1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.168 2005/06/22 14:45:05 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
10 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
12 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
13 command itself ended in error, the underlying message about the transport
14 filter timeout was being overwritten with the pipe command error. Now the
15 underlying error message should be appended to the second error message.
17 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
19 PH/02 Reduce the amount of output that "make" produces by default. Full output
20 can still be requested.
22 PH/03 The warning log line about a condition test deferring for a "warn" verb
23 was being output only once per connection, rather than after each
24 occurrence (because it was using the same function as for successful
25 "warn" verbs). This seems wrong, so I have changed it.
27 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
28 it should not have, which might have caused a crash in the right
29 circumstances, but probably never did.
31 PH/04 Installed a modified version of Tony Finch's patch to make submission
32 mode fix the return path as well as the Sender: header line, and to
33 add a /name= option so that you can make the user's friendly name appear
36 TF/03 Added the control = fakedefer ACL modifier.
38 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
39 Mark Lowes for thorough testing.
41 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
43 TK/03 Merged latest SRS patch from Miles Wilton.
45 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
46 with the definition in sysexits.h (which is #included earlier).
47 Fortunately, Exim does not actually use EX_OK. The code used to try to
48 preserve the sysexits.h value, by assumimg that macro definitions were
49 scanned for macro replacements. I have been disabused of this notion,
50 so now the code just undefines EX_OK before #including unistd.h.
52 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
53 in the smtp transport. When a block could not be written in a single
54 write() function, the timeout was being re-applied to each part-write.
55 This seems wrong - if the receiver was accepting one byte at a time it
56 would take for ever. The timeout is now adjusted when this happens. It
57 doesn't have to be particularly precise.
59 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
60 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
62 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
63 without a domain if the check to <postmaster@domain> fails.
65 SC/01 Eximstats: added -xls and the ability to specify output files
66 (patch written by Frank Heydlauf).
68 SC/02 Eximstats: use FileHandles for outputing results.
70 SC/03 Eximstats: allow any combination of xls, txt, and html output.
72 SC/04 Eximstats: fixed display of large numbers with -nvr option
74 SC/05 Eximstats: fixed merging of reports with empty tables.
76 SC/06 Eximstats: added the -include_original_destination flag
78 SC/07 Eximstats: removed tabs and trailing whitespace.
80 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
82 TK/06 MBOX spool code: Add real "From " MBOX separator line
83 so the .eml file is really in mbox format (even though
84 most programs do not really care). Patch from Alex Miller.
86 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
87 The latter is generated from $received_to and is only set if the
88 message has one envelope recipient. SA can use these headers,
89 obviously out-of-the-box. Patch from Alex Miller.
91 PH/08 The ${def test on a variable was returning false if the variable's
92 value was "0", contrary to what the specification has always said!
93 The result should be true unless the variable is empty.
95 PH/09 The syntax error of a character other than { following "${if
96 def:variable_name" (after optional whitespace) was not being diagnosed.
97 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
98 accidental colon was present, for example, could give incorrect results.
100 PH/10 Tidied the code in a number of places where the st_size field of a stat()
101 result is used (not including appendfile, where other changes are about
104 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
105 This involved changing a lot of size variables from int to off_t. It
106 should work with maildirs and everything.
108 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
109 spamd dying while we are connected to it.
111 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
114 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
115 function for BDB 4.3.
117 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
119 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
120 the sources for PCRE have been re-arranged and more files are now
123 PH/15 The code I had for printing potentially long long variables in PH/11
124 above was not the best (it lost precision). The length of off_t variables
125 is now inspected at build time, and an appropriate printing format (%ld
126 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
127 to be "long long int" or "long int". This is needed for the internal
128 formatting function string_vformat().
130 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
131 the configuration file to be ":syslog", then the script "guesses" where
132 the logs files are, rather than using the compiled in default. In our
133 case the guess is not the same as the compiled default, so the script
134 suddenly stopped working when I started to use syslog. The patch checks
135 to see if log_file_path is "". If so, it attempts to read it from exim
136 with no configuration file to get the compiled in version, before it
137 falls back to the previous guessing code."
139 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
140 implementing BATV in an Exim configuration. See NewStuff for the gory
143 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
144 Makefile that are specific to HP-UX.
146 PH/18 If the "use_postmaster" option was set for a recipient callout together
147 with the "random" option, the postmaster address was used as the MAIL
148 FROM address for the random test, but not for the subsequent recipient
149 test. It is now used for both.
151 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
152 patch removes a few documentation additions to RFC 3028, because the
153 latest draft now contains them. It adds the new en;ascii-case comparator
154 and a new error check for 8bit text in MIME parts. Comparator and
155 require names are now matched exactly. I enabled the subaddress
156 extension, but it is not well tested yet (read: it works for me)."
158 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
159 rework some of the code of TK/09 above to avoid the hardwired use of
160 "%lld" and "long long". Replaced the call to snprintf() with a call to
163 PH/21 Added another message to those in 4.51/PH/42, namely "All relevant MX
164 records point to non-existent hosts".
166 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
167 experimental DomainKeys support:
169 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
170 (2) On an error such as an illegally used "control", the wrong name for
171 the control was given.
173 These problems did NOT occur unless DomainKeys support was compiled.
175 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
177 PH/24 Added ${if match_ip condition.
179 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
180 databases so that it will be absolutely obvious if a crash occurs in the
181 DB library. This is a regular occurrence (often caused by mis-matched
188 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
189 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
191 TK/02 Fix ACL "control" statement not being available in MIME ACL.
193 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
195 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
196 to test Sieve filters that use "vacation".
198 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
199 that changes the way the GnuTLS parameters are stored in the cache file.
200 The new format can be generated externally. For backward compatibility,
201 if the data in the cache doesn't make sense, Exim assumes it has read an
202 old-format file, and it generates new data and writes a new file. This
203 means that you can't go back to an older release without removing the
206 PH/03 A redirect router that has both "unseen" and "one_time" set does not
207 work if there are any delivery delays because "one_time" forces the
208 parent to be marked "delivered", so its unseen clone is never tried
209 again. For this reason, Exim now forbids the simultaneous setting of
212 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
213 redirected to themselves ("homonym" addresses). Read the long ChangeLog
214 entry if you want to know the details. The fix, however, neglected to
215 consider the case when local delivery batching is involved. The test for
216 "previously delivered" was not happening when checking to see if an
217 address could be batched with a previous (undelivered) one; under
218 certain circumstances this could lead to multiple deliveries to the same
221 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
222 in its include files, and this causes problems building Exim.
224 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
225 header_syntax) but Exim was just ignoring anything given after a slash.
226 In particular, this caused confusion with an attempt to use "verify =
227 reverse_host_lookup/defer_ok". An error is now given when options are
228 supplied for verify items that do not have them. (Maybe reverse_host_
229 lookup should have a defer_ok option, but that's a different point.)
231 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
232 defined by RFC 821) to 2048, because there were problems with some AUTH
233 commands, and RFC 1869 says the size should be increased for extended
234 SMTP commands that take arguments.
236 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
239 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
240 "unknown" error; now it says that the functionality isn't in the binary.
242 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
243 an address' error message when a string expansion fails (syntax or
244 whatever). Otherwise the password may appear in the log. Following change
245 PH/42 below, there is no longer a chance of it appearing in a bounce
248 PH/11 Installed exipick version 20050225.0 from John Jetmore.
250 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
251 of its addresses was ever tried. (Bugzilla bug #2.)
253 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
254 the result incorrectly in the debug output. (It correctly added a newline
255 to what was transported.)
257 TF/01 Added $received_time.
259 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
260 commented out examples of how to interface to a virus scanner and to
261 SpamAssassin. Also added commented examples of av_scanner and
262 spamd_address settings.
264 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
265 and controls are allowed in which ACLs. There were a couple of minor
266 errors. Some of the entries in the conditions table (which is a table of
267 where they are NOT allowed) were getting very unwieldy; rewrote them as a
268 negation of where the condition IS allowed.
270 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
272 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
273 header file does not have a version number, so I've had to invent a new
274 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
275 API. The code is untested by me (my Linux distribution still has 0.3.2 of
276 radiusclient), but it was contributed by a Radius user.
278 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
279 files or directories, for OpenSSL.
281 PH/19 When an Exim process that is running as root has to create an Exim log
282 file, it does so in a subprocess that runs as exim:exim so as to get the
283 ownership right at creation (otherwise, other Exim processes might see
284 the file with the wrong ownership). There was no test for failure of this
285 fork() call, which would lead to the process getting stuck as it waited
286 for a non-existent subprocess. Forks do occasionally fail when resources
287 run out. I reviewed all the other calls to fork(); they all seem to check
290 PH/20 When checking for unexpected SMTP input at connect time (before writing
291 the banner), Exim was not dealing correctly with a non-positive return
292 from the read() function. If the client had disconnected by this time,
293 the result was a log entry for a synchronization error with an empty
294 string after "input=" when read() returned zero. If read() returned -1
295 (an event I could not check), uninitialized data bytes were printed.
296 There were reports of junk text (parts of files, etc) appearing after
299 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
301 PH/22 Added support for macro redefinition, and (re)definition in between
302 driver and ACL definitions.
304 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
305 forgetting to use the resulting value; it was using the unexpanded value.
307 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
308 hadn't been configured. The fix is from Juergen Kreileder, who
309 understands it better than I do:
311 "Here's what I see happening with three configured cyrus_sasl
312 authenticators configured (plain, login, cram-md5):
314 On startup auth_cyrus_sasl_init() gets called for each of these.
315 This means three calls to sasl_listmech() without a specified mech_list.
316 => SASL tests which mechs of all available mechs actually work
317 => three warnings about OTP not working
318 => the returned list contains: plain, login, cram-md5, digest-md5, ...
320 With the patch, sasl_listmech() also gets called three times. But now
321 SASL's mech_list option is set to the server_mech specified in the the
322 authenticator. Or in other words, the answer from sasl_listmech()
323 gets limited to just the mech you're testing for (which is different
325 => the return list contains just 'plain' or 'login', 'cram-md5' or
326 nothing depending on the value of ob->server_mech.
328 I've just tested the patch: Authentication still works fine,
329 unavailable mechs specified in the exim configuration are still
330 caught, and the auth.log warnings about OTP are gone."
332 PH/25 When debugging is enabled, the contents of the command line are added
333 to the debugging output, even when log_selector=+arguments is not
336 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
337 answer is "GNU", and only if the return is "GNU/something" is the answer
340 PH/27 $acl_verify_message is now set immediately after the failure of a
341 verification in an ACL, and so is available in subsequent modifiers. In
342 particular, the message can be preserved by coding like this:
344 warn !verify = sender
345 set acl_m0 = $acl_verify_message
347 Previously, $acl_verify_message was set only while expanding "message"
348 and "log_message" when a very denied access.
