 as a number of kilobytes instead of an absolute number of bytes. If a
1347 numbers is specified without 'K' or 'M', it is rounded up to the nearest
1348 kilobyte. This means that much larger values can be stored.
1350 8. Exim monitor: an attempt to get the action menu when not actually pointing
1351 at a message produces an empty menu entitled "No message selected". This
1352 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
1353 no entries in it ("Shell widget menu has zero width and/or height"). So I
1354 have added a single, blank menu entry in this case.
1356 9. Added ${quote_local_part.
135810. MIME decoding is now applied to the contents of Subject: header lines when
1359 they are logged.
136111. Now that a reference to $sender_host_address automatically causes a reverse
1362 lookup to occur if necessary (4.13/18), there is no need to arrange for a
1363 host lookup before query-style lookups in lists that might use this
1364 variable. This has therefore been abolished, and the "net-" prefix is no
1365 longer necessary for query-style lookups.
136712. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
1368 have been a typo for LFLAGS, so it has been changed.
137013. The install script calls Exim with "-C /dev/null" in order to find the
1371 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
1372 to be output. Howeve, since Exim outputs its version number before the
1373 error, it didn't break the script. It just looked ugly. I fixed this by
1374 always allowing "-C /dev/null" if the caller is root.
137614. Ignore overlarge ACL variable number when reading spool file - insurance
1377 against a later release with more variables having written the file.
137915. The standard form for an IPv6 address literal was being rejected by EHLO.
1380 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1381 this, as well as the form without the "IPv6" on the front.
138316. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
1384 OS/Makefile-Darwin file.
138617. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
1387 only to LDAP libraries that do not have LDAP_OPT_DEREF.
138918. After change 4.21/52, "%ld" was used to format the contents of the $inode
1390 variable. However, some OS use ints for inodes. I've added cast to long int
1391 to get rid of the compiler warning.
139319. I had forgotten to lock out "/../" in configuration file names when
1394 ALT_CONFIG_PREFIX was set.
139620. Routers used for verification do not need to specify transports. However,
1397 if such a router generated a host list, and callout was configured, Exim
1398 crashed, because it could not find a port number from the (non-existent)
1399 transport. It now assumes port 25 in this circumstance.
140121. Added the -t option to exigrep.
140322. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
1404 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
1405 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
1406 them is set, LOOKUP_LSEARCH is forced.
140823. "exim -bV" now outputs a list of lookups that are included in the binary.
141024. Added sender and host information to the "rejected by local_scan()" log
1411 line; previously there was no indication of these.
141325. Added .include_if_exists.
141526. Change 3.952/11 added an explicit directory sync on top of a file sync for
1416 Linux. It turns out that not all file systems support this. Apparently some
1417 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
1418 do it.) To cope with this, the error EINVAL, which means that sync-ing is
1419 not supported on the file descriptor, is now ignored when Exim is trying to
1420 sync a directory. This applies only to Linux.
142227. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
142428. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
1425 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
1426 this.
142829. Added support for maildirsize files from supplied patch (modified a bit).
143030. The use of :fail: followed by an empty string could lead Exim to respond to
1431 sender verification failures with (e.g.):
1433 550 Verification failed for <xxx>
1434 550 Sender verify failed
1436 where the first response line was missing the '-' that indicates it is not
1437 the final line of the response.
143931. The loop for finding the name of the user that called Exim had a hardwired
1440 limit of 10; it now uses the value of finduser_retries, which is used for
1441 all other user lookups.
144332. Added $received_count variable, available in data and not_smtp ACLs, and at
1444 delivery time.
144633. Exim was neglecting to zero errno before one call of strtol() when
1447 expanding a string and expecting an integer value. On some systems this
1448 resulted in spurious "integer overflow" errors. Also, it was casting the
1449 result into an int without checking.
145134. Testing for a connection timeout using "timeout_connect" in the retry rules
1452 did not work. The code looks as if it has *never* worked, though it appears
1453 to have been documented since at least releast 1.62. I have made it work.
145535. The "timeout_DNS" error in retry rules, also documented since at least
1456 1.62, also never worked. As it isn't clear exactly what this means, and
1457 clearly it isn't a major issue, I have abolished the feature by treating it
1458 as "timeout", and writing a warning to the main and panic logs.
146036. The display of retry rules for -brt wasn't always showing the error code
1461 correctly.
