The host_aton() buffer overflow: (1) Put a check in host_aton() itself;
[exim.git] / doc / doc-txt / ChangeLog
b975ba52 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.58 2004/12/29 10:16:52 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. The behaviour when one of the lookups defers can be
161 controlled by a keyword.
7bb56e1f 162
ea3bc19b 163 (4) It is now possible to specify the separator character for use when
7bb56e1f 164 multiple records are returned.
33397d19 165
16638. The dnslists ACL condition has been extended: it is now possible to supply
167 a list of IP addresses and/or domains to be looked up in a particular DNS
168 domain.
17039. Added log_selector=+queue_time_overall.
17240. When running the queue in the test harness, wait just a tad after forking a
173 delivery process, to get repeatability of debugging output.
17541. Include certificate and key file names in error message when GnuTLS fails
176 to set them up, because the GnuTLS error message doesn't include the name
177 of the failing file when there is a problem reading it.
17942. Allow both -bf and -bF in the same test run.
18143. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
18344. The "Exiscan patch" is now merged into the mainline Exim source.
18545. Sometimes the final signoff response after QUIT could fail to get
186 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
187 < 0 before doing a fflush(). This bug looks as though it goes back to the
188 introduction of TLS in release 3.20, but "sometimes" must have been rare
189 because the tests only now provoked it.
19146. Reset the locale to "C" after calling embedded Perl, in case it was changed
192 (this can affect the format of dates).
19447. exim_tidydb, when checking for the continued existence of a message for
195 which it has found a message-specific retry record, was not finding
196 messages that were in split spool directories. Consequently, it was
197 deleting retry records that should have stayed in existence.
19948. Steve fixed some bugs in eximstats.
20149. The SPA authentication driver was not abandoning authentication and moving
202 on to the next authenticator when an expansion was forced to fail,
203 contradicting the general specification for all authenticators. Instead it
204 was generating a temporary error. It now behaves as specified.
20650. The default ordering of permitted cipher suites for GnuTLS was pessimal
207 (the order specifies the preference for clients). The order is now AES256,
208 AES128, 3DES, ARCFOUR128.
21051. Small patch to Sieve code - explicitly set From: when generating an
211 autoreply.
21352. Exim crashed if a remote delivery caused a very long error message to be
214 recorded - for instance if somebody sent an entire SpamAssassin report back
215 as a large number of 550 error lines. This bug was coincidentally fixed by
216 increasing the size of one of Exim's internal buffers (big_buffer) that
217 happened as part of the Exiscan merge. However, to be on the safe side, I
218 have made the code more robust (and fixed the comments that describe what
219 is going on).
22153. Now that there can be additional text after "Completed" in log lines (if
222 the queue_time_overall log selector is set), a one-byte patch to exigrep
223 was needed to allow it to recognize "Completed" as not the last thing in
224 the line.
22654. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
227 patch that reportedly fixes this has been added. I am not expert enough to
228 create a test for it. This is what the patch creator wrote:
230 "I found a little strange behaviour of ldap code when working with
231 Windows 2003 AD Domain, where users was placed in more than one
232 Organization Units. When I tried to give exim partial DN, the exit code
233 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
234 But simultaneously result of request was absolutely normal ldap result,
235 so I produce this patch..."
23755. Some experimental protocols are using DNS PTR records for new purposes. The
238 keys for these records are domain names, not reversed IP addresses. The
239 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
240 leaves it alone. Component reversal etc. now happens only for IP addresses.
7102e136 241
24256. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
24457. Double the size of the debug message buffer (to 2048) so that more of very
245 long debug lines gets shown.
24758. The exicyclog utility now does better if the number of log files to keep
248 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
25059. Two changes related to the smtp_active_hostname option:
252 (1) $smtp_active_hostname is now available as a variable.
253 (2) The default for smtp_banner uses $smtp_active_hostname instead
254 of $primary_hostname.
25660. The host_aton() function is supposed to be passed a string that is known
257 to be a valid IP address. However, in the case of IPv6 addresses, it was
258 not checking this. This is a hostage to fortune. Exim now panics and dies
259 if the condition is not met. A case was found where this could be provoked
260 from a dnsdb lookup; fortuitously, this particular loophole had already
261 been fixed by change 4.50/55 above. If there are any other similar
262 loopholes, the new check should stop them being exploited.
265Exim version 4.43
268 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
269 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
270 Now it returns an integer. A number of calls were still expecting a T/F
271 return. Fortuitously, in all cases, the tests worked in OK situations,
272 which is the norm. However, things would have gone wrong on any write
273 failures on the smtp file descriptor. This function is used when sending
274 messages over SMTP and also when doing verify callouts.
276 2. When Exim is called to do synchronous delivery of a locally submitted
277 message (the -odf or -odi options), it no longer closes stderr before doing
278 the delivery.
280 3. Implemented the mua_wrapper option.
282 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
284 5. Implemented the functions header_remove(), header_testname(),
285 header_add_at_position(), and receive_remove_recipient(), and exported them
286 to local_scan().
288 6. If an ACL "warn" statement specified the addition of headers, Exim already
289 inserted X-ACL-Warn: at the start if there was no header name. However, it
290 was not making this test for the second and subsequent header lines if
291 there were newlines in the string. This meant that an invalid header could
292 be inserted if Exim was badly configured.
294 7. Allow an ACL "warn" statement to add header lines at the start or after all
295 the Received: headers, as well as at the end.
297 8. Added the rcpt_4xx retry error code.
299 9. Added postmaster_mailfrom=xxx to callout verification option.
30110. Added mailfrom=xxxx to the callout verification option, for verify=
302 header_sender only.
30411. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
305 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
30712. Inserted some casts to stop certain compilers warning when using pointer
308 differences as field lengths or precisions in printf-type calls (mostly
309 affecting debugging statements).
31113. Added optional readline() support for -be (dynamically loaded).
31314. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
314 same clock tick as a message's arrival, so that its received time was the
315 same as the "first fail" time on the retry record, and that message
316 remained on the queue past the ultimate address timeout, every queue runner
317 would try a delivery (because it was past the ultimate address timeout) but
318 after another failure, the ultimate address timeout, which should have then
319 bounced the address, did not kick in. This was a "< instead of <=" error;
320 in most cases the first failure would have been in the next clock tick
321 after the received time, and all would be well.
32315. The special items beginning with @ in domain lists (e.g. @mx_any) were not
324 being recognized when the domain list was tested by the match_domain
325 condition in an expansion string.
32716. Added the ${str2b64: operator.
32917. Exim was always calling setrlimit() to set a large limit for the number of
330 processes, without checking whether the existing limit was already
331 adequate. (It did check for the limit on file descriptors.) Furthermore,
332 errors from getrlimit() and setrlimit() were being ignored. Now they are
333 logged to the main and panic logs, but Exim does carry on, to try to do its
334 job under whatever limits there are.
33618. Imported PCRE 5.0.
33819. Trivial typo in log message " temporarily refused connection" (the leading
339 space).
34120. If the log selector return_path_on_delivery was set and an address was
342 redirected to /dev/null, the delivery process crashed because it assumed
343 that a return path would always be set for a "successful" delivery. In this
344 case, the whole delivery is bypassed as an optimization, and therefore no
345 return path is set.
34721. Internal re-arrangement: the function for sending a challenge and reading
348 a response while authentication was assuming a zero-terminated challenge
349 string. It's now changed to take a pointer and a length, to allow for
350 binary data in such strings.
35222. Added the cyrus_sasl authenticator (code supplied by MBM).
35423. Exim was not respecting finduser_retries when seeking the login of the
355 uid under which it was called; it was always trying 10 times. (The default
356 setting of finduser_retries is zero.) Also, it was sleeping after the final
357 failure, which is pointless.
35924. Implemented tls_on_connect_ports.
36125. Implemented acl_smtp_predata.
36326. If the domain in control=submission is set empty, Exim assumes that the
364 authenticated id is a complete email address when it generates From: or
365 Sender: header lines.
36727. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
368 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
369 chown and chgrp in /bin and hostname in /usr/bin.
37128. Exim was keeping the "process log" file open after each use, just as it
372 does for the main log. This opens the possibility of it remaining open for
373 long periods when the USR1 signal hits a daemon. Occasional processlog
374 errors were reported, that could have been caused by this. Anyway, it seems
375 much more sensible not to leave this file open at all, so that is what now
376 happens.
37829. The long-running daemon process does not normally write to the log once it
379 has entered its main loop, and it closes the log before doing so. This is
380 so that log files can straightforwardly be renamed and moved. However,
381 there are a couple of unusual error situations where the daemon does write
382 log entries, and I had neglected to close the log afterwards.
38430. The text of an SMTP error response that was received during a remote
385 delivery was being truncated at 512 bytes. This is too short for some of
386 the long messages that one sometimes sees. I've increased the limit to
387 1024.
38931. It is now possible to make retry rules that apply only when a message has a
390 specific sender, in particular, an empty sender.
39232. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
393 it possible to be selective about when SMTP synchronization is enforced.
39533. Added "control = caseful_local_part" and "control = "caselower_local_part".
39732. Implemented hosts_connection_nolog.
39933. Added an ACL for QUIT.
40134. Setting "delay_warning=" to disable warnings was not working; it gave a
402 syntax error.
40435. Added mailbox_size and mailbox_filecount to appendfile.
40636. Added control = no_multiline_responses to ACLs.
40837. There was a bug in the logic of the code that waits for the clock to tick
409 in the case where the clock went backwards by a substantial amount such
410 that the microsecond fraction of "now" was more than the microsecond
411 fraction of "then" (but the whole seconds number was less).
41338. Added support for the libradius Radius client library this is found on
414 FreeBSD (previously only the radiusclient library was supported).
417Exim version 4.42
420 1. When certain lookups returned multiple values in the form name=value, the
421 quoting of the values was not always being done properly. Specifically:
422 (a) If the value started with a double quote, but contained no whitespace,
423 it was not quoted.
424 (b) If the value contained whitespace other than a space character (i.e.
425 tabs or newlines or carriage returns) it was not quoted.
426 This fix has been applied to the mysql and pgsql lookups by writing a
427 separate quoting function and calling it from the lookup code. The fix
428 should probably also be applied to nisplus, ibase and oracle lookups, but
429 since I cannot test any of those, I have not disturbed their existing code.