350 PH/28 Modified OS/os.c-Linux with
352 -#ifndef OS_LOAD_AVERAGE
353 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
355 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
356 nomenclature these days.)
358 PH/29 Installed patch from the Sieve maintainer that adds the options
359 sieve_useraddress and sieve_subaddress to the redirect router.
361 PH/30 In these circumstances:
362 . Two addresses routed to the same list of hosts;
363 . First host does not offer TLS;
364 . First host accepts first address;
365 . First host gives temporary error to second address;
366 . Second host offers TLS and a TLS session is established;
367 . Second host accepts second address.
368 Exim incorrectly logged both deliveries with the TLS parameters (cipher
369 and peerdn, if requested) that were in fact used only for the second
372 PH/31 When doing a callout as part of verifying an address, Exim was not paying
373 attention to any local part prefix or suffix that was matched by the
374 router that accepted the address. It now behaves in the same way as it
375 does for delivery: the affixes are removed from the local part unless
376 rcpt_include_affixes is set on the transport.
378 PH/32 Add the sender address, as F=<...>, to the log line when logging a
379 timeout during the DATA phase of an incoming message.
381 PH/33 Sieve envelope tests were broken for match types other than :is. I have
382 applied a patch sanctioned by the Sieve maintainer.
384 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
385 the uid or gid is negative. A case of a negative gid caused this to be
386 noticed. The fix allows for either to be negative.
388 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
389 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
392 PH/36 The change PH/12 above was broken. Fixed it.
394 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
395 the grounds that routing the same address twice would always produce the
396 same answer. This might have been true once, but it is certainly no
397 longer true now. Routing a child address may depend on the previous
398 routing that produced that child. Some complicated redirection strategies
399 went wrong when messages had multiple recipients, and made Exim's
400 behaviour dependent on the order in which the addresses were given.
402 I have moved the duplicate checking until after the routing is complete.
403 Exim scans the addresses that are assigned to local and remote
404 transports, and removes any duplicates. This means that more work will be
405 done, as duplicates will always all be routed, but duplicates are
406 presumably rare, so I don't expect this is of any significance.
408 For deliveries to pipes, files, and autoreplies, the duplicate checking
409 still happens during the routing process, since they are not going to be
412 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
413 It corrects a timeout issue with spamd. This is Ian's comment: "The
414 background is that sometimes spamd either never reads data from a
415 connection it has accepted, or it never writes response data. The exiscan
416 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
417 blindly assumes that writes won't block so it may never time out."
419 PH/39 Allow G after quota size as well as K and M.
421 PH/40 The value set for $authenticated_id in an authenticator may not contain
422 binary zeroes or newlines because the value is written to log lines and
423 to spool files. There was no check on this. Now the value is run through
424 the string_printing() function so that such characters are converted to
425 printable escape sequences.
427 PH/41 $message_linecount is a new variable that contains the total number of
428 lines in the message. Compare $body_linecount, which is the count for the
431 PH/42 Exim no longer gives details of delivery errors for specific addresses in
432 bounce and delay warning messages, except in certain special cases, which
435 (a) An SMTP error message from a remote host;
436 (b) A message specified in a :fail: redirection;
437 (c) A message specified in a "fail" command in a system filter;
438 (d) A message specified in a FAIL return from the queryprogram router;
439 (e) A message specified by the cannot_route_message router option.
441 In these cases only, Exim does include the error details in bounce and
442 warning messages. There are also a few cases where bland messages such
443 as "unrouteable address" or "local delivery error" are given.
445 PH/43 $value is now also set for the "else" part of a ${run expansion.
447 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
448 being worked on, but at least Exim now implements the latest version to
451 PH/45 In a pipe transport, although a timeout while waiting for the pipe
452 process to complete was treated as a delivery failure, a timeout while
453 writing the message to the pipe was logged, but erroneously treated as a
454 successful delivery. Such timeouts include transport filter timeouts. For
455 consistency with the overall process timeout, these timeouts are now
456 treated as errors, giving rise to delivery failures by default. However,
457 there is now a new Boolean option for the pipe transport called
458 timeout_defer, which, if set TRUE, converts the failures into defers for
459 both kinds of timeout. A transport filter timeout is now identified in
462 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
463 systems where "make" and "gmake" are different, calling "gmake" at top
464 level broke things. I've arranged for the value of $(MAKE) to be passed
465 from the Makefile to this script so that it can call the same version of
469 A note about Exim versions 4.44 and 4.50
470 ----------------------------------------
472 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
473 changes of various kinds. As a consequence, a big documentation update was
474 needed. This delayed the release for rather longer than seemed good, especially
475 in the light of a couple of (minor) security issues. Therefore, the changes
476 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
477 release. So 4.44 and 4.50 are in effect two different branches that both start
480 I have left the 4.50 change log unchanged; it contains all the changes since
481 4.43. The change log for 4.44 is below; many of its items are identical to
482 those for 4.50. This seems to be the most sensible way to preserve the
483 historical information.
489 1. Minor wording change to the doc/README.SIEVE file.
491 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
492 computation of the current number of files was incorrect.
494 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
495 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
498 4. Give more explanation in the error message when the command for a transport
499 filter fails to execute.
501 5. There are several places where Exim runs a non-Exim command in a
502 subprocess. The SIGUSR1 signal should be disabled for these processes. This
503 was being done only for the command run by the queryprogram router. It is
504 now done for all such subprocesses. The other cases are: ${run, transport
505 filters, and the commands run by the lmtp and pipe transports.
507 6. Added CONFIGURE_GROUP build-time option.
509 7. Some older OS have a limit of 256 on the maximum number of file
510 descriptors. Exim was using setrlimit() to set 1000 as a large value
511 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
512 systems. I've change it so that if it can't get 1000, it tries for 256.
514 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
515 was an oversight, and furthermore, ever since the addition of extra
516 controls (e.g. 4.43/32), the checks on when to allow different forms of
517 "control" were broken. There should now be diagnostics for all cases when a
518 control that does not make sense is encountered.
520 9. Added the /retain_sender option to "control=submission".
522 10. $recipients is now available in the predata ACL (oversight).
524 11. Tidy the search cache before the fork to do a delivery from a message
525 received from the command line. Otherwise the child will trigger a lookup
526 failure and thereby defer the delivery if it tries to use (for example) a
527 cached ldap connection that the parent has called unbind on.
529 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
530 of $address_data from the recipient verification was clobbered by the
533 13. The value of address_data from a sender verification is now available in
534 $sender_address_data in subsequent conditions in the ACL statement.
536 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
538 15. Added a new option "connect=<time>" to callout options, to set a different
541 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
542 was its contents. (It was OK if the option was not defined at all.)
544 17. A "Completed" log line is now written for messages that are removed from
545 the spool by the -Mrm option.
547 18. New variables $sender_verify_failure and $recipient_verify_failure contain
548 information about exactly what failed.
550 19. Added -dd to debug only the daemon process.
552 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
553 handles timeouts, both on the server side and network timeouts. Renamed the
554 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
556 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
557 It is now set to "smtps".
559 22. $host_address is now set to the target address during the checking of
562 23. When checking ignore_target_hosts for an ipliteral router, no host name was
563 being passed; this would have caused $sender_host_name to have been used if
564 matching the list had actually called for a host name (not very likely,
565 since this list is usually IP addresses). A host name is now passed as
568 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
569 code that specifies a non-restarting handler (typically sigaction() in
570 modern systems) in an attempt to fix a rare and obscure crash bug.
572 25. Narrowed the window for a race in the daemon that could cause it to ignore
573 SIGCHLD signals. This is not a major problem, because they are used only to
574 wake it up if nothing else does.
576 26. A malformed maildirsize file could cause Exim to calculate negative values
577 for the mailbox size or file count. Odd effects could occur as a result.
578 The maildirsize information is now recalculated if the size or filecount
581 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
582 support for a long time. Removed HAVE_SYS_VFS_H.
584 28. Installed the latest version of exipick from John Jetmore.
586 29. In an address list, if the pattern was not a regular expression, an empty
587 subject address (from a bounce message) matched only if the pattern was an
588 empty string. Non-empty patterns were not even tested. This was the wrong
589 because it is perfectly reasonable to use an empty address as part of a
590 database query. An empty address is now tested by patterns that are
591 lookups. However, all the other forms of pattern expect the subject to
592 contain a local part and a domain, and therefore, for them, an empty
593 address still always fails if the pattern is not itself empty.
595 30. Exim went into a mad DNS loop when attempting to do a callout where the
596 host was specified on an smtp transport, and looking it up yielded more
599 31. Re-factored the code for checking spool and log partition space into a
600 function that finds that data and another that does the check. The former
601 is then used to implement four new variables: $spool_space, $log_space,
602 $spool_inodes, and $log_inodes.
604 32. The RFC2047 encoding function was originally intended for short strings
605 such as real names; it was not keeping to the 75-character limit for
606 encoded words that the RFC imposes. It now respects the limit, and
607 generates multiple encoded words if necessary. To be on the safe side, I
608 have increased the buffer size for the ${rfc2047: expansion operator from
611 33. It is now permitted to omit both strings after an "if" condition; if the
612 condition is true, the result is "true". As before, when the second string
613 is omitted, a false condition yields an empty string. This makes it less
614 cumbersome to write custom ACL and router conditions.
616 34. Failure to deliver a bounce message always caused it to be frozen, even if
617 there was an errors_to setting on the router. The errors_to setting is now
620 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
621 canonical form (fully expanded) before being placed in
622 $sender_host_address.
624 36. The table in the code that translates DNS record types into text (T_A to
625 "A" for instance) was missing entries for NS and CNAME. It is just possible
626 that this could have caused confusion if both these types were looked up
627 for the same domain, because the text type is used as part of Exim's
628 per-process caching. But the chance of anyone hitting this buglet seems
631 37. The dnsdb lookup has been extended in a number of ways.
633 (1) There is a new type, "zns", which walks up the domain tree until it
634 finds some nameserver records. It should be used with care.
636 (2) There is a new type, "mxh", which is like "mx" except that it returns
637 just the host names, not the priorities.
639 (3) It is now possible to give a list of domains (or IP addresses) to be
640 looked up. The behaviour when one of the lookups defers can be
641 controlled by a keyword.
643 (4) It is now possible to specify the separator character for use when
644 multiple records are returned.
646 38. The dnslists ACL condition has been extended: it is now possible to supply
647 a list of IP addresses and/or domains to be looked up in a particular DNS
650 39. Added log_selector=+queue_time_overall.