146337. Added new error conditions to retry rules: timeout_A, timeout_MX,
1464 timeout_connect_A, timeout_connect_MX.
146638. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
1467 to the empty sender.
146939. The daemon was not analysing the content of -oX till after it had closed
1470 stderr and disconnected from the controlling terminal. This meant that any
1471 syntax errors were only noted on the panic log, and the return code from
1472 the command was 0. By re-arranging the code a little, I've made the
1473 decoding happen first, so such errors now appear on stderr, and the return
1474 code is 1. However, the actual setting up of the sockets still happens in
1475 the disconnected process, so errors there are still only recorded on the
1476 panic log.
147840. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
1479 connections (as happens on some IP stacks) was logged at start up time as
1480 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
1481 it from "IPv6 and IPv4", which means that two separate sockets are being
1482 used.
148441. The debug output for gethostbyname2() or getipnodebyname() failures now
1485 says whether AF_INET or AF_INET6 was passed as an argument.
148742. Exiwhat output was messed up when time zones were included in log
1488 timestamps.
149043. Exiwhat now gives more information about the daemon's listening ports,
1491 and whether -tls-on-connect was used.
149344. The "port" option of the smtp transport is now expanded.
149545. A "message" modifier in a "warn" statement in a non-message ACL was being
1496 silently ignored. Now an error message is written to the main and panic
1497 logs.
149946. There's a new ACL modifier called "logwrite" which writes to a log file
1500 as soon as it is encountered.
150247. Added $local_user_uid and $local_user_gid at routing time.
150448. Exim crashed when trying to verify a sender address that was being
1505 rewritten to "<>".
150749. Exim was recognizing only a space character after ".include". It now also
1508 recognizes a tab character.
151050. Fixed several bugs in the Perl script that creates the exim.8 man page by
1511 extracting the relevant information from the specification. The man page no
1512 longer contains scrambled data for the -d option, and I've added a section
1513 at the front about calling Exim under different names.
151551. Added "extra_headers" argument to the "mail" command in filter files.
151752. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
1518 crash.
152053. Installed eximstats 1.29.
152254. Added transport_filter_timeout as a generic transport option.
152455. Exim no longer adds an empty Bcc: header to messages that have no To: or
1525 Cc: header lines. This was required by RFC 822, but it not required by RFC
1526 2822.
152856. Exim used to add From:, Date:, and Message-Id: header lines to any
1529 incoming messages that did not have them. Now it does so only if the
1530 message originates locally, that is, if there is no associated remote host
1531 address. When Resent- header lines are present, this applies to the Resent-
1532 lines rather than the non-Resent- lines.
153457. Drop incoming SMTP connection after too many syntax or protocol errors. The
1535 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
153758. Messages for configuration errors now include the name of the main
1538 configuration file - useful now that there may be more than one file in a
1539 list (.included file names were always shown).
154159. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
1542 for those rare installations that do not start the daemon as root or run it
1543 setuid root. I've cut out the call to initgroups() if the daemon is not
1544 root at that time.
154660. The Exim user and group can now be bound into the binary as text strings
1547 that are looked up at the start of Exim's processing.
154961. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
155162. Added $mailstore_basename variable.
155363. Installed patch to sieve.c from Michael Haardt.
155564. When Exim failed to open the panic log after failing to open the main log,
1556 the original message it was trying to log was written to stderr and debug
1557 output, but if they were not available (the usual case in production), it
1558 was lost. Now it is written to syslog before the two lines that record the
1559 failures to open the logs.
156165. Users' Exim filters run in subprocesses under the user's uid. It is
1562 possible for a "deliver" command or an alias in a "personal" command to
1563 provoke an address rewrite. If logging of address rewriting is configured,
1564 this fails because the process is not running as root or exim. There may be
1565 a better way of dealing with this, but for the moment (because 4.30 needs
1566 to be released), I have disabled address rewrite logging when running a
1567 filter in a non-root, non-exim process.
1570Exim version 4.24
1573 1. The buildconfig auxiliary program wasn't quoting the value set for
1574 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
1575 not defined. This bug was masked in 4.22 by the effect that was fixed in
1576 change 4.23/1.
1578 2. Some messages that were rejected after a message id was allocated were
1579 shown as "incomplete" by exigrep. It no longer does this for messages that
1580 are rejected by local_scan() or the DATA or non-SMTP ACLs.