431 2. A hit in the callout cache for a specific address caused a log line with no
432 reason for rejecting RCPT. Now it says "Previous (cached) callout
433 verification failure".
435 3. There was an off-by-one bug in the queryprogram router. An over-long
436 return line was truncated at 256 instead of 255 characters, thereby
437 overflowing its buffer with the terminating zero. As well as fixing this, I
438 have increased the buffer size to 1024 (and made a note to document this).
440 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
441 when Exim is waiting for an SMTP response from a remote server, Exim
442 restarts its select() call on the socket, thereby resetting its timeout.
443 This is not a problem when such interrupts are rare. Somebody set up a cron
444 job to run exiwhat every 2 minutes, which is less than the normal select()
445 timeout (5 or 10 minutes). This meant that the select() timeout never
446 kicked in because it was always reset. I have fixed this by comparing the
447 time when an interrupt arrives with the time at the start of the first call
448 to select(). If more time than the timeout has elapsed, the interrupt is
449 treated as a timeout.
451 5. Some internal re-factoring in preparation for the addition of Sieve
452 extensions (by MH). In particular, the "personal" test is moved to a
453 separate function, and given an option for scanning Cc: and Bcc: (which is
454 not set for Exim filters).
456 6. When Exim created an email address using the login of the caller as the
457 local part (e.g. when creating a From: or Sender: header line), it was not
458 quoting the local part when it contained special characters such as @.
460 7. Installed new OpenBSD configuration files.
462 8. Reworded some messages for syntax errors in "and" and "or" conditions to
463 try to make them clearer.
465 9. Callout options, other than the timeout value, were being ignored when
466 verifying sender addresses in header lines. For example, when using
468 verify = header_sender/callout=no_cache
470 the cache was (incorrectly) being used.
47210. Added a missing instance of ${EXE} to the exim_install script; this affects
473 only the Cygwin environment.
47511. When return_path_on_delivery was set as a log selector, if different remote
476 addresses in the same message used different return paths and parallel
477 remote delivery occurred, the wrong values would sometimes be logged.
478 (Whenever a remote delivery process finished, the return path value from
479 the most recently started remote delivery process was logged.)
48112. RFC 3848 specifies standard names for the "with" phrase in Received: header
482 lines when AUTH and/or TLS are in use. This is the "received protocol"
483 field. Exim used to use "asmtp" for authenticated SMTP, without any
484 indication (in the protocol name) for TLS use. Now it follows the RFC and
485 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
486 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
487 names appear in log lines as well as in Received: header lines.
48913. Installed MH's patches for Sieve to add the "copy" and "vacation"
490 extensions, and comparison tests, and to fix some bugs.
49214. Changes to the "personal" filter test:
494 (1) The test was buggy in that it was just doing the equivalent of
495 "contains" tests on header lines. For example, if a user's address was
496 anne@some.where, the "personal" test would incorrectly be true for
498 To: susanne@some.where
500 This test is now done by extracting each address from the header in turn,
501 and checking the entire address. Other tests that are part of "personal"
502 are now done using regular expressions (for example, to check local parts
503 of addresses in From: header lines).
505 (2) The list of non-personal local parts in From: addresses has been
506 extended to include "listserv", "majordomo", "*-request", and "owner-*",
507 taken from the Sieve specification recommendations.
509 (3) If the message contains any header line starting with "List-" it is
510 treated as non-personal.
512 (4) The test for "circular" in the Subject: header line has been removed
513 because it now seems ill-conceived.
51515. Minor typos in src/EDITME comments corrected.
51716. Installed latest exipick from John Jetmore.
51917. If headers_add on a router specified a text string that was too long for
520 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
521 of string_sprintf() is now avoided.
52318. $message_body_size was not set (it was always zero) when running the DATA
524 ACL and the local_scan() function.
52619. For the "mail" command in an Exim filter, no default was being set for
527 the once_repeat time, causing a random time value to be used if "once" was
528 specified. (If the value happened to be <= 0, no repeat happened.) The
529 default is now 0s, meaning "never repeat". The "vacation" command was OK
530 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
531 (I found it when inspecting the code).
53320. There is now an overall timeout for performing a callout verification. It
534 defaults to 4 times the callout timeout, which applies to individual SMTP
535 commands during the callout. The overall timeout applies when there is more
536 than one host that can be tried. The timeout is checked before trying the
537 next host. This prevents very long delays if there are a large number of
538 hosts and all are timing out (e.g. when the network connections are timing
539 out). The value of the overall timeout can be changed by specifying an
540 additional sub-option for "callout", called "maxwait". For example:
542 verify = sender/callout=5s,maxwait=20s
54421. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
545 to the end before writing, but this should make it even safer).
54722. Exim was forgetting that it had advertised PIPELINING for the second and
548 subsequent messages on an SMTP connection. It was also not resetting its
549 memory on STARTTLS and an internal HELO.
55123. When Exim logs an SMTP synchronization error within a session, it now
552 records whether PIPELINING has been advertised or not.
55424. Added 3 instances of "(long int)" casts to time_t variables that were being
555 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
556 rather than long int.
55825. Installed the latest Cygwin configuration files from the Cygwin maintainer.
56026. Added the never_mail option to autoreply.
563Exim version 4.41
566 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
567 crash if the getsockname() call failed; this can happen if a connection is
568 closed very soon after it is established. The problem was simply in the
569 order in which certain operations were done, causing Exim to try to write
570 to the SMTP stream before it had set up the file descriptor. The bug has
571 been fixed by making things happen in the correct order.
574Exim version 4.40
577 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
578 before the connection was closed, thus losing the rejection response.
580 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
581 some early Solaris releases, but causes trouble in current releases where
582 socklen_t is defined.
584 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
585 always exist.
587 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
588 configured.
590 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
591 that releasing the top of it at the end releases what was used for sub-
592 expansions (unless the block got too big). However, discard this block if
593 the first thing is a variable or header, so that we can use its block when
594 it is dynamic (useful for very large $message_headers, for example).
596 6. Lookups now cache *every* query, not just the most recent. A new, separate
597 store pool is used for this. It can be recovered when all lookup caches are
598 flushed. Lookups now release memory at the end of their result strings.
599 This has involved some general refactoring of the lookup sources.
601 7. Some code has been added to the store_xxx() functions to reduce the amount
602 of flapping under certain conditions.
604 8. log_incoming_interface used to affect only the <= reception log lines. Now
605 it causes the local interface and port to be added to several more SMTP log
606 lines, for example "SMTP connection from", and rejection lines.
608 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
61010. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
61211. If $host_data was set by virtue of a hosts lookup in an ACL, its value
613 could be overwritten at the end of the current message (or the start of a
614 new message if it was set in a HELO ACL). The value is now preserved for
615 the duration of the SMTP connection.
61712. If a transport had a headers_rewrite setting, and a matching header line
618 contained an unqualified address, that address was qualified, even if it
619 did not match any rewriting rules. The underlying bug was that the values
620 of the flags that permit the existence of unqualified sender and recipient
621 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
622 non-local messages, and by -bnq for local messages) were not being
623 preserved with the message after it was received.
62513. When Exim was logging an SMTP synchronization error, it could sometimes log
626 "next input=" as part of the text comprising the host identity instead of
627 the correct text. The code was using the same buffer for two different
628 strings. However, depending on which order the printing function evaluated
629 its arguments, the bug did not always show up. Under Linux, for example, my
630 test suite worked just fine.
63214. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
633 doesn't work with some older versions of Perl. It has been changed to "my",
634 which in any case is probably the better facility to use.
63615. A really picky compiler found some instances of statements for creating
637 error messages that either had too many or two few arguments for the format
638 string.
64016. The size of the buffer for calls to the DNS resolver has been increased
641 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
642 for addresses that have a lot of PTR records. This alleviates a problem; it
643 does not fully solve it.
64517. A dnsdb lookup for PTR records that receives more data than will fit in the
646 buffer now truncates the list and logs the incident, which is the same
647 action as happens when Exim is looking up a host name and its aliases.
648 Previously in this situation something unpredictable would happen;
649 sometimes it was "internal error: store_reset failed".
65118. If a server dropped the connection unexpectedly when an Exim client was
652 using GnuTLS and trying to read a response, the client delivery process
653 crashed while trying to generate an error log message.
65519. If a "warn" verb in an ACL added multiple headers to a message in a single
656 string, for example:
658 warn message = H1: something\nH2: something
660 the text was added as a single header line from Exim's point of view
661 though it ended up OK in the delivered message. However, searching for the
662 second and subsequent header lines using $h_h2: did not work. This has been
663 fixed. Similarly, if a system filter added multiple headers in this way,
664 the routers could not see them.
66620. Expanded the error message when iplsearch is called with an invalid key to
667 suggest using net-iplsearch in a host list.
66921. When running tests using -bh, any delays imposed by "delay" modifiers in
670 ACLs are no longer actually imposed (and a message to that effect is
671 output).
67322. If a "gecos" field in a passwd entry contained escaped characters, in
674 particular, if it contained a \" sequence, Exim got it wrong when building
675 a From: or a Sender: header from that name. A second bug also caused
676 incorrect handling when an unquoted " was present following a character
677 that needed quoting.
67923. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
680 was not being matched caselessly.
68224. Arranged for all hyphens in the exim.8 source to be escaped with
683 backslashes.
68525. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
686 properly. Recipient callout cache records were still being keyed to include
687 the sender, even when use_sender was set false. This led to far more
688 callouts that were necessary. The sender is no longer included in the key
689 when use_sender is false.
69126. Added "control = submission" modifier to ACLs.
69327. Added the ${base62d: operator to decode base 62 numbers.
69528. dnsdb lookups can now access SRV records.
69729. CONFIGURE_OWNER can be set at build time to define an alternative owner for
698 the configuration file.
70030. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
701 (-v) mode. This makes the output for a verbose queue run more intelligible.
70331. Added a use_postmaster feature to recipient callouts.
70532. Added the $body_zerocount variable, containing the number of binary zero
706 bytes in the message body.
70833. The time of last modification of the "new" subdirectory is now used as the
709 "mailbox time last read" when there is a quota error for a maildir
710 delivery.
71234. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
71435. Added +ignore_unknown as a special item in host lists.
71636. Code for decoding IPv6 addresses in host lists is now included, even if
717 IPv6 support is not being compiled. This fixes a bug in which an IPv6
718 address was recognized as an IP address, but was then not correctly decoded
719 into binary, causing unexpected and incorrect effects when compared with
720 another IP address.