652 40. When running the queue in the test harness, wait just a tad after forking a
653 delivery process, to get repeatability of debugging output.
655 41. Include certificate and key file names in error message when GnuTLS fails
656 to set them up, because the GnuTLS error message doesn't include the name
657 of the failing file when there is a problem reading it.
659 42. Allow both -bf and -bF in the same test run.
661 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
663 44. The "Exiscan patch" is now merged into the mainline Exim source.
665 45. Sometimes the final signoff response after QUIT could fail to get
666 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
667 < 0 before doing a fflush(). This bug looks as though it goes back to the
668 introduction of TLS in release 3.20, but "sometimes" must have been rare
669 because the tests only now provoked it.
671 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
672 (this can affect the format of dates).
674 47. exim_tidydb, when checking for the continued existence of a message for
675 which it has found a message-specific retry record, was not finding
676 messages that were in split spool directories. Consequently, it was
677 deleting retry records that should have stayed in existence.
679 48. Steve fixed some bugs in eximstats.
681 49. The SPA authentication driver was not abandoning authentication and moving
682 on to the next authenticator when an expansion was forced to fail,
683 contradicting the general specification for all authenticators. Instead it
684 was generating a temporary error. It now behaves as specified.
686 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
687 (the order specifies the preference for clients). The order is now AES256,
688 AES128, 3DES, ARCFOUR128.
690 51. Small patch to Sieve code - explicitly set From: when generating an
693 52. Exim crashed if a remote delivery caused a very long error message to be
694 recorded - for instance if somebody sent an entire SpamAssassin report back
695 as a large number of 550 error lines. This bug was coincidentally fixed by
696 increasing the size of one of Exim's internal buffers (big_buffer) that
697 happened as part of the Exiscan merge. However, to be on the safe side, I
698 have made the code more robust (and fixed the comments that describe what
701 53. Now that there can be additional text after "Completed" in log lines (if
702 the queue_time_overall log selector is set), a one-byte patch to exigrep
703 was needed to allow it to recognize "Completed" as not the last thing in
706 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
707 patch that reportedly fixes this has been added. I am not expert enough to
708 create a test for it. This is what the patch creator wrote:
710 "I found a little strange behaviour of ldap code when working with
711 Windows 2003 AD Domain, where users was placed in more than one
712 Organization Units. When I tried to give exim partial DN, the exit code
713 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
714 But simultaneously result of request was absolutely normal ldap result,
715 so I produce this patch..."
717 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
718 REFERENCE, so I have modified the code to exclude the patch when that macro
721 55. Some experimental protocols are using DNS PTR records for new purposes. The
722 keys for these records are domain names, not reversed IP addresses. The
723 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
724 leaves it alone. Component reversal etc. now happens only for IP addresses.
727 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
729 57. Double the size of the debug message buffer (to 2048) so that more of very
730 long debug lines gets shown.
732 58. The exicyclog utility now does better if the number of log files to keep
733 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
735 59. Two changes related to the smtp_active_hostname option:
737 (1) $smtp_active_hostname is now available as a variable.
738 (2) The default for smtp_banner uses $smtp_active_hostname instead
739 of $primary_hostname.
741 60. The host_aton() function is supposed to be passed a string that is known
742 to be a valid IP address. However, in the case of IPv6 addresses, it was
743 not checking this. This is a hostage to fortune. Exim now panics and dies
744 if the condition is not met. A case was found where this could be provoked
745 from a dnsdb PTR lookup with an IPv6 address that had more than 8
746 components; fortuitously, this particular loophole had already been fixed
747 by change 4.50/55 above.
749 If there are any other similar loopholes, the new check in host_aton()
750 itself should stop them being exploited. The report I received stated that
751 data on the command line could provoke the exploit when Exim was running as
752 exim, but did not say which command line option was involved. All I could
753 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
757 61. There was a buffer overflow vulnerability in the SPA authentication code
758 (which came originally from the Samba project). I have added a test to the
759 spa_base64_to_bits() function which I hope fixes it.
762 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
763 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
765 63. The daemon start-up calls getloadavg() while still root for those OS that
766 need the first call to be done as root, but it missed one case: when
767 deliver_queue_load_max is set with deliver_drop_privilege. This is
768 necessary for the benefit of the queue runner, because there is no re-exec
769 when deliver_drop_privilege is set.
771 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
774 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
775 was in use, was not putting the data itself into the right store pool;
776 consequently, it could be overwritten for a subsequent message in the same
777 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
780 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
782 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
783 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
784 the function were treating the return as a boolean value, which happened to
785 work because 0=false and not-0=true, but is not correct code.
787 68. The host_aton() function was not handling scoped IPv6 addresses (those
788 with, for example, "%eth0" on the end) correctly.
790 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
791 negated items (that is, ~something) in unsigned ints. Some compilers
792 apparently mutter when there is no cast.
794 70. If an address verification called from an ACL failed, and did not produce a
795 user-specific message (i.e. there was only a "system" message), nothing was
796 put in $acl_verify_message. In this situation, it now puts the system
799 71. Change 4.23/11 added synchronization checking at the start of an SMTP
800 session; change 4.31/43 added the unwanted input to the log line - except
801 that it did not do this in the start of session case. It now does.
803 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
804 This is wrong and can cause the other end to generate a synchronization
805 error if it is another Exim or anything else that does the synchronization
806 check. A QUIT command is no longer sent after a timeout.
808 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
811 74. The defer_ok option of callout verification was not working if it was used
812 when verifying addresses in header lines, that is, for this case:
814 verify = header_sender/callout=defer_ok
816 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
817 those file descriptors could be used for SMTP connections. If anything
818 wrote to stderr (the example that came up was "warn" in embedded Perl), it
819 could be sent to the SMTP client, causing chaos. The daemon now opens
820 stdin, stdout, and stderr to /dev/null when it puts itself into the
823 76. Arrange for output from Perl's "warn" command to be written to Exim's main
824 log by default. The user can override this with suitable Perl magic.
826 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
827 the log message when discard triggers, was not working for the DATA ACL or
828 for the non-SMTP ACL.
830 78. Error message wording change in sieve.c.
832 79. If smtp_accept_max_per_host was set, the number of connections could be
833 restricted to fewer than expected, because the daemon was trying to set up
834 a new connection before checking whether the processes handling previous
835 connections had finished. The check for completed processes is now done
836 earlier. On busy systems, this bug wouldn't be noticed because something
837 else would have woken the daemon, and it would have reaped the completed
840 80. If a message was submitted locally by a user whose login name contained one
841 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
842 It caused a spool format error. I have fixed the spool reading code. A
843 related problem was that the "from" clause in the Received: line became
844 illegal because of the space(s). It is now covered by ${quote_local_part.
846 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
849 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
851 83. Patch from Sieve maintainer to fix unterminated string problem in
854 84. Some minor changes to the Linux configuration files to help with other
855 OS variants using glibc.
857 85. One more patch for Sieve to update vacation handling to latest spec.
860 ----------------------------------------------------
861 See the note above about the 4.44 and 4.50 releases.
862 ----------------------------------------------------
868 1. Change 4.43/35 introduced a bug that caused file counts to be
869 incorrectly computed when quota_filecount was set in an appendfile
872 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
873 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
876 3. Give more explanation in the error message when the command for a transport
877 filter fails to execute.
879 4. There are several places where Exim runs a non-Exim command in a
880 subprocess. The SIGUSR1 signal should be disabled for these processes. This
881 was being done only for the command run by the queryprogram router. It is
882 now done for all such subprocesses. The other cases are: ${run, transport
883 filters, and the commands run by the lmtp and pipe transports.
885 5. Some older OS have a limit of 256 on the maximum number of file
886 descriptors. Exim was using setrlimit() to set 1000 as a large value
887 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
888 systems. I've change it so that if it can't get 1000, it tries for 256.
890 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
891 was an oversight, and furthermore, ever since the addition of extra
892 controls (e.g. 4.43/32), the checks on when to allow different forms of
893 "control" were broken. There should now be diagnostics for all cases when a
894 control that does not make sense is encountered.
896 7. $recipients is now available in the predata ACL (oversight).
898 8. Tidy the search cache before the fork to do a delivery from a message
899 received from the command line. Otherwise the child will trigger a lookup
900 failure and thereby defer the delivery if it tries to use (for example) a
901 cached ldap connection that the parent has called unbind on.
903 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
904 of $address_data from the recipient verification was clobbered by the
907 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
908 was its contents. (It was OK if the option was not defined at all.)
910 11. A "Completed" log line is now written for messages that are removed from
911 the spool by the -Mrm option.
913 12. $host_address is now set to the target address during the checking of
916 13. When checking ignore_target_hosts for an ipliteral router, no host name was
917 being passed; this would have caused $sender_host_name to have been used if
918 matching the list had actually called for a host name (not very likely,
919 since this list is usually IP addresses). A host name is now passed as
922 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
923 code that specifies a non-restarting handler (typically sigaction() in
924 modern systems) in an attempt to fix a rare and obscure crash bug.
926 15. Narrowed the window for a race in the daemon that could cause it to ignore
927 SIGCHLD signals. This is not a major problem, because they are used only to
928 wake it up if nothing else does.
930 16. A malformed maildirsize file could cause Exim to calculate negative values
931 for the mailbox size or file count. Odd effects could occur as a result.
932 The maildirsize information is now recalculated if the size or filecount
935 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
936 support for a long time. Removed HAVE_SYS_VFS_H.
938 18. Updated exipick to current release from John Jetmore.
940 19. Allow an empty sender to be matched against a lookup in an address list.
941 Previously the only cases considered were a regular expression, or an
944 20. Exim went into a mad DNS lookup loop when doing a callout where the
945 host was specified on the transport, if the DNS lookup yielded more than
948 21. The RFC2047 encoding function was originally intended for short strings
949 such as real names; it was not keeping to the 75-character limit for
950 encoded words that the RFC imposes. It now respects the limit, and
951 generates multiple encoded words if necessary. To be on the safe side, I
952 have increased the buffer size for the ${rfc2047: expansion operator from
955 22. Failure to deliver a bounce message always caused it to be frozen, even if
956 there was an errors_to setting on the router. The errors_to setting is now
959 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
960 canonical form (fully expanded) before being placed in
961 $sender_host_address.
963 24. Updated eximstats to version 1.33
965 25. Include certificate and key file names in error message when GnuTLS fails
966 to set them up, because the GnuTLS error message doesn't include the name
967 of the failing file when there is a problem reading it.
969 26. Expand error message when OpenSSL has problems setting up cert/key files.