1582 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
1583 have allow_domain_literals set, the ID did not get logged in the <= line.
1584 Domain literals are now always recognized in Message-ID: header lines.
1586 4. The first argument for a ${extract expansion item is the key name or field
1587 number. Leading and trailing spaces in this item were not being ignored,
1588 causing some misleading effects.
1590 5. When deliver_drop_privilege was set, single queue runner processes started
1591 manually (i.e. by the command "exim -q") or by the daemon (which uses the
1592 same command in the process it spins off) were not dropping privilege.
1594 6. When the daemon running as "exim" started a queue runner, it always
1595 re-executed Exim in the spun-off process. This is a waste of effort when
1596 deliver_drop_privilege is set. The new process now just calls the
1597 queue-runner function directly.
1600Exim version 4.23
1603 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
1606 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
1607 ignored. Though the use of -oP was forcing the writing of a pid file, it
1608 was always written to the default place.
1610 3. If the message "no IP address found for host xxxx" is generated during
1611 incoming verification, it is now followed by identification of the incoming
1612 connection (so you can more easily find what provoked it).
1614 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
1616 5. Added some features to "harden" Exim a bit more against certain attacks:
1618 (a) There is now a build-time option called FIXED_NEVER_USERS that can
1619 be put in Local/Makefile. This is like the never_users runtime option,
1620 but it cannot be overridden. The default setting is "root".
1622 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
1623 prefix string with which any file named in a -C command line option
1624 must start.
1626 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
1627 is retained for -C and -D only if the caller of Exim is root. Without
1628 it, the exim user may also use -C and -D and retain privilege.
1630 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
1631 command line option is disabled.
1633 6. Macro names set by the -D option must start with an upper case letter, just
1634 like macro names defined in the configuration file.
1636 7. Added "dereference=" facility to LDAP.
1638 8. Two instances of the typo "uknown" in the source files are fixed.
1640 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
1641 the Configure-Makefile script screwed up while processing it.
164310. Incorporated PCRE 4.4.
164511. The SMTP synchronization check was not operating right at the start of an
1646 SMTP session. For example, it could not catch a HELO sent before the client
1647 waited for the greeting. There is now a check for outstanding input at the
1648 point when the greeting is written. Because of the duplex, asynchronous
1649 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
1650 way, but not yet received, when the check is performed.
165212. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
1653 on TCP/IP sockets, because this apparently causes some broken clients to
1654 timeout.
165613. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
1657 unchanged) from the Cygwin maintainer.
165914. The code for -bV that shows what is in the binary showed "mbx" when maildir
1660 was supported instead of testing for mbx. Effectively a typo.
166215. The spa authenticator server code was not checking that the input it
1663 received was valid base64.
166516. The debug output line for the "set" modifier in ACLs was not showing the
1666 name of the variable that was being set.
166817. Code tidy: the variable type "vtype_string" was never used. Removed it.
167018. Previously, a reference to $sender_host_name did not cause a DNS reverse
1671 lookup on its own. Something else was needed to trigger the lookup. For
1672 example, a match in host_lookup or the need for a host name in a host list.
1673 Now, if $sender_host_name is referenced and the host name has not yet been
1674 looked up, a lookup is performed. If the lookup fails, the variable remains
1675 empty, and $host_lookup_failed is set to "1".
167719. Added "eqi" as a case-independent comparison operator.
167920. The saslauthd authentication condition could segfault if neither service
1680 nor realm was specified.
168221. If an overflowing value such as "2048M" was set for message_size_limit, the
1683 error message that was logged was misleading, and incoming SMTP
1684 connections were dropped. The message is now more accurate, and temporary
1685 errors are given to SMTP connections.
168722. In some error situations (such as 21 above) Exim rejects all SMTP commands
1688 (except RSET) with a 421 error, until QUIT is received. However, it was
1689 failing to send a response to QUIT.
169123. The HELO ACL was being run before the code for helo_try_verify_hosts,
1692 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
1693 ACL is now run after the helo_try_verify_hosts code.
169524. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
1696 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
1697 case-independent, so other case variants are also recognized). Apparently
1698 some systems use these upper case variants.
170025. If more than two messages were waiting for the same host, and a transport
1701 filter was specified for the transport, Exim sent two messages over the
1702 same TCP/IP connection, and then failed with "socket operation on non-
1703 socket" when it tried to send the third.