723Exim version 4.34
726 1. Very minor rewording of debugging text in manualroute to say "list of
727 hosts" instead of "hostlist".
729 2. If verify=header_syntax was set, and a header line with an unqualified
730 address (no domain) and a large number of spaces between the end of the
731 name and the colon was received, the reception process suffered a buffer
732 overflow, and (when I tested it) crashed. This was caused by some obsolete
733 code that should have been removed. The fix is to remove it!
735 3. When running in the test harness, delay a bit after writing a bounce
736 message to get a bit more predictability in the log output.
738 4. Added a call to search_tidyup() just before forking a reception process. In
739 theory, someone could use a lookup in the expansion of smtp_accept_max_
740 per_host which, without the tidyup, could leave open a database connection.
742 5. Added the variables $recipient_data and $sender_data which get set from a
743 lookup success in an ACL "recipients" or "senders" condition, or a router
744 "senders" option, similar to $domain_data and $local_part_data.
746 6. Moved the writing of debug_print from before to after the "senders" test
747 for routers.
749 7. Change 4.31/66 (moving the time when the Received: is generated) caused
750 problems for message scanning, either using a data ACL, or using
751 local_scan() because the Received: header was not generated till after they
752 were called (in order to set the time as the time of reception completion).
753 I have revised the way this works. The header is now generated after the
754 body is received, but before the ACL or local_scan() are called. After they
755 are run, the timestamp in the header is updated.
758Exim version 4.33
761 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
762 before starting a queue runner without re-exec. This happened only when
763 deliver_drop_privilege was set or when the Exim user was set to root. The
764 effect of the bug was that timeouts during subsequent deliveries caused
765 crashes instead of being properly handled. The handler is now left at its
766 default (and expected) setting.
768 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
769 message, again when deliver_drop_privilege is set or Exim is run as root.
770 The bug described in (1) was not present in this case, but the tidying up
771 of the other signals was missing. I have made the two cases consistent.
773 3. The ignore_target_hosts setting on a manualroute router was being ignored
774 for hosts that were looked up using the /MX notation.
776 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
777 in domain lists.
779 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
780 operated on the sender address. After changing the $sender_address to <>
781 for the sender address verify, Exim was re-instated it as the original
782 (before rewriting) address, but remembering that it had rewritten it, so it
783 wasn't rewriting it again. This bug also had the effect of breaking the
784 sender address verification caching when the sender address was rewritten.
786 6. The ignore_target_hosts option was being ignored by the ipliteral router.
787 This has been changed so that if the ip literal address matches
788 ignore_target_hosts, the router declines.
790 7. Added expansion conditions match_domain, match_address, and match_local_
791 part (NOT match_host).
793 8. The placeholder for the Received: header didn't have a length field set.
795 9. Added code to Exim itself and to exim_lock to test for a specific race
796 condition that could lead to file corruption when using MBX delivery. The
797 issue is with the lockfile that is created in /tmp. If this file is removed
798 after a process has opened it but before that process has acquired a lock,
799 there is the potential for a second process to recreate the file and also
800 acquire a lock. This could lead to two Exim processes writing to the file
801 at the same time. The added code performs the same test as UW imapd; it
802 checks after acquiring the lock that its file descriptor still refers to
803 the same named file.
80510. The buffer for building added header lines was of fixed size, 8192 bytes.
806 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
807 when Exim is built.
80911. Added the smtp_active_hostname option. If used, this will typically be made
810 to depend on the incoming interface address. Because $interface_address is
811 not set up until the daemon has forked a reception process, error responses
812 that can happen earlier (such as "too many connections") no longer contain
813 a host name.
81512. If an expansion in a condition on a "warn" statement fails because a lookup
816 defers, the "warn" statement is abandoned, and the next ACL statement is
817 processed. Previously this caused the whole ACL to be aborted.
81913. Added the iplsearch lookup type.
82114. Added ident_timeout as a log selector.
82315. Added tls_certificate_verified as a log selector.
82516. Added a global option tls_require_ciphers (compare the smtp transport
826 option of the same name). This controls incoming TLS connections.
82817. I finally figured out how to make tls_require_ciphers do a similar thing
829 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
830 before starting the TLS session.
83218. Tabs are now shown as \t in -bP output.
83419. If the log selector return_path_on_delivery was set, Exim crashed when
835 bouncing a message because it had too many Received: header lines.
83720. If two routers both had headers_remove settings, and the first one included
838 a superfluous trailing colon, the final name in the first list and the
839 first name in the second list were incorrectly joined into one item (with a
840 colon in the middle).
843Exim version 4.32
846 1. Added -C and -D options to the exinext utility, mainly to make it easier
847 to include in the automated testing, but these could be helpful when
848 multiple configurations are in use.
850 2. The exinext utility was not formatting the output nicely when there was
851 an alternate port involved in the retry record key, nor when there was a
852 message id as well (for retries that were specific to a specific message
853 and a specific host). It was also confused by IPv6 addresses, because of
854 the additional colons they contain. I have fixed the IPv4 problem, and
855 patched it up to do a reasonable job for IPv6.
857 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
858 delivery, the log line now contains "pipelined" if PIPELINING was used.
860 4. An SMTP transport process used to panic and die if the bind() call to set
861 an explicit outgoing interface failed. This has been changed; it is now
862 treated in the same way as a connect() failure.
864 5. A reference to $sender_host_name in the part of a conditional expansion
865 that was being skipped was still causing a DNS lookup. This no longer
866 occurs.
868 6. The def: expansion condition was not recognizing references to header lines
869 that used bh_ and bheader_.
871 7. Added the _cache feature to named lists.
873 8. The code for checking quota_filecount in the appendfile transport was
874 allowing one more file than it should have been.
876 9. For compatibility with Sendmail, the command line option
878 -prval:sval
880 is equivalent to
882 -oMr rval -oMs sval
884 and sets the incoming protocol and host name (for trusted callers). The
885 host name and its colon can be omitted when only the protocol is to be set.
886 Note the Exim already has two private options, -pd and -ps, that refer to
887 embedded Perl. It is therefore impossible to set a protocol value of "d" or
888 "s", but I don't think that's a major issue.
89010. A number of refactoring changes to the code, none of which should affect
891 Exim's behaviour:
893 (a) The number of logging options was getting close to filling up the
894 32-bit word that was used as a bit map. I have split them into two classes:
895 those that are passed in the argument to log_write(), and those that are
896 only ever tested independently outside of that function. These are now in
897 separate 32-bit words, so there is plenty of room for expansion again.
898 There is no change in the user interface or the logging behaviour.
900 (b) When building, for example, log lines, the code previously used a
901 macro that called string_cat() twice, in order to add two strings. This is
902 not really sufficiently general. Furthermore, there was one instance where
903 it was actually wrong because one of the argument was used twice, and in
904 one call a function was used. (As it happened, calling the function twice
905 did not affect the overall behaviour.) The macro has been replaced by a
906 function that can join an arbitrary number of extra strings onto a growing
907 string.
909 (c) The code for expansion conditions now uses a table and a binary chop
910 instead of a serial search (which was left over from when there were very
911 few conditions). Also, it now recognizes conditions like "pam" even when
912 the relevant support is not compiled in: a suitably worded error message is
913 given if an attempt is made to use such a condition.
91511. Added ${time_interval:xxxxx}.
91712. A bug was causing one of the ddress fields not to be passed back correctly
918 from remote delivery subprocesses. The field in question was not being
919 subsequently used, so this caused to problems in practice.
92113. Added new log selectors queue_time and deliver_time.
92314. Might have fixed a bug in maildirsizefile handling that threw up
924 "unexpected character" debug warnings, and recalculated the data
925 unnecessarily. In any case, I expanded the warning message to give more
926 information.
92815. Added the message "Restricted characters in address" to the statements in
929 the default ACL that block characters like @ and % in local parts.
93116. Change 71 for release 4.31 proved to be much less benign that I imagined.
932 Three changes have been made:
934 (a) There was a serious bug; a negative response to MAIL caused the whole
935 recipient domain to be cached as invalid, thereby blocking all messages
936 to all local parts at the same domain, from all senders. This bug has
937 been fixed. The domain is no longer cached after a negative response to
938 MAIL if the sender used is not empty.
940 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
941 been restored.
943 (c) A new callout option, "use_sender" has been added for people who want
944 the modified behaviour.
947Exim version 4.31
950 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
951 Larry Rosenman.
953 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
954 indeed breaks things for older releases.
956 3. Added additional logging to the case where there is a problem reading data
957 from a filter that is running in a subprocess using a pipe, in order to
958 try to track down a specific problem.
960 4. Testing facility fudge: when running in the test harness and attempting
961 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
962 getting "No route to host". Convert this to a timeout.
964 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
965 warning.
967 6. Some OS don't have socklen_t but use size_t instead. This affects the
968 fifth argument of getsockopt() amongst other things. This is now
969 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
970 can be set for individual OS. I have set it for SunOS5, OSF1, and
971 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
972 some earlier ones do not.
974 7. Change 4.30/15 was not doing the test caselessly.
976 8. The standard form for an IPv6 address literal was being rejected by address
977 parsing in, for example, MAIL and RCPT commands. An example of this kind of
978 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
979 this, as well as the form without the "IPv6" on the front (but only when
980 address literals are enabled, of course).
982 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
98410. Exim crashed if a message with an empty sender address specified by -f
985 encountered a router with an errors_to setting. This could be provoked only
986 by a command such as
988 exim -f "" ...
990 where an empty string was supplied; "<>" did not hit this bug.
99211. Installed PCRE release 4.5.
99412. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
995 remained set. It is now erased.
99713. exiqgrep wasn't working on MacOS X because it didn't correctly compute
998 times from message ids (which are base 36 rather than the normal 62).
100014. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
1001 were being counted as actual protocol errors, and logged if the log
1002 selector +smtp_protocol_error was set. One cannot be perfect in this test,
1003 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
1004 and DATA following a set of rejected RCPTs do not count as protocol errors.
1005 In other words, Exim assumes they were pipelined, though this may not
1006 actually be the case. Of course, in all cases the client gets an
1007 appropriate error code.