972 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
973 (this can affect the format of dates).
975 28. exim_tidydb, when checking for the continued existence of a message for
976 which it has found a message-specific retry record, was not finding
977 messages that were in split spool directories. Consequently, it was
978 deleting retry records that should have stayed in existence.
980 29. eximstats updated to version 1.35
981 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
982 1.35 - bugfix such that pie charts by volume are generated correctly
984 30. The SPA authentication driver was not abandoning authentication and moving
985 on to the next authenticator when an expansion was forced to fail,
986 contradicting the general specification for all authenticators. Instead it
987 was generating a temporary error. It now behaves as specified.
989 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
990 (the order specifies the preference for clients). The order is now AES256,
991 AES128, 3DES, ARCFOUR128.
993 31. Small patch to Sieve code - explicitly set From: when generating an
996 32. Exim crashed if a remote delivery caused a very long error message to be
997 recorded - for instance if somebody sent an entire SpamAssassin report back
998 as a large number of 550 error lines. This bug was coincidentally fixed by
999 increasing the size of one of Exim's internal buffers (big_buffer) that
1000 happened as part of the Exiscan merge. However, to be on the safe side, I
1001 have made the code more robust (and fixed the comments that describe what
1004 33. Some experimental protocols are using DNS PTR records for new purposes. The
1005 keys for these records are domain names, not reversed IP addresses. The
1006 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1007 leaves it alone. Component reversal etc. now happens only for IP addresses.
1010 34. The host_aton() function is supposed to be passed a string that is known
1011 to be a valid IP address. However, in the case of IPv6 addresses, it was
1012 not checking this. This is a hostage to fortune. Exim now panics and dies
1013 if the condition is not met. A case was found where this could be provoked
1014 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1015 components; fortuitously, this particular loophole had already been fixed
1016 by change 4.50/55 or 4.44/33 above.
1018 If there are any other similar loopholes, the new check in host_aton()
1019 itself should stop them being exploited. The report I received stated that
1020 data on the command line could provoke the exploit when Exim was running as
1021 exim, but did not say which command line option was involved. All I could
1022 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1023 running as the user.
1026 35. There was a buffer overflow vulnerability in the SPA authentication code
1027 (which came originally from the Samba project). I have added a test to the
1028 spa_base64_to_bits() function which I hope fixes it.
1031 36. The daemon start-up calls getloadavg() while still root for those OS that
1032 need the first call to be done as root, but it missed one case: when
1033 deliver_queue_load_max is set with deliver_drop_privilege. This is
1034 necessary for the benefit of the queue runner, because there is no re-exec
1035 when deliver_drop_privilege is set.
1037 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1038 was in use, was not putting the data itself into the right store pool;
1039 consequently, it could be overwritten for a subsequent message in the same
1040 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1043 38. Sometimes the final signoff response after QUIT could fail to get
1044 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1045 < 0 before doing a fflush(). This bug looks as though it goes back to the
1046 introduction of TLS in release 3.20, but "sometimes" must have been rare
1047 because the tests only now provoked it.
1053 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1054 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1055 Now it returns an integer. A number of calls were still expecting a T/F
1056 return. Fortuitously, in all cases, the tests worked in OK situations,
1057 which is the norm. However, things would have gone wrong on any write
1058 failures on the smtp file descriptor. This function is used when sending
1059 messages over SMTP and also when doing verify callouts.
1061 2. When Exim is called to do synchronous delivery of a locally submitted
1062 message (the -odf or -odi options), it no longer closes stderr before doing
1065 3. Implemented the mua_wrapper option.
1067 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1069 5. Implemented the functions header_remove(), header_testname(),
1070 header_add_at_position(), and receive_remove_recipient(), and exported them
1073 6. If an ACL "warn" statement specified the addition of headers, Exim already
1074 inserted X-ACL-Warn: at the start if there was no header name. However, it
1075 was not making this test for the second and subsequent header lines if
1076 there were newlines in the string. This meant that an invalid header could
1077 be inserted if Exim was badly configured.
1079 7. Allow an ACL "warn" statement to add header lines at the start or after all
1080 the Received: headers, as well as at the end.
1082 8. Added the rcpt_4xx retry error code.
1084 9. Added postmaster_mailfrom=xxx to callout verification option.
1086 10. Added mailfrom=xxxx to the callout verification option, for verify=
1089 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1090 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1092 12. Inserted some casts to stop certain compilers warning when using pointer
1093 differences as field lengths or precisions in printf-type calls (mostly
1094 affecting debugging statements).
1096 13. Added optional readline() support for -be (dynamically loaded).
1098 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1099 same clock tick as a message's arrival, so that its received time was the
1100 same as the "first fail" time on the retry record, and that message
1101 remained on the queue past the ultimate address timeout, every queue runner
1102 would try a delivery (because it was past the ultimate address timeout) but
1103 after another failure, the ultimate address timeout, which should have then
1104 bounced the address, did not kick in. This was a "< instead of <=" error;
1105 in most cases the first failure would have been in the next clock tick
1106 after the received time, and all would be well.
1108 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1109 being recognized when the domain list was tested by the match_domain
1110 condition in an expansion string.
1112 16. Added the ${str2b64: operator.
1114 17. Exim was always calling setrlimit() to set a large limit for the number of
1115 processes, without checking whether the existing limit was already
1116 adequate. (It did check for the limit on file descriptors.) Furthermore,
1117 errors from getrlimit() and setrlimit() were being ignored. Now they are
1118 logged to the main and panic logs, but Exim does carry on, to try to do its
1119 job under whatever limits there are.
1121 18. Imported PCRE 5.0.
1123 19. Trivial typo in log message " temporarily refused connection" (the leading
1126 20. If the log selector return_path_on_delivery was set and an address was
1127 redirected to /dev/null, the delivery process crashed because it assumed
1128 that a return path would always be set for a "successful" delivery. In this
1129 case, the whole delivery is bypassed as an optimization, and therefore no
1132 21. Internal re-arrangement: the function for sending a challenge and reading
1133 a response while authentication was assuming a zero-terminated challenge
1134 string. It's now changed to take a pointer and a length, to allow for
1135 binary data in such strings.
1137 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1139 23. Exim was not respecting finduser_retries when seeking the login of the
1140 uid under which it was called; it was always trying 10 times. (The default
1141 setting of finduser_retries is zero.) Also, it was sleeping after the final
1142 failure, which is pointless.
1144 24. Implemented tls_on_connect_ports.
1146 25. Implemented acl_smtp_predata.
1148 26. If the domain in control=submission is set empty, Exim assumes that the
1149 authenticated id is a complete email address when it generates From: or
1150 Sender: header lines.
1152 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1153 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1154 chown and chgrp in /bin and hostname in /usr/bin.
1156 28. Exim was keeping the "process log" file open after each use, just as it
1157 does for the main log. This opens the possibility of it remaining open for
1158 long periods when the USR1 signal hits a daemon. Occasional processlog
1159 errors were reported, that could have been caused by this. Anyway, it seems
1160 much more sensible not to leave this file open at all, so that is what now
1163 29. The long-running daemon process does not normally write to the log once it
1164 has entered its main loop, and it closes the log before doing so. This is
1165 so that log files can straightforwardly be renamed and moved. However,
1166 there are a couple of unusual error situations where the daemon does write
1167 log entries, and I had neglected to close the log afterwards.
1169 30. The text of an SMTP error response that was received during a remote
1170 delivery was being truncated at 512 bytes. This is too short for some of
1171 the long messages that one sometimes sees. I've increased the limit to
1174 31. It is now possible to make retry rules that apply only when a message has a
1175 specific sender, in particular, an empty sender.
1177 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1178 it possible to be selective about when SMTP synchronization is enforced.
1180 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1182 32. Implemented hosts_connection_nolog.
1184 33. Added an ACL for QUIT.
1186 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1189 35. Added mailbox_size and mailbox_filecount to appendfile.
1191 36. Added control = no_multiline_responses to ACLs.
1193 37. There was a bug in the logic of the code that waits for the clock to tick
1194 in the case where the clock went backwards by a substantial amount such
1195 that the microsecond fraction of "now" was more than the microsecond
1196 fraction of "then" (but the whole seconds number was less).
1198 38. Added support for the libradius Radius client library this is found on
1199 FreeBSD (previously only the radiusclient library was supported).
1205 1. When certain lookups returned multiple values in the form name=value, the
1206 quoting of the values was not always being done properly. Specifically:
1207 (a) If the value started with a double quote, but contained no whitespace,
1209 (b) If the value contained whitespace other than a space character (i.e.
1210 tabs or newlines or carriage returns) it was not quoted.
1211 This fix has been applied to the mysql and pgsql lookups by writing a
1212 separate quoting function and calling it from the lookup code. The fix
1213 should probably also be applied to nisplus, ibase and oracle lookups, but
1214 since I cannot test any of those, I have not disturbed their existing code.
1216 2. A hit in the callout cache for a specific address caused a log line with no
1217 reason for rejecting RCPT. Now it says "Previous (cached) callout
1218 verification failure".
1220 3. There was an off-by-one bug in the queryprogram router. An over-long
1221 return line was truncated at 256 instead of 255 characters, thereby
1222 overflowing its buffer with the terminating zero. As well as fixing this, I
1223 have increased the buffer size to 1024 (and made a note to document this).
1225 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1226 when Exim is waiting for an SMTP response from a remote server, Exim
1227 restarts its select() call on the socket, thereby resetting its timeout.
1228 This is not a problem when such interrupts are rare. Somebody set up a cron
1229 job to run exiwhat every 2 minutes, which is less than the normal select()
1230 timeout (5 or 10 minutes). This meant that the select() timeout never
1231 kicked in because it was always reset. I have fixed this by comparing the
1232 time when an interrupt arrives with the time at the start of the first call
1233 to select(). If more time than the timeout has elapsed, the interrupt is
1234 treated as a timeout.
1236 5. Some internal re-factoring in preparation for the addition of Sieve
1237 extensions (by MH). In particular, the "personal" test is moved to a
1238 separate function, and given an option for scanning Cc: and Bcc: (which is
1239 not set for Exim filters).
1241 6. When Exim created an email address using the login of the caller as the
1242 local part (e.g. when creating a From: or Sender: header line), it was not
1243 quoting the local part when it contained special characters such as @.
1245 7. Installed new OpenBSD configuration files.
1247 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1248 try to make them clearer.
1250 9. Callout options, other than the timeout value, were being ignored when
1251 verifying sender addresses in header lines. For example, when using
1253 verify = header_sender/callout=no_cache
1255 the cache was (incorrectly) being used.
1257 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1258 only the Cygwin environment.