170526. Added Exim::debug_write and Exim::log_write for embedded Perl use.
170727. The extern definition of crypt16() in expand.c was not being excluded when
1708 the OS had its own crypt16() function.
171028. Added bounce_return_body as a new option, and bounce_return_size_limit
1711 as a preferred synonym for return_size_limit, both as an option and as an
1712 expansion variable.
171429. Added LIBS=-liconv to OS/Makefile-OSF1.
171630. Changed the default configuration ACL to relax the local part checking rule
1717 for addresses that are not in any local domains. For these addresses,
1718 slashes and pipe symbols are allowed within local parts, but the sequence
1719 /../ is explicitly forbidden.
172131. SPA server authentication was not clearing the challenge buffer before
1722 using it.
172432. log_message in a "warn" ACL statement was writing to the reject log as
1725 well as to the main log, which contradicts the documentation and doesn't
1726 seem right (because no rejection is happening). So I have stopped it.
172833. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
1729 However, I am unable to do any testing of this.
173134. Fixed an infelicity in the appendfile transport. When checking directories
1732 for a mailbox, to see if any needed to be created, it was accidentally
1733 using path names with one or more superfluous leading slashes; tracing
1734 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
173635. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
1737 contents are added to the log line that is written for every discarded
1738 recipient. (Previously a log_message setting was ignored.)
174036. The ${quote: operator now quotes the string if it is empty.
174237. The install script runs exim in order to find its version number. If for
1743 some reason other than non-existence or emptiness, which it checks, it
1744 could not run './exim', it was installing it with an empty version number,
1745 i.e. as "exim-". This error state is now caught, and the installation is
1746 aborted.
174838. An argument was missing from the function that creates an error message
1749 when Exim fails to connect to the socket for saslauthd authentication.
1750 This could cause Exim to crash, or give a corrupted message.
175239. Added isip, isip4, and isip6 to ${if conditions.
175440. The ACL variables $acl_xx are now saved with the message, and can be
1755 accessed later in routers, transports, and filters.
175741. The new lookup type nwildlsearch is like wildlsearch, except that the key
1758 strings in the file are not string-expanded.
176042. If a MAIL command specified a SIZE value that was too large to fit into an
1761 int variable, the check against message_size_limit failed. Such values are
1762 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
1763 day this will have to be increased, but I don't think I want to be around
1764 when emails are that large.
1768Exim version 4.22
1771 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
1772 iconv() is not standard in FreeBSD.
1774 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
1775 IPv6 enabled. The observed symptom was a segmentation fault on return from
1776 the function os_common_find_running_interfaces() in src/os.c.
1778 3. In the check_special_case() function in daemon.c I had used "errno" as an
1779 argument name, which causes warnings on some systems. This was basically a
1780 typo, since it was named "eno" in the comments!
1782 4. The code that waits for the clock to tick (at a resolution of some fraction
1783 of a second) so as to ensure message-id uniqueness was always waiting for
1784 at least one whole tick, when it could have waited for less. [This is
1785 almost certainly not relevant at current processor speeds, where it is
1786 unlikely to ever wait at all. But we try to future-proof.]
1788 5. The function that sleeps for a time interval that includes fractions of a
1789 second contained a race. It did not block SIGALRM between setting the
1790 timer, and suspending (a couple of lines later). If the interval was short
1791 and the sigsuspend() was delayed until after it had expired, the suspension
1792 never ended. On busy systems this could lead to processes getting stuck for
1793 ever.
1795 6. Some uncommon configurations may cause a lookup to happen in a queue runner
1796 process, before it forks any delivery processes. The open lookup caching
1797 mechanism meant that the open file or database connection was passed into
1798 the delivery process. The problem was that delivery processes always tidy
1799 up cached lookup data. This could cause a problem for the next delivery
1800 process started by the queue runner, because the external queue runner
1801 process does not know about the closure. So the next delivery process
1802 still has data in the lookup cache. In the case of a file lookup, there was
1803 no problem because closing a file descriptor in a subprocess doesn't affect
1804 the parent. However, if the lookup was caching a connection to a database,
1805 the connection was closed, and the second delivery process was likely to
1806 see errors such as "PGSQL: query failed: server closed the connection
1807 unexpectedly". The problem has been fixed by closing all cached lookups
1808 in a queue runner before running a delivery process.
1810 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
1811 seem to have the "const" qualifier which it has on other OS. I've
1812 parameterised it.