100915. If a lookup fails in an ACL condition, a message about the failure may
1010 be available; it is used if testing the ACL cannot continue, because most
1011 such messages specify what the cause of the deferral is. However, some
1012 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
1013 that caused an old message to be retained and used if a later statement
1014 caused a defer, replacing the real cause of the deferral.
101616. If an IP address had so many PTR records that the DNS lookup buffer
1017 was not large enough to hold them, Exim could crash while trying to process
1018 the truncated data. It now detects and logs this case.
102017. Further to 4.21/58, another change has been made: if (and only if) the
1021 first line of a message (the first header line) ends with CRLF, a bare LF
1022 in a subsequent header line has a space inserted after it, so as not to
1023 terminate the header.
102518. Refactoring: tidied an ugly bit of code in appendfile that copied data
1026 unnecessarily, used atoi() instead of strtol(), and didn't check the
1027 termination when getting file sizes from file names by regex.
102919. Completely re-implemented the support for maildirsize files, in the light
1030 of a number of problems with the previous contributed implementation
1031 (4.30/29). In particular:
1033 . If the quota is zero, the maildirsize file is maintained, but no quota is
1034 imposed.
1036 . If the maildir directory does not exist, it is created before any attempt
1037 to write a maildirsize file.
1039 . The quota value in the file is just a cache; if the quota is changed in
1040 the transport, the new value overrides.
1042 . A regular expression is available for excluding directories from the
1043 count.
104520. The autoreply transport checks the characters in options that define the
1046 message's headers; it allows continued headers, but it was checking with
1047 isspace() after an embedded newline instead of explicitly looking for a
1048 space or a tab.
105021. If all the "regular" hosts to which an address was routed had passed their
1051 expiry times, and had not reached their retry times, the address was
1052 bounced, even if fallback hosts were defined. Now Exim should go on to try
1053 the fallback hosts.
105522. Increased buffer sizes in the callout code from 1024 to 4096 to match the
1056 equivalent code in the SMTP transport. Some hosts send humungous responses
1057 to HELO/EHLO, more than 1024 it seems.
105923. Refactoring: code in filter.c used (void *) for "any old type" but this
1060 gives compiler warnings in some environments. I've now done it "properly",
1061 using a union.
106324. The replacement for inet_ntoa() that is used with gcc on IRIX systems
1064 (because of problems with the built-in one) was declared to return uschar *
1065 instead of char *, causing compiler failure.
106725. Fixed a file descriptor leak when processing alias/forward files.
106926. Fixed a minor format string issue in dbfn.c.
107127. Typo in exim.c: ("dmbnz" for "dbmnz").
107328. If a filter file refered to $h_xxx or $message_headers, and the headers
1074 contained RFC 2047 "words", Exim's memory could, under certain conditions,
1075 become corrupted.
107729. When a sender address is verified, it is cached, to save repeating the test
1078 when there is more than one recipient in a message. However, when the
1079 verification involves a callout, it is possible for different callout
1080 options to be set for different recipients. It is too complicated to keep
1081 track of this in the cache, so now Exim always runs a verification when a
1082 callout is required, relying on the callout cache for the optimization.
1083 The overhead is duplication of the address routing, but this should not be
1084 too great.
108630. Fixed a bug in callout caching. If a RCPT command caused the sender address
1087 to be verified with callout=postmaster, and the main callout worked but the
1088 postmaster check failed, the verification correctly failed. However, if a
1089 subsequent RCPT command asked for sender verification *without* the
1090 postmaster check, incorrect caching caused this verification also to fail,
1091 incorrectly.
109331. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
1094 it was not caching the DNS options (qualify_single, search_parents) that
1095 were used when the lookup failed. A subsequent lookup with different
1096 options therefore always gave the same answer, though there were cases
1097 where it should not have. (Example: a "domains = !$mx_any" option on a
1098 dnslookup router: the "domains" option is always processed without any
1099 widening, but the router might have qualify_single set.) Now Exim uses the
1100 cached value only when the same options are set.
110232. Added John Jetmore's "exipick" utility to the distribution.
110433. GnuTLS: When an attempt to start a TLS session fails for any reason other
1105 than a timeout (e.g. a certificate is required, and is not provided), an
1106 Exim server now closes the connection immediately. Previously it waited for
1107 the client to close - but if the client is SSL, it seems that they each
1108 wait for each other, leading to a delay before one of them times out.
111034: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
1111 maintained 0.8.x compatibility because I don't think many are using it, and
1112 it is clearly obsolete.
111435. Added TLS support for CRLs: a tls_crl global option and one for the smtp
1115 transport.
111736. OpenSSL: $tls_certificate_verified was being set to 1 even if the
1118 client certificate was expired. A simple patch fixes this, though I don't
1119 understand the full logic of why the verify callback is called multiple
1120 times.
112237. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
1123 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
1124 which causes problems with some clients (such as the Certicom SSL Plus
1125 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
1126 disables the coutermeasure allowing Eudora to connect."
112838. Exim was not checking that a write() to a log file succeeded. This could
1129 lead to Bad Things if a log got too big, in particular if it hit a file
1130 size limit. Exim now panics and dies if it cannot write to a log file, just
1131 as it does if it cannot open a log file.
113339. Modified OS/Makefile-Linux so that it now contains
1137 The two -D definitions ensure that Exim is compiled with large file
1138 support, which makes it possible to handle log files that are bigger than
1139 2^31.
114140. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
1142 instance) a domain was checked against a named list that involved a lookup,
1143 causing $domain_data to be set, then another domain was checked against the
1144 same list, then the first domain was re-checked, the value of $domain_data
1145 after the final check could be wrong. In particular, if the second check
1146 failed, it could be set empty. This bug probably also applied to
1147 $localpart_data.
114941. The strip_trailing_dot option was not being applied to the address given
1150 with the -f command-line option.
115242. The code for reading a message's header from the spool was incrementing
1153 $received_count, but never initializing it. This meant that the value was
1154 incorrect (doubled) while delivering a message in the same process in which
1155 it was received. In the most common configuration of Exim, this never
1156 happens - a fresh exec is done - but it can happen when
1157 deliver_drop_privilege is set.
115943. When Exim logs an SMTP synchronization error - client data sent too soon -
1160 it now includes up to 150 characters of the unexpected data in the log
1161 line.
116344. The exim_dbmbuild utility uses fixed size buffers for reading input lines
1164 and building data strings. The size of both of these buffers was 10 000
1165 bytes - far larger than anybody would *ever* want, thought I. Needless to
1166 say, somebody hit the limit. I have increased the maximum line length to
1167 20 000 and the maximum data length of concatenated lines to 100 000. I have
1168 also fixed two bugs, because there was no checking on these buffers. Tsk,
1169 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1170 buffer is too small.
117245. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1173 lsearch lookups. Now it does.
117546. When parsing a route_list item in a manualroute router, a fixed-length
1176 buffer was used for the list of hosts. I made this 1024 bytes long,
1177 thinking that nobody would ever have a list of hosts that long. Wrong.
1178 Somebody had a whole pile of complicated expansion conditions, and the
1179 string was silently truncated, leading to an expansion error. It turns out
1180 that it is easier to change to an unlimited length (owing to other changes
1181 that have happened since this code was originally written) than to build
1182 structure for giving a limitation error. The length of the item that
1183 expands into the list of hosts is now unlimited.
118547. The lsearch lookup could not handle data where the length of text line was
1186 more than 4095 characters. Such lines were truncated, leading to shortened
1187 data being returned. It should now handle lines of any length.
118948. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
1190 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
1191 ACL").
119349. Cosmetic tidy to scripts like exicyclog that are generated by globally
1194 replacing strings such as BIN_DIRECTORY in a source file: the replacement
1195 no longer happens in comment lines. A list of replacements is now placed
1196 at the head of all of the source files, except those whose only change is
1197 to replace PERL_COMMAND in the very first #! line.
119950. Replaced the slow insertion sort in queue.c, for sorting the list of
1200 messages on the queue, with a bottom-up merge sort, using code contributed
1201 by Michael Haardt. This should make operations like -bp somewhat faster on
1202 large queues. It won't affect queue runners, except when queue_run_in_order
1203 is set.
120551. Installed eximstats 1.31 in the distribution.
120752. Added support for SRV lookups to the dnslookup router.
120953. If an ACL referred to $message_body or $message_body_end, the value was not
1210 reset for any messages that followed in the same SMTP session.
121254. The store-handling optimization for building very long strings was not
1213 differentiating between the different store pools. I don't think this
1214 actually made any difference in practice, but I've tidied it.
121655. While running the routers to verify a sender address, $sender_address
1217 was still set to the sender address. This is wrong, because when routing to
1218 send a bounce to the sender, it would be empty. Therefore, I have changed
1219 it so that, while verifying a sender address, $sender_address is set to <>.
1220 (There is no change to what happens when verifying a recipient address.)
122256. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
1223 target A or AAAA records (if not already returned) without resetting the
1224 qualify_single or search_parents options of the DNS resolver. These are
1225 inappropriate in this case because the targets of MX and SRV records must
1226 be FQDNs. A broken DNS record could cause trouble if it happened to have a
1227 target that, when qualified, matched something in the local domain. These
1228 two options are now turned off when doing these lookups.
123057. It seems that at least some releases of Reiserfs (which does not have the
1231 concept of a fixed number of inodes) returns zero and not -1 for the
1232 number of available inodes. This interacted badly with check_spool_inodes,
1233 which assumed that -1 was the "no such thing" setting. What I have done is
1234 to check that the total number of inodes is greater than zero before doing
1235 the test of how many are available.
123758. When a "warn" ACL statement has a log_message modifier, the message is
1238 remembered, and not repeated. This is to avoid a lot of repetition when a
1239 message has many recipients that cause the same warning to be written.
1240 Howewer, Exim was preserving the list of already written lines for an
1241 entire SMTP session, which doesn't seem right. The memory is now reset if a
1242 new message is started.
124459. The "rewrite" debugging flag was not showing the result of rewriting in the
1245 debugging output unless log_rewrite was also set.
124760. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
1248 of (int)(handle) when we know that handle contains (void *)(-1).
125061. The Exim daemon panic-logs an error return when it closes the incoming
1251 connection. However "connection reset by peer" seems to be common, and
1252 isn't really an error worthy of noting specially, so that particular error
1253 is no long logged.