1260 11. When return_path_on_delivery was set as a log selector, if different remote
1261 addresses in the same message used different return paths and parallel
1262 remote delivery occurred, the wrong values would sometimes be logged.
1263 (Whenever a remote delivery process finished, the return path value from
1264 the most recently started remote delivery process was logged.)
1266 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1267 lines when AUTH and/or TLS are in use. This is the "received protocol"
1268 field. Exim used to use "asmtp" for authenticated SMTP, without any
1269 indication (in the protocol name) for TLS use. Now it follows the RFC and
1270 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1271 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1272 names appear in log lines as well as in Received: header lines.
1274 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1275 extensions, and comparison tests, and to fix some bugs.
1277 14. Changes to the "personal" filter test:
1279 (1) The test was buggy in that it was just doing the equivalent of
1280 "contains" tests on header lines. For example, if a user's address was
1281 anne@some.where, the "personal" test would incorrectly be true for
1283 To: susanne@some.where
1285 This test is now done by extracting each address from the header in turn,
1286 and checking the entire address. Other tests that are part of "personal"
1287 are now done using regular expressions (for example, to check local parts
1288 of addresses in From: header lines).
1290 (2) The list of non-personal local parts in From: addresses has been
1291 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1292 taken from the Sieve specification recommendations.
1294 (3) If the message contains any header line starting with "List-" it is
1295 treated as non-personal.
1297 (4) The test for "circular" in the Subject: header line has been removed
1298 because it now seems ill-conceived.
1300 15. Minor typos in src/EDITME comments corrected.
1302 16. Installed latest exipick from John Jetmore.
1304 17. If headers_add on a router specified a text string that was too long for
1305 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1306 of string_sprintf() is now avoided.
1308 18. $message_body_size was not set (it was always zero) when running the DATA
1309 ACL and the local_scan() function.
1311 19. For the "mail" command in an Exim filter, no default was being set for
1312 the once_repeat time, causing a random time value to be used if "once" was
1313 specified. (If the value happened to be <= 0, no repeat happened.) The
1314 default is now 0s, meaning "never repeat". The "vacation" command was OK
1315 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1316 (I found it when inspecting the code).
1318 20. There is now an overall timeout for performing a callout verification. It
1319 defaults to 4 times the callout timeout, which applies to individual SMTP
1320 commands during the callout. The overall timeout applies when there is more
1321 than one host that can be tried. The timeout is checked before trying the
1322 next host. This prevents very long delays if there are a large number of
1323 hosts and all are timing out (e.g. when the network connections are timing
1324 out). The value of the overall timeout can be changed by specifying an
1325 additional sub-option for "callout", called "maxwait". For example:
1327 verify = sender/callout=5s,maxwait=20s
1329 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1330 to the end before writing, but this should make it even safer).
1332 22. Exim was forgetting that it had advertised PIPELINING for the second and
1333 subsequent messages on an SMTP connection. It was also not resetting its
1334 memory on STARTTLS and an internal HELO.
1336 23. When Exim logs an SMTP synchronization error within a session, it now
1337 records whether PIPELINING has been advertised or not.
1339 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1340 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1341 rather than long int.
1343 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1345 26. Added the never_mail option to autoreply.
1351 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1352 crash if the getsockname() call failed; this can happen if a connection is
1353 closed very soon after it is established. The problem was simply in the
1354 order in which certain operations were done, causing Exim to try to write
1355 to the SMTP stream before it had set up the file descriptor. The bug has
1356 been fixed by making things happen in the correct order.
1362 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1363 before the connection was closed, thus losing the rejection response.
1365 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1366 some early Solaris releases, but causes trouble in current releases where
1367 socklen_t is defined.
1369 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1372 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1375 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1376 that releasing the top of it at the end releases what was used for sub-
1377 expansions (unless the block got too big). However, discard this block if
1378 the first thing is a variable or header, so that we can use its block when
1379 it is dynamic (useful for very large $message_headers, for example).
1381 6. Lookups now cache *every* query, not just the most recent. A new, separate
1382 store pool is used for this. It can be recovered when all lookup caches are
1383 flushed. Lookups now release memory at the end of their result strings.
1384 This has involved some general refactoring of the lookup sources.
1386 7. Some code has been added to the store_xxx() functions to reduce the amount
1387 of flapping under certain conditions.
1389 8. log_incoming_interface used to affect only the <= reception log lines. Now
1390 it causes the local interface and port to be added to several more SMTP log
1391 lines, for example "SMTP connection from", and rejection lines.
1393 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1395 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1397 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1398 could be overwritten at the end of the current message (or the start of a
1399 new message if it was set in a HELO ACL). The value is now preserved for
1400 the duration of the SMTP connection.
1402 12. If a transport had a headers_rewrite setting, and a matching header line
1403 contained an unqualified address, that address was qualified, even if it
1404 did not match any rewriting rules. The underlying bug was that the values
1405 of the flags that permit the existence of unqualified sender and recipient
1406 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1407 non-local messages, and by -bnq for local messages) were not being
1408 preserved with the message after it was received.
1410 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1411 "next input=" as part of the text comprising the host identity instead of
1412 the correct text. The code was using the same buffer for two different
1413 strings. However, depending on which order the printing function evaluated
1414 its arguments, the bug did not always show up. Under Linux, for example, my
1415 test suite worked just fine.
1417 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1418 doesn't work with some older versions of Perl. It has been changed to "my",
1419 which in any case is probably the better facility to use.
1421 15. A really picky compiler found some instances of statements for creating
1422 error messages that either had too many or two few arguments for the format
1425 16. The size of the buffer for calls to the DNS resolver has been increased
1426 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1427 for addresses that have a lot of PTR records. This alleviates a problem; it
1428 does not fully solve it.
1430 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1431 buffer now truncates the list and logs the incident, which is the same
1432 action as happens when Exim is looking up a host name and its aliases.
1433 Previously in this situation something unpredictable would happen;
1434 sometimes it was "internal error: store_reset failed".
1436 18. If a server dropped the connection unexpectedly when an Exim client was
1437 using GnuTLS and trying to read a response, the client delivery process
1438 crashed while trying to generate an error log message.
1440 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1441 string, for example:
1443 warn message = H1: something\nH2: something
1445 the text was added as a single header line from Exim's point of view
1446 though it ended up OK in the delivered message. However, searching for the
1447 second and subsequent header lines using $h_h2: did not work. This has been
1448 fixed. Similarly, if a system filter added multiple headers in this way,
1449 the routers could not see them.
1451 20. Expanded the error message when iplsearch is called with an invalid key to
1452 suggest using net-iplsearch in a host list.
1454 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1455 ACLs are no longer actually imposed (and a message to that effect is
1458 22. If a "gecos" field in a passwd entry contained escaped characters, in
1459 particular, if it contained a \" sequence, Exim got it wrong when building
1460 a From: or a Sender: header from that name. A second bug also caused
1461 incorrect handling when an unquoted " was present following a character
1462 that needed quoting.
1464 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1465 was not being matched caselessly.
1467 24. Arranged for all hyphens in the exim.8 source to be escaped with
1470 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1471 properly. Recipient callout cache records were still being keyed to include
1472 the sender, even when use_sender was set false. This led to far more
1473 callouts that were necessary. The sender is no longer included in the key
1474 when use_sender is false.
1476 26. Added "control = submission" modifier to ACLs.
1478 27. Added the ${base62d: operator to decode base 62 numbers.
1480 28. dnsdb lookups can now access SRV records.
1482 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1483 the configuration file.
1485 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1486 (-v) mode. This makes the output for a verbose queue run more intelligible.
1488 31. Added a use_postmaster feature to recipient callouts.
1490 32. Added the $body_zerocount variable, containing the number of binary zero
1491 bytes in the message body.
1493 33. The time of last modification of the "new" subdirectory is now used as the
1494 "mailbox time last read" when there is a quota error for a maildir
1497 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1499 35. Added +ignore_unknown as a special item in host lists.
1501 36. Code for decoding IPv6 addresses in host lists is now included, even if
1502 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1503 address was recognized as an IP address, but was then not correctly decoded
1504 into binary, causing unexpected and incorrect effects when compared with
1511 1. Very minor rewording of debugging text in manualroute to say "list of
1512 hosts" instead of "hostlist".
1514 2. If verify=header_syntax was set, and a header line with an unqualified
1515 address (no domain) and a large number of spaces between the end of the
1516 name and the colon was received, the reception process suffered a buffer
1517 overflow, and (when I tested it) crashed. This was caused by some obsolete
1518 code that should have been removed. The fix is to remove it!
1520 3. When running in the test harness, delay a bit after writing a bounce
1521 message to get a bit more predictability in the log output.
1523 4. Added a call to search_tidyup() just before forking a reception process. In
1524 theory, someone could use a lookup in the expansion of smtp_accept_max_
1525 per_host which, without the tidyup, could leave open a database connection.
1527 5. Added the variables $recipient_data and $sender_data which get set from a
1528 lookup success in an ACL "recipients" or "senders" condition, or a router
1529 "senders" option, similar to $domain_data and $local_part_data.
1531 6. Moved the writing of debug_print from before to after the "senders" test
1534 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1535 problems for message scanning, either using a data ACL, or using
1536 local_scan() because the Received: header was not generated till after they
1537 were called (in order to set the time as the time of reception completion).
1538 I have revised the way this works. The header is now generated after the
1539 body is received, but before the ACL or local_scan() are called. After they
1540 are run, the timestamp in the header is updated.
1546 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1547 before starting a queue runner without re-exec. This happened only when
1548 deliver_drop_privilege was set or when the Exim user was set to root. The
1549 effect of the bug was that timeouts during subsequent deliveries caused
1550 crashes instead of being properly handled. The handler is now left at its
1551 default (and expected) setting.
1553 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1554 message, again when deliver_drop_privilege is set or Exim is run as root.
1555 The bug described in (1) was not present in this case, but the tidying up
1556 of the other signals was missing. I have made the two cases consistent.
1558 3. The ignore_target_hosts setting on a manualroute router was being ignored
1559 for hosts that were looked up using the /MX notation.
1561 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1564 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1565 operated on the sender address. After changing the $sender_address to <>
1566 for the sender address verify, Exim was re-instated it as the original
1567 (before rewriting) address, but remembering that it had rewritten it, so it
1568 wasn't rewriting it again. This bug also had the effect of breaking the
1569 sender address verification caching when the sender address was rewritten.
1571 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1572 This has been changed so that if the ip literal address matches
1573 ignore_target_hosts, the router declines.
1575 7. Added expansion conditions match_domain, match_address, and match_local_
1576 part (NOT match_host).