1814 8. Change 4.21/2 was too strict. It is only if there are two authenticators
1815 *of the same type* (client or server) with the same public name that an
1816 error should be diagnosed.
1818 9. When Exim looked up a host name for an IP address, but failed to find the
1819 original IP address when looking up the host name (a safety check), it
1820 output the message "<ip address> does not match any IP for NULL", which was
1821 confusing, to say the least. The bug was that the host name should have
1822 appeared instead of "NULL".
182410. Since release 3.03, if Exim is called by a uid other than root or the Exim
1825 user that is built into the binary, and the -C or -D options is used, root
1826 privilege is dropped before the configuration file is read. In addition,
1827 logging is switched to stderr instead of the normal log files. If the
1828 configuration then re-defines the Exim user, the unprivileged environment
1829 is probably not what is expected, so Exim logs a panic warning message (but
1830 proceeds).
1832 However, if deliver_drop_privilege is set, the unprivileged state may well
1833 be exactly what is intended, so the warning has been cut out in that case,
1834 and Exim is allowed to try to write to its normal log files.
1837Exim version 4.21
1840 1. smtp_return_error_details was not giving details for temporary sender
1841 or receiver verification errors.
1843 2. Diagnose a configuration error if two authenticators have the same public
1844 name.
1846 3. Exim used not to create the message log file for a message until the first
1847 delivery attempt. This could be confusing when incoming messages were held
1848 for policy or load reasons. The message log file is now created at the time
1849 the message is received, and an initial "Received" line is written to it.
1851 4. The automatically generated man page for command line options had a minor
1852 bug that caused no ill effects; however, a more serious problem was that
1853 the procedure for building the man page automatically didn't always
1854 operate. Consequently, release 4.20 contains an out-of-date version. This
1855 shouldn't happen again.
1857 5. When building Exim with embedded Perl support, the script that builds the
1858 Makefile was calling 'perl' to find its compile-time parameters, ignoring
1859 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
1861 6. The freeze_tell option was not being used for messages that were frozen on
1862 arrival, either by an ACL or by local_scan().
1864 7. Added the smtp_incomplete_transaction log selector.
1866 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
1867 was accepting AUTH without a new EHLO.
1869 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
1870 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
1871 RFC.
187310. Logging of TCP/IP connections (when configured) now happens in the main
1874 daemon process instead of the child process, so that the TCP/IP connection
1875 count is more accurate (but it can never be perfect).
187711. The use of "drop" in a nested ACL was not being handled correctly in the
1878 outer ACL. Now, if condition failure induced by the nested "drop" causes
1879 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
1880 or "require"), the connection is dropped.
188212. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
1883 that yield "discard" can now be used with an "accept" or a "discard" verb,
1884 but an error is generated for any others (because I can't see a useful way
1885 to define what should happen).
188713. When an ACL is read dynamically from a file (or anywhere else), the lines
1888 are now processed in the same way as lines in the Exim configuration file.
1889 In particular, continuation lines are supported.
189114. Added the "dnslists = a.b.c!=n.n.n.n" feature.
189315. Added -ti meaning -t -i.
189516. Check for letters, digits, hyphens, and dots in the names of dnslist
1896 domains, and warn by logging if others are found.
189817. At least on BSD, alignment is not guarenteed for the array of ifreq's
1899 returned from GIFCONF when Exim is trying to find the list of interfaces on
1900 a host. The code in os.c has been modified to copy each ifreq to an aligned
1901 structure in all cases.
1903 Also, in some cases, the returned ifreq's were being copied to a 'struct
1904 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
1905 means the last couple of bytes of an IPv6 address could be chopped if the
1906 ifreq contained only a normal sockaddr (14 bytes storage).
190818. Named domain lists were not supported in the hosts_treat_as_local option.
1909 An entry such as +xxxx was not recognized, and was treated as a literal
1910 domain name.
191219. Ensure that header lines added by a DATA ACL are included in the reject log
1913 if the ACL subsequently rejects the message.
191520. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
1916 MD5 (which is deprecated).
191821. When testing a filter file using -bf, Exim was writing a message when it
1919 took the sender from a "From " line in the message, but it was not doing so
1920 when it took $return_path from a Return-Path: header line. It now does.
192222. If the contents of a "message" modifier for a "warn" ACL verb do not begin
1923 with a valid header line field name (a series of printing characters
1924 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
192623. Changed "disc" in the source to "disk" to conform to the documentation and
1927 the book and for uniformity.