125562. When Exim is trying to find all the local interfaces, it used to panic and
1256 die if the ioctl to get the interface flags failed. However, it seems that
1257 on at least one OS (Solaris 9) it is possible to have an interface that is
1258 included in the list of interfaces, but for which you get a failure error
1259 for this call. This happens when the interface is not "plumbed" into a
1260 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
1261 failure of the "get flags" call assumes that the interface is down.
126363. Added a ${eval10: operator, which assumes all numbers are decimal. This
1264 makes life easier for people who are doing arithmetic on fields extracted
1265 from dates, where you often get leading zeros that should not be
1266 interpreted as octal.
126864. Added qualify_domain to the redirect router, to override the global
1269 setting.
127165. If a pathologically long header line contained very many addresses (the
1272 report of this problem mentioned 10 000) and each of them was rewritten,
1273 Exim could use up a very large amount of memory. (It kept on making new
1274 copies of the header line as it rewrote, and never released the old ones.)
1275 At the expense of a bit more processing, the header rewriting function has
1276 been changed so that it no longer eats memory in this way.
127866. The generation of the Received: header has been moved from the time that a
1279 message starts to be received, to the time that it finishes. The timestamp
1280 in the Received: header should now be very close to that of the <= log
1281 line. There are two side-effects of this change:
1283 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
1284 logged header lines no longer include the local Received: line, because
1285 it has not yet been created. The same applies to a copy of the message
1286 that is returned to a non-SMTP sender when a message is rejected.
1288 (b) When a filter file is tested using -bf, no additional Received: header
1289 is added to the test message. After some thought, I decided that this
1290 is a bug fix.
1292 This change does not affect the value of $received_for. It is still set
1293 after address rewriting, but before local_scan() is called.
129567. Installed the latest Cygwin-specific files from the Cygwin maintainer.
129768. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
1298 gave an unhelpful panic error message, and a defer error. I have managed to
1299 change this behaviour so that it now rejects any supplied certificate,
1300 which seems right, as the list of acceptable certificates is empty.
130269. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
1303 gave an unhelpful defer error. I have not managed to make this reject any
1304 supplied certificates, but the error message it gives is "no certificate
1305 supplied", which is not helpful.
130770. exigrep's output now also includes lines that are not associated with any
1308 message, but which match the given pattern. Implemented by a patch from
1309 Martin Sluka, which also tidied up the Perl a bit.
131171. Recipient callout verification, like sender verification, was using <> in
1312 the MAIL FROM command. This isn't really the right thing, since the actual
1313 sender may affect whether the remote host accepts the recipient or not. I
1314 have changed it to use the actual sender in the callout; this means that
1315 the cache record is now keyed on a recipient/sender pair, not just the
1316 recipient address. There doesn't seem to be a real danger of callout loops,
1317 since a callout by the remote host to check the sender would use <>.
1318 [SEE ABOVE: changed after hitting problems.]
132072. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
1321 temporary errors. However, in the case of such a code being given after
1322 the end of a data transmission (i.e. after ".") Exim was failing to write
1323 a retry record for the message. (Yes, there was some broken host that was
1324 actually sending 8xx at this point.)
132673. An unknown lookup type in a host list could cause Exim to panic-die when
1327 the list was checked. (An example that provoked this was putting <; in the
1328 middle of a list instead of at the start.) If this happened during a DATA
1329 ACL check, a -D file could be left lying around. This kind of configuration
1330 error no longer causes Exim to die; instead it causes a defer errror. The
1331 incident is still logged to the main and panic logs.
133374. Buglet left over from Exim 3 conversion. The message "too many messages
1334 in one connection" was written to the rejectlog but not the mainlog, except
1335 when address rewriting (yes!) was being logged.
133775. Added write_rejectlog option.
133976. When a system filter was run not as root (that is, when system_filter_user
1340 was set), the values of the $n variables were not being returned to the
1341 main process; thus, they were not subsequently available in the $sn
1342 variables.
134477. Added +return_path_on_delivery log selector.
134678. A connection timeout was being treated differently from recipients deferred
1347 when testing hosts_max_try with a message that was older than the host's
1348 retry timeout. (The host should not be counted, thus allowing all hosts to
1349 be tried at least once before bouncing.) This may have been the cause of an
1350 occasionally reported bug whereby a message would remain on the queue
1351 longer than the retry timeout, but would be bounced if a delivery was
1352 forced. I say "may" because I never totally pinned down the problem;
1353 setting up timeout/retry tests is difficult. See also the next item.
135579. The ultimate address timeout was not being applied to errors that involved
1356 a combination of host plus message (for example, a timeout on a MAIL
1357 command). When an address resolved to a number of possible hosts, and they
1358 were not all tried for each delivery (e.g. because of hosts_max_try), a
1359 message could remain on the queue longer than the retry timeout.
136180. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
1362 Haardt.
136481. Fixed an obscure SMTP outgoing bug which required at least the following
1365 conditions: (a) there was another message waiting for the same server;
1366 (b) the server returned 5xx to all RCPT commands in the first message so
1367 that the message was not completed; (c) the server dropped the connection
1368 or gave a negative response to the RSET that Exim sends to abort the
1369 transaction. The observed case was a dropped connection after DATA that had
1370 been sent in pipelining mode. That is, the server had advertised PIPELINING
1371 but was not implementing it correctly. The effect of the bug was incorrect
1372 behaviour, such as trying another host, and this could lead to a crash.
1375Exim version 4.30
1378 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
1379 and daemon.c were passed as pointers to ints; they should have been
1380 pointers to socklen_t variables (which are typically unsigned ints).
1382 2. Some signed/unsigned type warnings in the os.c file for Linux have been
1383 fixed.
1385 3. Fixed a really odd bug that affected only the testing scheme; patching a
1386 certain fixed string in the binary changed the value of another string that
1387 happened to be identical to the end of the original first string.
1389 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
1390 name", it returns that address as the IP address. On some operating
1391 systems (e.g. Solaris), it also passes back the IP address string as the
1392 "host name". However, on others (e.g. Linux), it passes back an empty
1393 string. Exim wasn't checking for this, and was changing the host name to an
1394 empty string, assuming it had been canonicized.
1396 5. Although rare, it is permitted to have more than one PTR record for a given
1397 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
1398 all the names associated with an address, because they do in Solaris.
1399 However, it seems that they do not in Linux for data that comes from the
1400 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
1401 I found this out when I moved to a new Linux workstation and tried to run
1402 the Exim test suite.
1404 To get round this problem I have changed the code so that it now does its
1405 own call to the DNS to look up PTR records when searching for a host name.
1406 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
1407 addresses that are only in /etc/hosts are still found.
1409 This behaviour is, however, controlled by an option called host_lookup_
1410 order, which defaults to "bydns:byaddr". If people want to use the other
1411 order, or indeed, just use one or the other means of lookup, they can
1412 specify it in this variable.
1414 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
1415 some operating systems, this comes back from gethostbyaddr() as an empty
1416 string, and this is what Exim used to test for. However, it seems that in
1417 other systems, "." is yielded. Exim now tests for this case too.
1419 7. The values of check_spool_space and check_log_space are now held internally
1420 as a number of kilobytes instead of an absolute number of bytes. If a
1421 numbers is specified without 'K' or 'M', it is rounded up to the nearest
1422 kilobyte. This means that much larger values can be stored.
1424 8. Exim monitor: an attempt to get the action menu when not actually pointing
1425 at a message produces an empty menu entitled "No message selected". This
1426 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
1427 no entries in it ("Shell widget menu has zero width and/or height"). So I
1428 have added a single, blank menu entry in this case.
1430 9. Added ${quote_local_part.
143210. MIME decoding is now applied to the contents of Subject: header lines when
1433 they are logged.
143511. Now that a reference to $sender_host_address automatically causes a reverse
1436 lookup to occur if necessary (4.13/18), there is no need to arrange for a
1437 host lookup before query-style lookups in lists that might use this
1438 variable. This has therefore been abolished, and the "net-" prefix is no
1439 longer necessary for query-style lookups.
144112. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
1442 have been a typo for LFLAGS, so it has been changed.
144413. The install script calls Exim with "-C /dev/null" in order to find the
1445 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
1446 to be output. Howeve, since Exim outputs its version number before the
1447 error, it didn't break the script. It just looked ugly. I fixed this by
1448 always allowing "-C /dev/null" if the caller is root.
145014. Ignore overlarge ACL variable number when reading spool file - insurance
1451 against a later release with more variables having written the file.
145315. The standard form for an IPv6 address literal was being rejected by EHLO.
1454 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1455 this, as well as the form without the "IPv6" on the front.
145716. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
1458 OS/Makefile-Darwin file.
146017. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
1461 only to LDAP libraries that do not have LDAP_OPT_DEREF.
146318. After change 4.21/52, "%ld" was used to format the contents of the $inode
1464 variable. However, some OS use ints for inodes. I've added cast to long int
1465 to get rid of the compiler warning.
146719. I had forgotten to lock out "/../" in configuration file names when
1468 ALT_CONFIG_PREFIX was set.
147020. Routers used for verification do not need to specify transports. However,
1471 if such a router generated a host list, and callout was configured, Exim
1472 crashed, because it could not find a port number from the (non-existent)
1473 transport. It now assumes port 25 in this circumstance.
147521. Added the -t option to exigrep.
147722. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
1478 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
1479 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
1480 them is set, LOOKUP_LSEARCH is forced.
148223. "exim -bV" now outputs a list of lookups that are included in the binary.
148424. Added sender and host information to the "rejected by local_scan()" log
1485 line; previously there was no indication of these.
148725. Added .include_if_exists.
148926. Change 3.952/11 added an explicit directory sync on top of a file sync for
1490 Linux. It turns out that not all file systems support this. Apparently some
1491 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
1492 do it.) To cope with this, the error EINVAL, which means that sync-ing is
1493 not supported on the file descriptor, is now ignored when Exim is trying to
1494 sync a directory. This applies only to Linux.
149627. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
149828. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
1499 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
1500 this.
150229. Added support for maildirsize files from supplied patch (modified a bit).
150430. The use of :fail: followed by an empty string could lead Exim to respond to
1505 sender verification failures with (e.g.):
1507 550 Verification failed for <xxx>
1508 550 Sender verify failed
1510 where the first response line was missing the '-' that indicates it is not
1511 the final line of the response.
151331. The loop for finding the name of the user that called Exim had a hardwired
1514 limit of 10; it now uses the value of finduser_retries, which is used for
1515 all other user lookups.