1578 8. The placeholder for the Received: header didn't have a length field set.
1580 9. Added code to Exim itself and to exim_lock to test for a specific race
1581 condition that could lead to file corruption when using MBX delivery. The
1582 issue is with the lockfile that is created in /tmp. If this file is removed
1583 after a process has opened it but before that process has acquired a lock,
1584 there is the potential for a second process to recreate the file and also
1585 acquire a lock. This could lead to two Exim processes writing to the file
1586 at the same time. The added code performs the same test as UW imapd; it
1587 checks after acquiring the lock that its file descriptor still refers to
1588 the same named file.
1590 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1591 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1594 11. Added the smtp_active_hostname option. If used, this will typically be made
1595 to depend on the incoming interface address. Because $interface_address is
1596 not set up until the daemon has forked a reception process, error responses
1597 that can happen earlier (such as "too many connections") no longer contain
1600 12. If an expansion in a condition on a "warn" statement fails because a lookup
1601 defers, the "warn" statement is abandoned, and the next ACL statement is
1602 processed. Previously this caused the whole ACL to be aborted.
1604 13. Added the iplsearch lookup type.
1606 14. Added ident_timeout as a log selector.
1608 15. Added tls_certificate_verified as a log selector.
1610 16. Added a global option tls_require_ciphers (compare the smtp transport
1611 option of the same name). This controls incoming TLS connections.
1613 17. I finally figured out how to make tls_require_ciphers do a similar thing
1614 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1615 before starting the TLS session.
1617 18. Tabs are now shown as \t in -bP output.
1619 19. If the log selector return_path_on_delivery was set, Exim crashed when
1620 bouncing a message because it had too many Received: header lines.
1622 20. If two routers both had headers_remove settings, and the first one included
1623 a superfluous trailing colon, the final name in the first list and the
1624 first name in the second list were incorrectly joined into one item (with a
1625 colon in the middle).
1631 1. Added -C and -D options to the exinext utility, mainly to make it easier
1632 to include in the automated testing, but these could be helpful when
1633 multiple configurations are in use.
1635 2. The exinext utility was not formatting the output nicely when there was
1636 an alternate port involved in the retry record key, nor when there was a
1637 message id as well (for retries that were specific to a specific message
1638 and a specific host). It was also confused by IPv6 addresses, because of
1639 the additional colons they contain. I have fixed the IPv4 problem, and
1640 patched it up to do a reasonable job for IPv6.
1642 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1643 delivery, the log line now contains "pipelined" if PIPELINING was used.
1645 4. An SMTP transport process used to panic and die if the bind() call to set
1646 an explicit outgoing interface failed. This has been changed; it is now
1647 treated in the same way as a connect() failure.
1649 5. A reference to $sender_host_name in the part of a conditional expansion
1650 that was being skipped was still causing a DNS lookup. This no longer
1653 6. The def: expansion condition was not recognizing references to header lines
1654 that used bh_ and bheader_.
1656 7. Added the _cache feature to named lists.
1658 8. The code for checking quota_filecount in the appendfile transport was
1659 allowing one more file than it should have been.
1661 9. For compatibility with Sendmail, the command line option
1669 and sets the incoming protocol and host name (for trusted callers). The
1670 host name and its colon can be omitted when only the protocol is to be set.
1671 Note the Exim already has two private options, -pd and -ps, that refer to
1672 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1673 "s", but I don't think that's a major issue.
1675 10. A number of refactoring changes to the code, none of which should affect
1678 (a) The number of logging options was getting close to filling up the
1679 32-bit word that was used as a bit map. I have split them into two classes:
1680 those that are passed in the argument to log_write(), and those that are
1681 only ever tested independently outside of that function. These are now in
1682 separate 32-bit words, so there is plenty of room for expansion again.
1683 There is no change in the user interface or the logging behaviour.
1685 (b) When building, for example, log lines, the code previously used a
1686 macro that called string_cat() twice, in order to add two strings. This is
1687 not really sufficiently general. Furthermore, there was one instance where
1688 it was actually wrong because one of the argument was used twice, and in
1689 one call a function was used. (As it happened, calling the function twice
1690 did not affect the overall behaviour.) The macro has been replaced by a
1691 function that can join an arbitrary number of extra strings onto a growing
1694 (c) The code for expansion conditions now uses a table and a binary chop
1695 instead of a serial search (which was left over from when there were very
1696 few conditions). Also, it now recognizes conditions like "pam" even when
1697 the relevant support is not compiled in: a suitably worded error message is
1698 given if an attempt is made to use such a condition.
1700 11. Added ${time_interval:xxxxx}.
1702 12. A bug was causing one of the ddress fields not to be passed back correctly
1703 from remote delivery subprocesses. The field in question was not being
1704 subsequently used, so this caused to problems in practice.
1706 13. Added new log selectors queue_time and deliver_time.
1708 14. Might have fixed a bug in maildirsizefile handling that threw up
1709 "unexpected character" debug warnings, and recalculated the data
1710 unnecessarily. In any case, I expanded the warning message to give more
1713 15. Added the message "Restricted characters in address" to the statements in
1714 the default ACL that block characters like @ and % in local parts.
1716 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
1717 Three changes have been made:
1719 (a) There was a serious bug; a negative response to MAIL caused the whole
1720 recipient domain to be cached as invalid, thereby blocking all messages
1721 to all local parts at the same domain, from all senders. This bug has
1722 been fixed. The domain is no longer cached after a negative response to
1723 MAIL if the sender used is not empty.
1725 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
1728 (c) A new callout option, "use_sender" has been added for people who want
1729 the modified behaviour.
1735 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
1738 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
1739 indeed breaks things for older releases.
1741 3. Added additional logging to the case where there is a problem reading data
1742 from a filter that is running in a subprocess using a pipe, in order to
1743 try to track down a specific problem.
1745 4. Testing facility fudge: when running in the test harness and attempting
1746 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
1747 getting "No route to host". Convert this to a timeout.
1749 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
1752 6. Some OS don't have socklen_t but use size_t instead. This affects the
1753 fifth argument of getsockopt() amongst other things. This is now
1754 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
1755 can be set for individual OS. I have set it for SunOS5, OSF1, and
1756 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
1757 some earlier ones do not.
1759 7. Change 4.30/15 was not doing the test caselessly.
1761 8. The standard form for an IPv6 address literal was being rejected by address
1762 parsing in, for example, MAIL and RCPT commands. An example of this kind of
1763 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1764 this, as well as the form without the "IPv6" on the front (but only when
1765 address literals are enabled, of course).
1767 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
1769 10. Exim crashed if a message with an empty sender address specified by -f
1770 encountered a router with an errors_to setting. This could be provoked only
1771 by a command such as
1775 where an empty string was supplied; "<>" did not hit this bug.
1777 11. Installed PCRE release 4.5.
1779 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
1780 remained set. It is now erased.
1782 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
1783 times from message ids (which are base 36 rather than the normal 62).
1785 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
1786 were being counted as actual protocol errors, and logged if the log
1787 selector +smtp_protocol_error was set. One cannot be perfect in this test,
1788 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
1789 and DATA following a set of rejected RCPTs do not count as protocol errors.
1790 In other words, Exim assumes they were pipelined, though this may not
1791 actually be the case. Of course, in all cases the client gets an
1792 appropriate error code.
1794 15. If a lookup fails in an ACL condition, a message about the failure may
1795 be available; it is used if testing the ACL cannot continue, because most
1796 such messages specify what the cause of the deferral is. However, some
1797 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
1798 that caused an old message to be retained and used if a later statement
1799 caused a defer, replacing the real cause of the deferral.
1801 16. If an IP address had so many PTR records that the DNS lookup buffer
1802 was not large enough to hold them, Exim could crash while trying to process
1803 the truncated data. It now detects and logs this case.
1805 17. Further to 4.21/58, another change has been made: if (and only if) the
1806 first line of a message (the first header line) ends with CRLF, a bare LF
1807 in a subsequent header line has a space inserted after it, so as not to
1808 terminate the header.
1810 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
1811 unnecessarily, used atoi() instead of strtol(), and didn't check the
1812 termination when getting file sizes from file names by regex.
1814 19. Completely re-implemented the support for maildirsize files, in the light
1815 of a number of problems with the previous contributed implementation
1816 (4.30/29). In particular:
1818 . If the quota is zero, the maildirsize file is maintained, but no quota is
1821 . If the maildir directory does not exist, it is created before any attempt
1822 to write a maildirsize file.
1824 . The quota value in the file is just a cache; if the quota is changed in
1825 the transport, the new value overrides.
1827 . A regular expression is available for excluding directories from the
1830 20. The autoreply transport checks the characters in options that define the
1831 message's headers; it allows continued headers, but it was checking with
1832 isspace() after an embedded newline instead of explicitly looking for a
1835 21. If all the "regular" hosts to which an address was routed had passed their
1836 expiry times, and had not reached their retry times, the address was
1837 bounced, even if fallback hosts were defined. Now Exim should go on to try
1840 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
1841 equivalent code in the SMTP transport. Some hosts send humungous responses
1842 to HELO/EHLO, more than 1024 it seems.
1844 23. Refactoring: code in filter.c used (void *) for "any old type" but this
1845 gives compiler warnings in some environments. I've now done it "properly",
1848 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
1849 (because of problems with the built-in one) was declared to return uschar *
1850 instead of char *, causing compiler failure.
1852 25. Fixed a file descriptor leak when processing alias/forward files.
1854 26. Fixed a minor format string issue in dbfn.c.
1856 27. Typo in exim.c: ("dmbnz" for "dbmnz").
1858 28. If a filter file refered to $h_xxx or $message_headers, and the headers
1859 contained RFC 2047 "words", Exim's memory could, under certain conditions,
1862 29. When a sender address is verified, it is cached, to save repeating the test
1863 when there is more than one recipient in a message. However, when the
1864 verification involves a callout, it is possible for different callout
1865 options to be set for different recipients. It is too complicated to keep
1866 track of this in the cache, so now Exim always runs a verification when a
1867 callout is required, relying on the callout cache for the optimization.
1868 The overhead is duplication of the address routing, but this should not be
1871 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
1872 to be verified with callout=postmaster, and the main callout worked but the
1873 postmaster check failed, the verification correctly failed. However, if a
1874 subsequent RCPT command asked for sender verification *without* the
1875 postmaster check, incorrect caching caused this verification also to fail,
1878 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
1879 it was not caching the DNS options (qualify_single, search_parents) that
1880 were used when the lookup failed. A subsequent lookup with different
1881 options therefore always gave the same answer, though there were cases
1882 where it should not have. (Example: a "domains = !$mx_any" option on a
1883 dnslookup router: the "domains" option is always processed without any
1884 widening, but the router might have qualify_single set.) Now Exim uses the
1885 cached value only when the same options are set.