192924. Ignore Sendmail's -Ooption=value command line item.
193125. When execve() failed while trying to run a command in a pipe transport,
1932 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
1933 could be confused with a return value of 69 from the command itself. This
1934 has been changed to 127, the value the shell returns if it is asked to run
1935 a non-existent command. The wording for the related log line suggests a
1936 non-existent command as the problem.
193826. If received_header_text expands to an empty string, do not add a Received:
1939 header line to the message. (Well, it adds a token one on the spool, but
1940 marks it "old" so that it doesn't get used or transmitted.)
194227. Installed eximstats 1.28 (addition of -nt option).
194428. There was no check for failure on the call to getsockname() in the daemon
1945 code. This can fail if there is a shortage of resources on the system, with
1946 ENOMEM, for example. A temporary error is now given on failure.
194829. Contrary to the C standard, it seems that in some environments, the
1949 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
1950 program. Exim now does this explicitly; it affects the formatting of
1951 timestamps using strftime().
195330. If exiqsumm was given junk data, it threw up some uninitialized variable
1954 complaints. I've now initialized all the variables, to avoid this.
195632. Header lines added by a system filter were not being "seen" during
1957 transport-time rewrites.
195933. The info_callback() function passed to OpenSSL is set up with type void
1960 (*)(SSL *, int, int), as described somewhere. However, when calling the
1961 function (actually a macro) that sets it up, the type void(*)() is
1962 expected. I've put in a cast to prevent warnings from picky compilers.
196434. If a DNS black list lookup found a CNAME record, but there were no A
1965 records associated with the domain it pointed at, Exim crashed.
196735. If a DNS black list lookup returned more than one A record, Exim ignored
1968 all but the first. It now scans all returned addresses if a particular IP
1969 value is being sought. In this situation, the contents of the
1970 $dnslist_value variable are a list of all the addresses, separated by a
1971 comma and a space.
197336. Tightened up the rules for host name lookups using reverse DNS. Exim used
1974 to accept a host name and all its aliases if the forward lookup for any of
1975 them yielded the IP address of the incoming connection. Now it accepts only
1976 those names whose forward lookup yields the correct IP address. Any other
1977 names are discarded. This closes a loophole whereby a rogue DNS
1978 administrator could create reverse DNS records to break through a
1979 wildcarded host restriction in an ACL.
198137. If a user filter or a system filter that ran in a subprocess used any of
1982 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
1983 the wrong values were passed to the pipe command ($thisaddress had the
1984 value of $0, $0 had the value of $1, etc). This bug was introduced by
1985 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
198738. Improved the line breaking for long SMTP error messages from ACLs.
1988 Previously, if there was no break point between 40 and 75 characters, Exim
1989 left the rest of the message alone. Two changes have been made: (a) I've
1990 reduced the minimum length to 35 characters; (b) if it can't find a break
1991 point between 35 and 75 characters, it looks ahead and uses the first one
1992 that it finds. This may give the occasional overlong line, but at least the
1993 remaining text gets split now.
199539. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
1996 file descriptors might be low, and that setting 1000 would always raise it.
1997 It turns out that in some environments, the limit is already over 1000 and
1998 that lowering it causes trouble. So now Exim takes care not to decrease it.
200040. When delivering a message, the value of $return_path is set to $sender_
2001 address at the start of routing (routers may change the value). By an
2002 oversight, this default was not being set up when an address was tested by
2003 -bt or -bv, which affected the outcome if any router or filter referred to
2004 $return_path.
200641. The idea of the "warn" ACL verb is that it adds a header or writes to the
2007 log only when "message" or "log_message" are set. However, if one of the
2008 conditions was an address verification, or a call to a nested ACL, the
2009 messages generated by the underlying test were being passed through. This
2010 no longer happens. The underlying message is available in $acl_verify_
2011 message for both "message" and "log_message" expansions, so it can be
2012 passed through if needed.
201442. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
2015 new expansion $bh_ to give the encoded byte string without charset
2016 translation. Translation happens only if iconv() is available; HAVE_ICONV
2017 indicates this at build time. HEADERS_CHARSET gives the charset to
2018 translate to; headers_charset can change it in the configuration, and
2019 "headers charset" can change it in an individual filter file.