151732. Added $received_count variable, available in data and not_smtp ACLs, and at
1518 delivery time.
152033. Exim was neglecting to zero errno before one call of strtol() when
1521 expanding a string and expecting an integer value. On some systems this
1522 resulted in spurious "integer overflow" errors. Also, it was casting the
1523 result into an int without checking.
152534. Testing for a connection timeout using "timeout_connect" in the retry rules
1526 did not work. The code looks as if it has *never* worked, though it appears
1527 to have been documented since at least releast 1.62. I have made it work.
152935. The "timeout_DNS" error in retry rules, also documented since at least
1530 1.62, also never worked. As it isn't clear exactly what this means, and
1531 clearly it isn't a major issue, I have abolished the feature by treating it
1532 as "timeout", and writing a warning to the main and panic logs.
153436. The display of retry rules for -brt wasn't always showing the error code
1535 correctly.
153737. Added new error conditions to retry rules: timeout_A, timeout_MX,
1538 timeout_connect_A, timeout_connect_MX.
154038. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
1541 to the empty sender.
154339. The daemon was not analysing the content of -oX till after it had closed
1544 stderr and disconnected from the controlling terminal. This meant that any
1545 syntax errors were only noted on the panic log, and the return code from
1546 the command was 0. By re-arranging the code a little, I've made the
1547 decoding happen first, so such errors now appear on stderr, and the return
1548 code is 1. However, the actual setting up of the sockets still happens in
1549 the disconnected process, so errors there are still only recorded on the
1550 panic log.
155240. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
1553 connections (as happens on some IP stacks) was logged at start up time as
1554 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
1555 it from "IPv6 and IPv4", which means that two separate sockets are being
1556 used.
155841. The debug output for gethostbyname2() or getipnodebyname() failures now
1559 says whether AF_INET or AF_INET6 was passed as an argument.
156142. Exiwhat output was messed up when time zones were included in log
1562 timestamps.
156443. Exiwhat now gives more information about the daemon's listening ports,
1565 and whether -tls-on-connect was used.
156744. The "port" option of the smtp transport is now expanded.
156945. A "message" modifier in a "warn" statement in a non-message ACL was being
1570 silently ignored. Now an error message is written to the main and panic
1571 logs.
157346. There's a new ACL modifier called "logwrite" which writes to a log file
1574 as soon as it is encountered.
157647. Added $local_user_uid and $local_user_gid at routing time.
157848. Exim crashed when trying to verify a sender address that was being
1579 rewritten to "<>".
158149. Exim was recognizing only a space character after ".include". It now also
1582 recognizes a tab character.
158450. Fixed several bugs in the Perl script that creates the exim.8 man page by
1585 extracting the relevant information from the specification. The man page no
1586 longer contains scrambled data for the -d option, and I've added a section
1587 at the front about calling Exim under different names.
158951. Added "extra_headers" argument to the "mail" command in filter files.
159152. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
1592 crash.
159453. Installed eximstats 1.29.
159654. Added transport_filter_timeout as a generic transport option.
159855. Exim no longer adds an empty Bcc: header to messages that have no To: or
1599 Cc: header lines. This was required by RFC 822, but it not required by RFC
1600 2822.
160256. Exim used to add From:, Date:, and Message-Id: header lines to any
1603 incoming messages that did not have them. Now it does so only if the
1604 message originates locally, that is, if there is no associated remote host
1605 address. When Resent- header lines are present, this applies to the Resent-
1606 lines rather than the non-Resent- lines.
160857. Drop incoming SMTP connection after too many syntax or protocol errors. The
1609 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
161158. Messages for configuration errors now include the name of the main
1612 configuration file - useful now that there may be more than one file in a
1613 list (.included file names were always shown).
161559. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
1616 for those rare installations that do not start the daemon as root or run it
1617 setuid root. I've cut out the call to initgroups() if the daemon is not
1618 root at that time.
162060. The Exim user and group can now be bound into the binary as text strings
1621 that are looked up at the start of Exim's processing.
162361. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
162562. Added $mailstore_basename variable.
162763. Installed patch to sieve.c from Michael Haardt.
162964. When Exim failed to open the panic log after failing to open the main log,
1630 the original message it was trying to log was written to stderr and debug
1631 output, but if they were not available (the usual case in production), it
1632 was lost. Now it is written to syslog before the two lines that record the
1633 failures to open the logs.
163565. Users' Exim filters run in subprocesses under the user's uid. It is
1636 possible for a "deliver" command or an alias in a "personal" command to
1637 provoke an address rewrite. If logging of address rewriting is configured,
1638 this fails because the process is not running as root or exim. There may be
1639 a better way of dealing with this, but for the moment (because 4.30 needs
1640 to be released), I have disabled address rewrite logging when running a
1641 filter in a non-root, non-exim process.
1644Exim version 4.24
1647 1. The buildconfig auxiliary program wasn't quoting the value set for
1648 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
1649 not defined. This bug was masked in 4.22 by the effect that was fixed in
1650 change 4.23/1.
1652 2. Some messages that were rejected after a message id was allocated were
1653 shown as "incomplete" by exigrep. It no longer does this for messages that
1654 are rejected by local_scan() or the DATA or non-SMTP ACLs.
1656 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
1657 have allow_domain_literals set, the ID did not get logged in the <= line.
1658 Domain literals are now always recognized in Message-ID: header lines.
1660 4. The first argument for a ${extract expansion item is the key name or field
1661 number. Leading and trailing spaces in this item were not being ignored,
1662 causing some misleading effects.
1664 5. When deliver_drop_privilege was set, single queue runner processes started
1665 manually (i.e. by the command "exim -q") or by the daemon (which uses the
1666 same command in the process it spins off) were not dropping privilege.
1668 6. When the daemon running as "exim" started a queue runner, it always
1669 re-executed Exim in the spun-off process. This is a waste of effort when
1670 deliver_drop_privilege is set. The new process now just calls the
1671 queue-runner function directly.
1674Exim version 4.23
1677 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
1680 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
1681 ignored. Though the use of -oP was forcing the writing of a pid file, it
1682 was always written to the default place.
1684 3. If the message "no IP address found for host xxxx" is generated during
1685 incoming verification, it is now followed by identification of the incoming
1686 connection (so you can more easily find what provoked it).
1688 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
1690 5. Added some features to "harden" Exim a bit more against certain attacks:
1692 (a) There is now a build-time option called FIXED_NEVER_USERS that can
1693 be put in Local/Makefile. This is like the never_users runtime option,
1694 but it cannot be overridden. The default setting is "root".
1696 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
1697 prefix string with which any file named in a -C command line option
1698 must start.
1700 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
1701 is retained for -C and -D only if the caller of Exim is root. Without
1702 it, the exim user may also use -C and -D and retain privilege.
1704 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
1705 command line option is disabled.
1707 6. Macro names set by the -D option must start with an upper case letter, just
1708 like macro names defined in the configuration file.
1710 7. Added "dereference=" facility to LDAP.
1712 8. Two instances of the typo "uknown" in the source files are fixed.
1714 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
1715 the Configure-Makefile script screwed up while processing it.
171710. Incorporated PCRE 4.4.
171911. The SMTP synchronization check was not operating right at the start of an
1720 SMTP session. For example, it could not catch a HELO sent before the client
1721 waited for the greeting. There is now a check for outstanding input at the
1722 point when the greeting is written. Because of the duplex, asynchronous
1723 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
1724 way, but not yet received, when the check is performed.
172612. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
1727 on TCP/IP sockets, because this apparently causes some broken clients to
1728 timeout.
173013. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
1731 unchanged) from the Cygwin maintainer.
173314. The code for -bV that shows what is in the binary showed "mbx" when maildir
1734 was supported instead of testing for mbx. Effectively a typo.
173615. The spa authenticator server code was not checking that the input it
1737 received was valid base64.
173916. The debug output line for the "set" modifier in ACLs was not showing the
1740 name of the variable that was being set.
174217. Code tidy: the variable type "vtype_string" was never used. Removed it.
174418. Previously, a reference to $sender_host_name did not cause a DNS reverse
1745 lookup on its own. Something else was needed to trigger the lookup. For
1746 example, a match in host_lookup or the need for a host name in a host list.
1747 Now, if $sender_host_name is referenced and the host name has not yet been
1748 looked up, a lookup is performed. If the lookup fails, the variable remains
1749 empty, and $host_lookup_failed is set to "1".
175119. Added "eqi" as a case-independent comparison operator.
175320. The saslauthd authentication condition could segfault if neither service
1754 nor realm was specified.
175621. If an overflowing value such as "2048M" was set for message_size_limit, the
1757 error message that was logged was misleading, and incoming SMTP
1758 connections were dropped. The message is now more accurate, and temporary
1759 errors are given to SMTP connections.
176122. In some error situations (such as 21 above) Exim rejects all SMTP commands
1762 (except RSET) with a 421 error, until QUIT is received. However, it was
1763 failing to send a response to QUIT.
176523. The HELO ACL was being run before the code for helo_try_verify_hosts,
1766 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
1767 ACL is now run after the helo_try_verify_hosts code.
176924. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
1770 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
1771 case-independent, so other case variants are also recognized). Apparently
1772 some systems use these upper case variants.
177425. If more than two messages were waiting for the same host, and a transport
1775 filter was specified for the transport, Exim sent two messages over the
1776 same TCP/IP connection, and then failed with "socket operation on non-
1777 socket" when it tried to send the third.
177926. Added Exim::debug_write and Exim::log_write for embedded Perl use.
178127. The extern definition of crypt16() in expand.c was not being excluded when
1782 the OS had its own crypt16() function.
178428. Added bounce_return_body as a new option, and bounce_return_size_limit
1785 as a preferred synonym for return_size_limit, both as an option and as an
1786 expansion variable.
178829. Added LIBS=-liconv to OS/Makefile-OSF1.
179030. Changed the default configuration ACL to relax the local part checking rule
1791 for addresses that are not in any local domains. For these addresses,
1792 slashes and pipe symbols are allowed within local parts, but the sequence
1793 /../ is explicitly forbidden.
179531. SPA server authentication was not clearing the challenge buffer before
1796 using it.
179832. log_message in a "warn" ACL statement was writing to the reject log as
1799 well as to the main log, which contradicts the documentation and doesn't
1800 seem right (because no rejection is happening). So I have stopped it.