1887 32. Added John Jetmore's "exipick" utility to the distribution.
1889 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
1890 than a timeout (e.g. a certificate is required, and is not provided), an
1891 Exim server now closes the connection immediately. Previously it waited for
1892 the client to close - but if the client is SSL, it seems that they each
1893 wait for each other, leading to a delay before one of them times out.
1895 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
1896 maintained 0.8.x compatibility because I don't think many are using it, and
1897 it is clearly obsolete.
1899 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
1902 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
1903 client certificate was expired. A simple patch fixes this, though I don't
1904 understand the full logic of why the verify callback is called multiple
1907 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
1908 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
1909 which causes problems with some clients (such as the Certicom SSL Plus
1910 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
1911 disables the coutermeasure allowing Eudora to connect."
1913 38. Exim was not checking that a write() to a log file succeeded. This could
1914 lead to Bad Things if a log got too big, in particular if it hit a file
1915 size limit. Exim now panics and dies if it cannot write to a log file, just
1916 as it does if it cannot open a log file.
1918 39. Modified OS/Makefile-Linux so that it now contains
1920 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
1922 The two -D definitions ensure that Exim is compiled with large file
1923 support, which makes it possible to handle log files that are bigger than
1926 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
1927 instance) a domain was checked against a named list that involved a lookup,
1928 causing $domain_data to be set, then another domain was checked against the
1929 same list, then the first domain was re-checked, the value of $domain_data
1930 after the final check could be wrong. In particular, if the second check
1931 failed, it could be set empty. This bug probably also applied to
1934 41. The strip_trailing_dot option was not being applied to the address given
1935 with the -f command-line option.
1937 42. The code for reading a message's header from the spool was incrementing
1938 $received_count, but never initializing it. This meant that the value was
1939 incorrect (doubled) while delivering a message in the same process in which
1940 it was received. In the most common configuration of Exim, this never
1941 happens - a fresh exec is done - but it can happen when
1942 deliver_drop_privilege is set.
1944 43. When Exim logs an SMTP synchronization error - client data sent too soon -
1945 it now includes up to 150 characters of the unexpected data in the log
1948 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
1949 and building data strings. The size of both of these buffers was 10 000
1950 bytes - far larger than anybody would *ever* want, thought I. Needless to
1951 say, somebody hit the limit. I have increased the maximum line length to
1952 20 000 and the maximum data length of concatenated lines to 100 000. I have
1953 also fixed two bugs, because there was no checking on these buffers. Tsk,
1954 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1955 buffer is too small.
1957 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1958 lsearch lookups. Now it does.
1960 46. When parsing a route_list item in a manualroute router, a fixed-length
1961 buffer was used for the list of hosts. I made this 1024 bytes long,
1962 thinking that nobody would ever have a list of hosts that long. Wrong.
1963 Somebody had a whole pile of complicated expansion conditions, and the
1964 string was silently truncated, leading to an expansion error. It turns out
1965 that it is easier to change to an unlimited length (owing to other changes
1966 that have happened since this code was originally written) than to build
1967 structure for giving a limitation error. The length of the item that
1968 expands into the list of hosts is now unlimited.
1970 47. The lsearch lookup could not handle data where the length of text line was
1971 more than 4095 characters. Such lines were truncated, leading to shortened
1972 data being returned. It should now handle lines of any length.
1974 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
1975 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
1978 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
1979 replacing strings such as BIN_DIRECTORY in a source file: the replacement
1980 no longer happens in comment lines. A list of replacements is now placed
1981 at the head of all of the source files, except those whose only change is
1982 to replace PERL_COMMAND in the very first #! line.
1984 50. Replaced the slow insertion sort in queue.c, for sorting the list of
1985 messages on the queue, with a bottom-up merge sort, using code contributed
1986 by Michael Haardt. This should make operations like -bp somewhat faster on
1987 large queues. It won't affect queue runners, except when queue_run_in_order
1990 51. Installed eximstats 1.31 in the distribution.
1992 52. Added support for SRV lookups to the dnslookup router.
1994 53. If an ACL referred to $message_body or $message_body_end, the value was not
1995 reset for any messages that followed in the same SMTP session.
1997 54. The store-handling optimization for building very long strings was not
1998 differentiating between the different store pools. I don't think this
1999 actually made any difference in practice, but I've tidied it.
2001 55. While running the routers to verify a sender address, $sender_address
2002 was still set to the sender address. This is wrong, because when routing to
2003 send a bounce to the sender, it would be empty. Therefore, I have changed
2004 it so that, while verifying a sender address, $sender_address is set to <>.
2005 (There is no change to what happens when verifying a recipient address.)
2007 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2008 target A or AAAA records (if not already returned) without resetting the
2009 qualify_single or search_parents options of the DNS resolver. These are
2010 inappropriate in this case because the targets of MX and SRV records must
2011 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2012 target that, when qualified, matched something in the local domain. These
2013 two options are now turned off when doing these lookups.
2015 57. It seems that at least some releases of Reiserfs (which does not have the
2016 concept of a fixed number of inodes) returns zero and not -1 for the
2017 number of available inodes. This interacted badly with check_spool_inodes,
2018 which assumed that -1 was the "no such thing" setting. What I have done is
2019 to check that the total number of inodes is greater than zero before doing
2020 the test of how many are available.
2022 58. When a "warn" ACL statement has a log_message modifier, the message is
2023 remembered, and not repeated. This is to avoid a lot of repetition when a
2024 message has many recipients that cause the same warning to be written.
2025 Howewer, Exim was preserving the list of already written lines for an
2026 entire SMTP session, which doesn't seem right. The memory is now reset if a
2027 new message is started.
2029 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2030 debugging output unless log_rewrite was also set.
2032 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2033 of (int)(handle) when we know that handle contains (void *)(-1).
2035 61. The Exim daemon panic-logs an error return when it closes the incoming
2036 connection. However "connection reset by peer" seems to be common, and
2037 isn't really an error worthy of noting specially, so that particular error
2040 62. When Exim is trying to find all the local interfaces, it used to panic and
2041 die if the ioctl to get the interface flags failed. However, it seems that
2042 on at least one OS (Solaris 9) it is possible to have an interface that is
2043 included in the list of interfaces, but for which you get a failure error
2044 for this call. This happens when the interface is not "plumbed" into a
2045 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2046 failure of the "get flags" call assumes that the interface is down.
2048 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2049 makes life easier for people who are doing arithmetic on fields extracted
2050 from dates, where you often get leading zeros that should not be
2051 interpreted as octal.
2053 64. Added qualify_domain to the redirect router, to override the global
2056 65. If a pathologically long header line contained very many addresses (the
2057 report of this problem mentioned 10 000) and each of them was rewritten,
2058 Exim could use up a very large amount of memory. (It kept on making new
2059 copies of the header line as it rewrote, and never released the old ones.)
2060 At the expense of a bit more processing, the header rewriting function has
2061 been changed so that it no longer eats memory in this way.
2063 66. The generation of the Received: header has been moved from the time that a
2064 message starts to be received, to the time that it finishes. The timestamp
2065 in the Received: header should now be very close to that of the <= log
2066 line. There are two side-effects of this change:
2068 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2069 logged header lines no longer include the local Received: line, because
2070 it has not yet been created. The same applies to a copy of the message
2071 that is returned to a non-SMTP sender when a message is rejected.
2073 (b) When a filter file is tested using -bf, no additional Received: header
2074 is added to the test message. After some thought, I decided that this
2077 This change does not affect the value of $received_for. It is still set
2078 after address rewriting, but before local_scan() is called.
2080 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2082 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2083 gave an unhelpful panic error message, and a defer error. I have managed to
2084 change this behaviour so that it now rejects any supplied certificate,
2085 which seems right, as the list of acceptable certificates is empty.
2087 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2088 gave an unhelpful defer error. I have not managed to make this reject any
2089 supplied certificates, but the error message it gives is "no certificate
2090 supplied", which is not helpful.
2092 70. exigrep's output now also includes lines that are not associated with any
2093 message, but which match the given pattern. Implemented by a patch from
2094 Martin Sluka, which also tidied up the Perl a bit.
2096 71. Recipient callout verification, like sender verification, was using <> in
2097 the MAIL FROM command. This isn't really the right thing, since the actual
2098 sender may affect whether the remote host accepts the recipient or not. I
2099 have changed it to use the actual sender in the callout; this means that
2100 the cache record is now keyed on a recipient/sender pair, not just the
2101 recipient address. There doesn't seem to be a real danger of callout loops,
2102 since a callout by the remote host to check the sender would use <>.
2103 [SEE ABOVE: changed after hitting problems.]
2105 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2106 temporary errors. However, in the case of such a code being given after
2107 the end of a data transmission (i.e. after ".") Exim was failing to write
2108 a retry record for the message. (Yes, there was some broken host that was
2109 actually sending 8xx at this point.)
2111 73. An unknown lookup type in a host list could cause Exim to panic-die when
2112 the list was checked. (An example that provoked this was putting <; in the
2113 middle of a list instead of at the start.) If this happened during a DATA
2114 ACL check, a -D file could be left lying around. This kind of configuration
2115 error no longer causes Exim to die; instead it causes a defer errror. The
2116 incident is still logged to the main and panic logs.
2118 74. Buglet left over from Exim 3 conversion. The message "too many messages
2119 in one connection" was written to the rejectlog but not the mainlog, except
2120 when address rewriting (yes!) was being logged.
2122 75. Added write_rejectlog option.
2124 76. When a system filter was run not as root (that is, when system_filter_user
2125 was set), the values of the $n variables were not being returned to the
2126 main process; thus, they were not subsequently available in the $sn
2129 77. Added +return_path_on_delivery log selector.
2131 78. A connection timeout was being treated differently from recipients deferred
2132 when testing hosts_max_try with a message that was older than the host's
2133 retry timeout. (The host should not be counted, thus allowing all hosts to
2134 be tried at least once before bouncing.) This may have been the cause of an
2135 occasionally reported bug whereby a message would remain on the queue
2136 longer than the retry timeout, but would be bounced if a delivery was
2137 forced. I say "may" because I never totally pinned down the problem;
2138 setting up timeout/retry tests is difficult. See also the next item.
2140 79. The ultimate address timeout was not being applied to errors that involved
2141 a combination of host plus message (for example, a timeout on a MAIL
2142 command). When an address resolved to a number of possible hosts, and they
2143 were not all tried for each delivery (e.g. because of hosts_max_try), a
2144 message could remain on the queue longer than the retry timeout.