202143. Now that we have a default RFC 2047 charset (see above), the code in Exim
2022 that creates RFC 2047 encoded "words" labels them as that charset instead
2023 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
2024 expansion operator; (ii) when Exim creates a From: line for a local
2025 message; (iii) when a header line is rewritten to include a "phrase" part.
202744. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
2028 buggy, causing it to skip the first lines of messages whose message ID
2029 ended in 'D'. This would not have bitten before Exim release 4.14, because
2030 message IDs were unlikely to end in 'D' before then. The effect was to have
2031 incorrect size information for certain domains.
203345. #include "config.h" was missing at the start of the crypt16.c module. This
2034 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
2035 noticed.
203746. If there was a timeout during a "random" callout check, Exim treated it as
2038 a failure of the random address, and carried on sending RSET and the real
2039 address. If the delay was just some slowness somewhere, the response to the
2040 original RCPT would be taken as a response to RSET and so on, causing
2041 mayhem of various kinds.
204347. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
2044 when I implemented it. It didn't allow for the fact that some option values
2045 may legitimatetly be negative (e.g. size_addition), and it didn't even do
2046 the right test for positive values.
204848. Domain names in DNS records are case-independent. Exim always looks them up
2049 in lower case. Some resolvers return domain names in exactly the case they
2050 appear in the zone file, that is, they may contain uppercase letters. Not
2051 all resolvers do this - some return always lower case. Exim was treating a
2052 change of case by a resolver as a change of domain, similar to a widening
2053 of a domain abbreviation. This triggered its re-routing code and so it was
2054 trying to route what was effectively the same domain again. This normally
2055 caused routing to fail (because the router wouldn't handle the domain
2056 twice). Now Exim checks for this case specially, and just changes the
2057 casing of the domain that it ultimately uses when it transmits the message
2058 envelope.
206049. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
2061 module.
206350. If a filter generated a file delivery with a non-absolute name (possible if
2064 no home directory exists for the router), the forbid_file option was not
2065 forbidding it.
206751. Added '&' feature to dnslists, to provide bit mask matching in addition to
2068 the existing equality matching.
207052. Exim was using ints instead of ino_t variables in some places where it was
2071 dealing with inode numbers.
207353. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
2074 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
2075 called TMPDIR, and if it finds it is different, it changes its value.
207754. The smtp_printf() function is now made available to local_scan() so
2078 additional output lines can be written before returning. There is also an
2079 smtp_fflush() function to enable the detection of a dropped connection.
2080 The variables smtp_input and smtp_batched_input are exported to
2081 local_scan().
208355. Changed the default runtime configuration: the message "Unknown user"
2084 has been removed from the ACL, and instead placed on the localuser router,
2085 using the cannot_route_message feature. This means that any verification
2086 failures that generate their own messages won't get overridden. Similarly,
2087 the "Unrouteable address" message that was in the ACL for unverifiable
2088 relay addresses has also been removed.
209056. Added hosts_avoid_esmtp to the smtp transport.
209257. The exicyclog script was not checking for the esoteric option
2093 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2094 will work only if exicyclog is run under the appropriate euid.
209658. Following a discussion on the list, the rules by which Exim recognises line
2097 endings on incoming messages have been changed. The -dropcr and drop_cr
2098 options are now no-ops, retained only for backwards compatibility. The
2099 following line terminators are recognized: LF CRLF CR. However, special
2100 processing applies to CR:
2102 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2103 nor a local message in the state where . is a terminator.
2105 (ii) If a bare CR is encountered in a header line, an extra space is added
2106 after the line terminator so as not to end the header. The reasoning
2107 behind this is that bare CRs in header lines are most likely either
2108 to be mistakes, or people trying to play silly games.
211059. The size of a message, as listed by "-bp" or in the Exim monitor window,
2111 was being incorrectly given as 18 bytes larger than it should have been.
2112 This is a VOB (very old bug).
211460. This may never have affected anything current, but just in case it has:
2115 When the local host is found other than at the start of a list of hosts,
2116 the local host, those with the same MX, and any that follow, are discarded.
2117 When the list in question was part of a longer list of hosts, the following
2118 hosts (not currently being processed) were also being discarded. This no
2119 longer happens. I'm not sure if this situation could ever has previously
2120 arisen.
212261. Added the "/MX" feature to lists of hosts in the manualroute and query
2123 program routers.