180233. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
1803 However, I am unable to do any testing of this.
180534. Fixed an infelicity in the appendfile transport. When checking directories
1806 for a mailbox, to see if any needed to be created, it was accidentally
1807 using path names with one or more superfluous leading slashes; tracing
1808 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
181035. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
1811 contents are added to the log line that is written for every discarded
1812 recipient. (Previously a log_message setting was ignored.)
181436. The ${quote: operator now quotes the string if it is empty.
181637. The install script runs exim in order to find its version number. If for
1817 some reason other than non-existence or emptiness, which it checks, it
1818 could not run './exim', it was installing it with an empty version number,
1819 i.e. as "exim-". This error state is now caught, and the installation is
1820 aborted.
182238. An argument was missing from the function that creates an error message
1823 when Exim fails to connect to the socket for saslauthd authentication.
1824 This could cause Exim to crash, or give a corrupted message.
182639. Added isip, isip4, and isip6 to ${if conditions.
182840. The ACL variables $acl_xx are now saved with the message, and can be
1829 accessed later in routers, transports, and filters.
183141. The new lookup type nwildlsearch is like wildlsearch, except that the key
1832 strings in the file are not string-expanded.
183442. If a MAIL command specified a SIZE value that was too large to fit into an
1835 int variable, the check against message_size_limit failed. Such values are
1836 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
1837 day this will have to be increased, but I don't think I want to be around
1838 when emails are that large.
1842Exim version 4.22
1845 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
1846 iconv() is not standard in FreeBSD.
1848 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
1849 IPv6 enabled. The observed symptom was a segmentation fault on return from
1850 the function os_common_find_running_interfaces() in src/os.c.
1852 3. In the check_special_case() function in daemon.c I had used "errno" as an
1853 argument name, which causes warnings on some systems. This was basically a
1854 typo, since it was named "eno" in the comments!
1856 4. The code that waits for the clock to tick (at a resolution of some fraction
1857 of a second) so as to ensure message-id uniqueness was always waiting for
1858 at least one whole tick, when it could have waited for less. [This is
1859 almost certainly not relevant at current processor speeds, where it is
1860 unlikely to ever wait at all. But we try to future-proof.]
1862 5. The function that sleeps for a time interval that includes fractions of a
1863 second contained a race. It did not block SIGALRM between setting the
1864 timer, and suspending (a couple of lines later). If the interval was short
1865 and the sigsuspend() was delayed until after it had expired, the suspension
1866 never ended. On busy systems this could lead to processes getting stuck for
1867 ever.
1869 6. Some uncommon configurations may cause a lookup to happen in a queue runner
1870 process, before it forks any delivery processes. The open lookup caching
1871 mechanism meant that the open file or database connection was passed into
1872 the delivery process. The problem was that delivery processes always tidy
1873 up cached lookup data. This could cause a problem for the next delivery
1874 process started by the queue runner, because the external queue runner
1875 process does not know about the closure. So the next delivery process
1876 still has data in the lookup cache. In the case of a file lookup, there was
1877 no problem because closing a file descriptor in a subprocess doesn't affect
1878 the parent. However, if the lookup was caching a connection to a database,
1879 the connection was closed, and the second delivery process was likely to
1880 see errors such as "PGSQL: query failed: server closed the connection
1881 unexpectedly". The problem has been fixed by closing all cached lookups
1882 in a queue runner before running a delivery process.
1884 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
1885 seem to have the "const" qualifier which it has on other OS. I've
1886 parameterised it.
1888 8. Change 4.21/2 was too strict. It is only if there are two authenticators
1889 *of the same type* (client or server) with the same public name that an
1890 error should be diagnosed.
1892 9. When Exim looked up a host name for an IP address, but failed to find the
1893 original IP address when looking up the host name (a safety check), it
1894 output the message "<ip address> does not match any IP for NULL", which was
1895 confusing, to say the least. The bug was that the host name should have
1896 appeared instead of "NULL".
189810. Since release 3.03, if Exim is called by a uid other than root or the Exim
1899 user that is built into the binary, and the -C or -D options is used, root
1900 privilege is dropped before the configuration file is read. In addition,
1901 logging is switched to stderr instead of the normal log files. If the
1902 configuration then re-defines the Exim user, the unprivileged environment
1903 is probably not what is expected, so Exim logs a panic warning message (but
1904 proceeds).
1906 However, if deliver_drop_privilege is set, the unprivileged state may well
1907 be exactly what is intended, so the warning has been cut out in that case,
1908 and Exim is allowed to try to write to its normal log files.
1911Exim version 4.21
1914 1. smtp_return_error_details was not giving details for temporary sender
1915 or receiver verification errors.
1917 2. Diagnose a configuration error if two authenticators have the same public
1918 name.
1920 3. Exim used not to create the message log file for a message until the first
1921 delivery attempt. This could be confusing when incoming messages were held
1922 for policy or load reasons. The message log file is now created at the time
1923 the message is received, and an initial "Received" line is written to it.
1925 4. The automatically generated man page for command line options had a minor
1926 bug that caused no ill effects; however, a more serious problem was that
1927 the procedure for building the man page automatically didn't always
1928 operate. Consequently, release 4.20 contains an out-of-date version. This
1929 shouldn't happen again.
1931 5. When building Exim with embedded Perl support, the script that builds the
1932 Makefile was calling 'perl' to find its compile-time parameters, ignoring
1933 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
1935 6. The freeze_tell option was not being used for messages that were frozen on
1936 arrival, either by an ACL or by local_scan().
1938 7. Added the smtp_incomplete_transaction log selector.
1940 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
1941 was accepting AUTH without a new EHLO.
1943 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
1944 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
1945 RFC.
194710. Logging of TCP/IP connections (when configured) now happens in the main
1948 daemon process instead of the child process, so that the TCP/IP connection
1949 count is more accurate (but it can never be perfect).
195111. The use of "drop" in a nested ACL was not being handled correctly in the
1952 outer ACL. Now, if condition failure induced by the nested "drop" causes
1953 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
1954 or "require"), the connection is dropped.
195612. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
1957 that yield "discard" can now be used with an "accept" or a "discard" verb,
1958 but an error is generated for any others (because I can't see a useful way
1959 to define what should happen).
196113. When an ACL is read dynamically from a file (or anywhere else), the lines
1962 are now processed in the same way as lines in the Exim configuration file.
1963 In particular, continuation lines are supported.
196514. Added the "dnslists = a.b.c!=n.n.n.n" feature.
196715. Added -ti meaning -t -i.
196916. Check for letters, digits, hyphens, and dots in the names of dnslist
1970 domains, and warn by logging if others are found.
197217. At least on BSD, alignment is not guarenteed for the array of ifreq's
1973 returned from GIFCONF when Exim is trying to find the list of interfaces on
1974 a host. The code in os.c has been modified to copy each ifreq to an aligned
1975 structure in all cases.
1977 Also, in some cases, the returned ifreq's were being copied to a 'struct
1978 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
1979 means the last couple of bytes of an IPv6 address could be chopped if the
1980 ifreq contained only a normal sockaddr (14 bytes storage).
198218. Named domain lists were not supported in the hosts_treat_as_local option.
1983 An entry such as +xxxx was not recognized, and was treated as a literal
1984 domain name.
198619. Ensure that header lines added by a DATA ACL are included in the reject log
1987 if the ACL subsequently rejects the message.
198920. Upgrade the utility script to use Digest::MD5 instead of just
1990 MD5 (which is deprecated).
199221. When testing a filter file using -bf, Exim was writing a message when it
1993 took the sender from a "From " line in the message, but it was not doing so
1994 when it took $return_path from a Return-Path: header line. It now does.
199622. If the contents of a "message" modifier for a "warn" ACL verb do not begin
1997 with a valid header line field name (a series of printing characters
1998 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
200023. Changed "disc" in the source to "disk" to conform to the documentation and
2001 the book and for uniformity.
200324. Ignore Sendmail's -Ooption=value command line item.
200525. When execve() failed while trying to run a command in a pipe transport,
2006 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
2007 could be confused with a return value of 69 from the command itself. This
2008 has been changed to 127, the value the shell returns if it is asked to run
2009 a non-existent command. The wording for the related log line suggests a
2010 non-existent command as the problem.
201226. If received_header_text expands to an empty string, do not add a Received:
2013 header line to the message. (Well, it adds a token one on the spool, but
2014 marks it "old" so that it doesn't get used or transmitted.)
201627. Installed eximstats 1.28 (addition of -nt option).
201828. There was no check for failure on the call to getsockname() in the daemon
2019 code. This can fail if there is a shortage of resources on the system, with
2020 ENOMEM, for example. A temporary error is now given on failure.
202229. Contrary to the C standard, it seems that in some environments, the
2023 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
2024 program. Exim now does this explicitly; it affects the formatting of
2025 timestamps using strftime().
202730. If exiqsumm was given junk data, it threw up some uninitialized variable
2028 complaints. I've now initialized all the variables, to avoid this.
203032. Header lines added by a system filter were not being "seen" during
2031 transport-time rewrites.
203333. The info_callback() function passed to OpenSSL is set up with type void
2034 (*)(SSL *, int, int), as described somewhere. However, when calling the
2035 function (actually a macro) that sets it up, the type void(*)() is
2036 expected. I've put in a cast to prevent warnings from picky compilers.
203834. If a DNS black list lookup found a CNAME record, but there were no A
2039 records associated with the domain it pointed at, Exim crashed.
204135. If a DNS black list lookup returned more than one A record, Exim ignored
2042 all but the first. It now scans all returned addresses if a particular IP
2043 value is being sought. In this situation, the contents of the
2044 $dnslist_value variable are a list of all the addresses, separated by a
2045 comma and a space.
204736. Tightened up the rules for host name lookups using reverse DNS. Exim used
2048 to accept a host name and all its aliases if the forward lookup for any of
2049 them yielded the IP address of the incoming connection. Now it accepts only
2050 those names whose forward lookup yields the correct IP address. Any other
2051 names are discarded. This closes a loophole whereby a rogue DNS
2052 administrator could create reverse DNS records to break through a
2053 wildcarded host restriction in an ACL.