212562. Whenever Exim generates a new message, it now adds an Auto-Submitted:
2126 header. This is something that is recommended in a new Internet Draft, and
2127 is something that is documented as being done by Sendmail. There are two
2128 possible values. For messages generated by the autoreply transport, Exim
2129 adds:
2131 Auto-Submitted: auto-replied
2133 whereas for all other generated messages (e.g. bounces) it adds
2135 Auto-Submitted: auto-generated
213763. The "personal" condition in filters now includes a test for the
2138 Auto-Submitted: header. If it contains the string "auto-" the message it
2139 not considered personal.
214164. Added rcpt_include_affixes as a generic transport option.
214365. Added queue_only_override (default true).
214566. Added the syslog_duplication option.
214767. If what should have been the first header line of a message consisted of
2148 a space followed by a colon, Exim was mis-interpreting it as a header line.
2149 It isn't of course - it is syntactically invalid and should therefore be
2150 treated as the start of the message body. The misbehaviour could have
2151 caused a number of strange effects, including loss of data in subsequent
2152 header lines, and spool format errors.
215468. Formerly, the AUTH parameter on a MAIL command was trusted only if the
2155 client host had authenticated. This control can now be exercised by an ACL
2156 for more flexibility.
215869. By default, callouts do not happen when testing with -bh. There is now a
2159 variant, -bhc, which does actually run the callout code, including
2160 consulting and updating the callout cache.
216270. Added support for saslauthd authentication, courtesy of Alexander
2163 Sabourenkov.
216571. If statvfs() failed on the spool or log directories while checking their
2166 size for availability, Exim confusingly gave the error "space shortage".
2167 Furthermore, in debugging mode it crashed with a floating point exception.
2168 These checks are done if check_{spool,log}_{space,inodes} are set, and when
2169 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
2170 serious problem, Exim now writes to the main and panic logs when this
2171 happens, with details of the failure. It then refuses to accept the
2172 incoming message, giving the message "spool directory problem" or "log
2173 directory problem" with a 421 code for SMTP messages.
217572. When Exim is about to re-exec itself, it ensures that the file descriptors
2176 0, 1, and 2 exist, because some OS complain for execs without them (see
2177 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
2178 descriptors. However, the code omitted to check that the open succeeded,
2179 causing mysterious errors if for some reason the permissions on /dev/null
2180 got screwed. Now Exim writes a message to the main and panic logs, and
2181 bombs out if it can't open /dev/null.
218373. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
2184 interact so that it is all more flexible. It is supposed to remain
2185 backwards compatible. Also added extra_local_interfaces.
218774. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
2188 to bomb out with an assertion failure - to the client this appears as a
2189 connection drop. This problem occurs in the part of the code that was taken
2190 from the Samba project. Fortunately, the assertion is in a very simple
2191 function, so I have fixed this by reproducing the function inline in the
2192 one place where it is called, and arranging for authentication to fail
2193 instead of killing the process with assert().
219575. The SPA client code was not working when the server requested OEM rather
2196 than Unicode encoding.
219876. Added code to make require_files with a specific uid setting more usable in
2199 the case where statting the file as root fails - usually a non-root-mounted
2200 NFS file system. When this happens and the failure is EACCES, Exim now
2201 forks a subprocess and does the per-uid checking as the relevant uid.
220377. Added process_log_path.
220578. If log_file_path was not explicitly set, a setting of check_log_space or
2206 check_log_inodes was ignored.
220879. If a space check for the spool or log partitions fails, the incident is now
2209 logged. Of course, in the latter case the data may get lost...
221180. Added the %p formatting code to string_format() so that it can be used to
2212 print addresses in debug_print(). Adjusted all the address printing in the
2213 debugging in store.c to use %p rather than %d.
221581. There was a concern that a line of code in smtp_in.c could overflow a
2216 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
2217 initially expressed, the concern was not well-founded, because trailing
2218 spaces are removed early. However, if the trailing spaces were followed by
2219 a NULL, they did not get removed, so the overflow was possible. Two fixes
2220 were applied:
2222 (a) I re-wrote the offending code in a cleaner fashion.
2223 (b) If an incoming SMTP command contains a NULL character, it is rejected
2224 as invalid.
222682. When Exim changes uid/gid to the Exim user at daemon start time, it now
2227 runs initgroups(), so that if the Exim user is in any additional groups,
2228 they will be used during message reception.
2231Exim version 4.20
2234The change log for 4.20 and earlier releases has been archived.