205537. If a user filter or a system filter that ran in a subprocess used any of
2056 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
2057 the wrong values were passed to the pipe command ($thisaddress had the
2058 value of $0, $0 had the value of $1, etc). This bug was introduced by
2059 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
206138. Improved the line breaking for long SMTP error messages from ACLs.
2062 Previously, if there was no break point between 40 and 75 characters, Exim
2063 left the rest of the message alone. Two changes have been made: (a) I've
2064 reduced the minimum length to 35 characters; (b) if it can't find a break
2065 point between 35 and 75 characters, it looks ahead and uses the first one
2066 that it finds. This may give the occasional overlong line, but at least the
2067 remaining text gets split now.
206939. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
2070 file descriptors might be low, and that setting 1000 would always raise it.
2071 It turns out that in some environments, the limit is already over 1000 and
2072 that lowering it causes trouble. So now Exim takes care not to decrease it.
207440. When delivering a message, the value of $return_path is set to $sender_
2075 address at the start of routing (routers may change the value). By an
2076 oversight, this default was not being set up when an address was tested by
2077 -bt or -bv, which affected the outcome if any router or filter referred to
2078 $return_path.
208041. The idea of the "warn" ACL verb is that it adds a header or writes to the
2081 log only when "message" or "log_message" are set. However, if one of the
2082 conditions was an address verification, or a call to a nested ACL, the
2083 messages generated by the underlying test were being passed through. This
2084 no longer happens. The underlying message is available in $acl_verify_
2085 message for both "message" and "log_message" expansions, so it can be
2086 passed through if needed.
208842. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
2089 new expansion $bh_ to give the encoded byte string without charset
2090 translation. Translation happens only if iconv() is available; HAVE_ICONV
2091 indicates this at build time. HEADERS_CHARSET gives the charset to
2092 translate to; headers_charset can change it in the configuration, and
2093 "headers charset" can change it in an individual filter file.
209543. Now that we have a default RFC 2047 charset (see above), the code in Exim
2096 that creates RFC 2047 encoded "words" labels them as that charset instead
2097 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
2098 expansion operator; (ii) when Exim creates a From: line for a local
2099 message; (iii) when a header line is rewritten to include a "phrase" part.
210144. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
2102 buggy, causing it to skip the first lines of messages whose message ID
2103 ended in 'D'. This would not have bitten before Exim release 4.14, because
2104 message IDs were unlikely to end in 'D' before then. The effect was to have
2105 incorrect size information for certain domains.
210745. #include "config.h" was missing at the start of the crypt16.c module. This
2108 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
2109 noticed.
211146. If there was a timeout during a "random" callout check, Exim treated it as
2112 a failure of the random address, and carried on sending RSET and the real
2113 address. If the delay was just some slowness somewhere, the response to the
2114 original RCPT would be taken as a response to RSET and so on, causing
2115 mayhem of various kinds.
211747. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
2118 when I implemented it. It didn't allow for the fact that some option values
2119 may legitimatetly be negative (e.g. size_addition), and it didn't even do
2120 the right test for positive values.
212248. Domain names in DNS records are case-independent. Exim always looks them up
2123 in lower case. Some resolvers return domain names in exactly the case they
2124 appear in the zone file, that is, they may contain uppercase letters. Not
2125 all resolvers do this - some return always lower case. Exim was treating a
2126 change of case by a resolver as a change of domain, similar to a widening
2127 of a domain abbreviation. This triggered its re-routing code and so it was
2128 trying to route what was effectively the same domain again. This normally
2129 caused routing to fail (because the router wouldn't handle the domain
2130 twice). Now Exim checks for this case specially, and just changes the
2131 casing of the domain that it ultimately uses when it transmits the message
2132 envelope.
213449. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
2135 module.
213750. If a filter generated a file delivery with a non-absolute name (possible if
2138 no home directory exists for the router), the forbid_file option was not
2139 forbidding it.
214151. Added '&' feature to dnslists, to provide bit mask matching in addition to
2142 the existing equality matching.
214452. Exim was using ints instead of ino_t variables in some places where it was
2145 dealing with inode numbers.
214753. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
2148 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
2149 called TMPDIR, and if it finds it is different, it changes its value.
215154. The smtp_printf() function is now made available to local_scan() so
2152 additional output lines can be written before returning. There is also an
2153 smtp_fflush() function to enable the detection of a dropped connection.
2154 The variables smtp_input and smtp_batched_input are exported to
2155 local_scan().
215755. Changed the default runtime configuration: the message "Unknown user"
2158 has been removed from the ACL, and instead placed on the localuser router,
2159 using the cannot_route_message feature. This means that any verification
2160 failures that generate their own messages won't get overridden. Similarly,
2161 the "Unrouteable address" message that was in the ACL for unverifiable
2162 relay addresses has also been removed.
216456. Added hosts_avoid_esmtp to the smtp transport.
216657. The exicyclog script was not checking for the esoteric option
2167 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2168 will work only if exicyclog is run under the appropriate euid.
217058. Following a discussion on the list, the rules by which Exim recognises line
2171 endings on incoming messages have been changed. The -dropcr and drop_cr
2172 options are now no-ops, retained only for backwards compatibility. The
2173 following line terminators are recognized: LF CRLF CR. However, special
2174 processing applies to CR:
2176 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2177 nor a local message in the state where . is a terminator.
2179 (ii) If a bare CR is encountered in a header line, an extra space is added
2180 after the line terminator so as not to end the header. The reasoning
2181 behind this is that bare CRs in header lines are most likely either
2182 to be mistakes, or people trying to play silly games.
218459. The size of a message, as listed by "-bp" or in the Exim monitor window,
2185 was being incorrectly given as 18 bytes larger than it should have been.
2186 This is a VOB (very old bug).
218860. This may never have affected anything current, but just in case it has:
2189 When the local host is found other than at the start of a list of hosts,
2190 the local host, those with the same MX, and any that follow, are discarded.
2191 When the list in question was part of a longer list of hosts, the following
2192 hosts (not currently being processed) were also being discarded. This no
2193 longer happens. I'm not sure if this situation could ever has previously
2194 arisen.
219661. Added the "/MX" feature to lists of hosts in the manualroute and query
2197 program routers.
219962. Whenever Exim generates a new message, it now adds an Auto-Submitted:
2200 header. This is something that is recommended in a new Internet Draft, and
2201 is something that is documented as being done by Sendmail. There are two
2202 possible values. For messages generated by the autoreply transport, Exim
2203 adds:
2205 Auto-Submitted: auto-replied
2207 whereas for all other generated messages (e.g. bounces) it adds
2209 Auto-Submitted: auto-generated
221163. The "personal" condition in filters now includes a test for the
2212 Auto-Submitted: header. If it contains the string "auto-" the message it
2213 not considered personal.
221564. Added rcpt_include_affixes as a generic transport option.
221765. Added queue_only_override (default true).
221966. Added the syslog_duplication option.
222167. If what should have been the first header line of a message consisted of
2222 a space followed by a colon, Exim was mis-interpreting it as a header line.
2223 It isn't of course - it is syntactically invalid and should therefore be
2224 treated as the start of the message body. The misbehaviour could have
2225 caused a number of strange effects, including loss of data in subsequent
2226 header lines, and spool format errors.
222868. Formerly, the AUTH parameter on a MAIL command was trusted only if the
2229 client host had authenticated. This control can now be exercised by an ACL
2230 for more flexibility.
223269. By default, callouts do not happen when testing with -bh. There is now a
2233 variant, -bhc, which does actually run the callout code, including
2234 consulting and updating the callout cache.
223670. Added support for saslauthd authentication, courtesy of Alexander
2237 Sabourenkov.
223971. If statvfs() failed on the spool or log directories while checking their
2240 size for availability, Exim confusingly gave the error "space shortage".
2241 Furthermore, in debugging mode it crashed with a floating point exception.
2242 These checks are done if check_{spool,log}_{space,inodes} are set, and when
2243 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
2244 serious problem, Exim now writes to the main and panic logs when this
2245 happens, with details of the failure. It then refuses to accept the
2246 incoming message, giving the message "spool directory problem" or "log
2247 directory problem" with a 421 code for SMTP messages.
224972. When Exim is about to re-exec itself, it ensures that the file descriptors
2250 0, 1, and 2 exist, because some OS complain for execs without them (see
2251 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
2252 descriptors. However, the code omitted to check that the open succeeded,
2253 causing mysterious errors if for some reason the permissions on /dev/null
2254 got screwed. Now Exim writes a message to the main and panic logs, and
2255 bombs out if it can't open /dev/null.
225773. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
2258 interact so that it is all more flexible. It is supposed to remain
2259 backwards compatible. Also added extra_local_interfaces.
226174. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
2262 to bomb out with an assertion failure - to the client this appears as a
2263 connection drop. This problem occurs in the part of the code that was taken
2264 from the Samba project. Fortunately, the assertion is in a very simple
2265 function, so I have fixed this by reproducing the function inline in the
2266 one place where it is called, and arranging for authentication to fail
2267 instead of killing the process with assert().
226975. The SPA client code was not working when the server requested OEM rather
2270 than Unicode encoding.
227276. Added code to make require_files with a specific uid setting more usable in
2273 the case where statting the file as root fails - usually a non-root-mounted
2274 NFS file system. When this happens and the failure is EACCES, Exim now
2275 forks a subprocess and does the per-uid checking as the relevant uid.
227777. Added process_log_path.
227978. If log_file_path was not explicitly set, a setting of check_log_space or
2280 check_log_inodes was ignored.
228279. If a space check for the spool or log partitions fails, the incident is now
2283 logged. Of course, in the latter case the data may get lost...
228580. Added the %p formatting code to string_format() so that it can be used to
2286 print addresses in debug_print(). Adjusted all the address printing in the
2287 debugging in store.c to use %p rather than %d.
228981. There was a concern that a line of code in smtp_in.c could overflow a
2290 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
2291 initially expressed, the concern was not well-founded, because trailing
2292 spaces are removed early. However, if the trailing spaces were followed by
2293 a NULL, they did not get removed, so the overflow was possible. Two fixes
2294 were applied:
2296 (a) I re-wrote the offending code in a cleaner fashion.
2297 (b) If an incoming SMTP command contains a NULL character, it is rejected
2298 as invalid.
230082. When Exim changes uid/gid to the Exim user at daemon start time, it now
2301 runs initgroups(), so that if the Exim user is in any additional groups,
2302 they will be used during message reception.
2305Exim version 4.20
2308The change log for 4.20 and earlier releases has been archived.