2146 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2149 81. Fixed an obscure SMTP outgoing bug which required at least the following
2150 conditions: (a) there was another message waiting for the same server;
2151 (b) the server returned 5xx to all RCPT commands in the first message so
2152 that the message was not completed; (c) the server dropped the connection
2153 or gave a negative response to the RSET that Exim sends to abort the
2154 transaction. The observed case was a dropped connection after DATA that had
2155 been sent in pipelining mode. That is, the server had advertised PIPELINING
2156 but was not implementing it correctly. The effect of the bug was incorrect
2157 behaviour, such as trying another host, and this could lead to a crash.
2163 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2164 and daemon.c were passed as pointers to ints; they should have been
2165 pointers to socklen_t variables (which are typically unsigned ints).
2167 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2170 3. Fixed a really odd bug that affected only the testing scheme; patching a
2171 certain fixed string in the binary changed the value of another string that
2172 happened to be identical to the end of the original first string.
2174 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2175 name", it returns that address as the IP address. On some operating
2176 systems (e.g. Solaris), it also passes back the IP address string as the
2177 "host name". However, on others (e.g. Linux), it passes back an empty
2178 string. Exim wasn't checking for this, and was changing the host name to an
2179 empty string, assuming it had been canonicized.
2181 5. Although rare, it is permitted to have more than one PTR record for a given
2182 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2183 all the names associated with an address, because they do in Solaris.
2184 However, it seems that they do not in Linux for data that comes from the
2185 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2186 I found this out when I moved to a new Linux workstation and tried to run
2187 the Exim test suite.
2189 To get round this problem I have changed the code so that it now does its
2190 own call to the DNS to look up PTR records when searching for a host name.
2191 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2192 addresses that are only in /etc/hosts are still found.
2194 This behaviour is, however, controlled by an option called host_lookup_
2195 order, which defaults to "bydns:byaddr". If people want to use the other
2196 order, or indeed, just use one or the other means of lookup, they can
2197 specify it in this variable.
2199 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2200 some operating systems, this comes back from gethostbyaddr() as an empty
2201 string, and this is what Exim used to test for. However, it seems that in
2202 other systems, "." is yielded. Exim now tests for this case too.
2204 7. The values of check_spool_space and check_log_space are now held internally
2205 as a number of kilobytes instead of an absolute number of bytes. If a
2206 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2207 kilobyte. This means that much larger values can be stored.
2209 8. Exim monitor: an attempt to get the action menu when not actually pointing
2210 at a message produces an empty menu entitled "No message selected". This
2211 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2212 no entries in it ("Shell widget menu has zero width and/or height"). So I
2213 have added a single, blank menu entry in this case.
2215 9. Added ${quote_local_part.
2217 10. MIME decoding is now applied to the contents of Subject: header lines when
2220 11. Now that a reference to $sender_host_address automatically causes a reverse
2221 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2222 host lookup before query-style lookups in lists that might use this
2223 variable. This has therefore been abolished, and the "net-" prefix is no
2224 longer necessary for query-style lookups.
2226 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2227 have been a typo for LFLAGS, so it has been changed.
2229 13. The install script calls Exim with "-C /dev/null" in order to find the
2230 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2231 to be output. Howeve, since Exim outputs its version number before the
2232 error, it didn't break the script. It just looked ugly. I fixed this by
2233 always allowing "-C /dev/null" if the caller is root.
2235 14. Ignore overlarge ACL variable number when reading spool file - insurance
2236 against a later release with more variables having written the file.
2238 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2239 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2240 this, as well as the form without the "IPv6" on the front.
2242 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2243 OS/Makefile-Darwin file.
2245 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2246 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2248 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2249 variable. However, some OS use ints for inodes. I've added cast to long int
2250 to get rid of the compiler warning.
2252 19. I had forgotten to lock out "/../" in configuration file names when
2253 ALT_CONFIG_PREFIX was set.
2255 20. Routers used for verification do not need to specify transports. However,
2256 if such a router generated a host list, and callout was configured, Exim
2257 crashed, because it could not find a port number from the (non-existent)
2258 transport. It now assumes port 25 in this circumstance.
2260 21. Added the -t option to exigrep.
2262 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2263 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2264 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2265 them is set, LOOKUP_LSEARCH is forced.
2267 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2269 24. Added sender and host information to the "rejected by local_scan()" log
2270 line; previously there was no indication of these.
2272 25. Added .include_if_exists.
2274 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2275 Linux. It turns out that not all file systems support this. Apparently some
2276 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2277 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2278 not supported on the file descriptor, is now ignored when Exim is trying to
2279 sync a directory. This applies only to Linux.
2281 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2283 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2284 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2287 29. Added support for maildirsize files from supplied patch (modified a bit).
2289 30. The use of :fail: followed by an empty string could lead Exim to respond to
2290 sender verification failures with (e.g.):
2292 550 Verification failed for <xxx>
2293 550 Sender verify failed
2295 where the first response line was missing the '-' that indicates it is not
2296 the final line of the response.
2298 31. The loop for finding the name of the user that called Exim had a hardwired
2299 limit of 10; it now uses the value of finduser_retries, which is used for
2300 all other user lookups.
2302 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2305 33. Exim was neglecting to zero errno before one call of strtol() when
2306 expanding a string and expecting an integer value. On some systems this
2307 resulted in spurious "integer overflow" errors. Also, it was casting the
2308 result into an int without checking.
2310 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2311 did not work. The code looks as if it has *never* worked, though it appears
2312 to have been documented since at least releast 1.62. I have made it work.
2314 35. The "timeout_DNS" error in retry rules, also documented since at least
2315 1.62, also never worked. As it isn't clear exactly what this means, and
2316 clearly it isn't a major issue, I have abolished the feature by treating it
2317 as "timeout", and writing a warning to the main and panic logs.
2319 36. The display of retry rules for -brt wasn't always showing the error code
2322 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2323 timeout_connect_A, timeout_connect_MX.
2325 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2326 to the empty sender.
2328 39. The daemon was not analysing the content of -oX till after it had closed
2329 stderr and disconnected from the controlling terminal. This meant that any
2330 syntax errors were only noted on the panic log, and the return code from
2331 the command was 0. By re-arranging the code a little, I've made the
2332 decoding happen first, so such errors now appear on stderr, and the return
2333 code is 1. However, the actual setting up of the sockets still happens in
2334 the disconnected process, so errors there are still only recorded on the
2337 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2338 connections (as happens on some IP stacks) was logged at start up time as
2339 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2340 it from "IPv6 and IPv4", which means that two separate sockets are being
2343 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2344 says whether AF_INET or AF_INET6 was passed as an argument.
2346 42. Exiwhat output was messed up when time zones were included in log
2349 43. Exiwhat now gives more information about the daemon's listening ports,
2350 and whether -tls-on-connect was used.
2352 44. The "port" option of the smtp transport is now expanded.
2354 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2355 silently ignored. Now an error message is written to the main and panic
2358 46. There's a new ACL modifier called "logwrite" which writes to a log file
2359 as soon as it is encountered.
2361 47. Added $local_user_uid and $local_user_gid at routing time.
2363 48. Exim crashed when trying to verify a sender address that was being
2366 49. Exim was recognizing only a space character after ".include". It now also
2367 recognizes a tab character.
2369 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2370 extracting the relevant information from the specification. The man page no
2371 longer contains scrambled data for the -d option, and I've added a section
2372 at the front about calling Exim under different names.
2374 51. Added "extra_headers" argument to the "mail" command in filter files.
2376 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2379 53. Installed eximstats 1.29.
2381 54. Added transport_filter_timeout as a generic transport option.
2383 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2384 Cc: header lines. This was required by RFC 822, but it not required by RFC
2387 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2388 incoming messages that did not have them. Now it does so only if the
2389 message originates locally, that is, if there is no associated remote host
2390 address. When Resent- header lines are present, this applies to the Resent-
2391 lines rather than the non-Resent- lines.
2393 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2394 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2396 58. Messages for configuration errors now include the name of the main
2397 configuration file - useful now that there may be more than one file in a
2398 list (.included file names were always shown).
2400 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2401 for those rare installations that do not start the daemon as root or run it
2402 setuid root. I've cut out the call to initgroups() if the daemon is not
2405 60. The Exim user and group can now be bound into the binary as text strings
2406 that are looked up at the start of Exim's processing.
2408 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2410 62. Added $mailstore_basename variable.
2412 63. Installed patch to sieve.c from Michael Haardt.
2414 64. When Exim failed to open the panic log after failing to open the main log,
2415 the original message it was trying to log was written to stderr and debug
2416 output, but if they were not available (the usual case in production), it
2417 was lost. Now it is written to syslog before the two lines that record the
2418 failures to open the logs.
2420 65. Users' Exim filters run in subprocesses under the user's uid. It is
2421 possible for a "deliver" command or an alias in a "personal" command to
2422 provoke an address rewrite. If logging of address rewriting is configured,
2423 this fails because the process is not running as root or exim. There may be
2424 a better way of dealing with this, but for the moment (because 4.30 needs
2425 to be released), I have disabled address rewrite logging when running a
2426 filter in a non-root, non-exim process.
2432 1. The buildconfig auxiliary program wasn't quoting the value set for
2433 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2434 not defined. This bug was masked in 4.22 by the effect that was fixed in
2437 2. Some messages that were rejected after a message id was allocated were
2438 shown as "incomplete" by exigrep. It no longer does this for messages that
2439 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2441 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2442 have allow_domain_literals set, the ID did not get logged in the <= line.
2443 Domain literals are now always recognized in Message-ID: header lines.
2445 4. The first argument for a ${extract expansion item is the key name or field
2446 number. Leading and trailing spaces in this item were not being ignored,
2447 causing some misleading effects.
2449 5. When deliver_drop_privilege was set, single queue runner processes started
2450 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2451 same command in the process it spins off) were not dropping privilege.
2453 6. When the daemon running as "exim" started a queue runner, it always
2454 re-executed Exim in the spun-off process. This is a waste of effort when
2455 deliver_drop_privilege is set. The new process now just calls the
2456 queue-runner function directly.
2462 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2465 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2466 ignored. Though the use of -oP was forcing the writing of a pid file, it
2467 was always written to the default place.
2469 3. If the message "no IP address found for host xxxx" is generated during
2470 incoming verification, it is now followed by identification of the incoming
2471 connection (so you can more easily find what provoked it).
2473 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2475 5. Added some features to "harden" Exim a bit more against certain attacks:
2477 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2478 be put in Local/Makefile. This is like the never_users runtime option,
2479 but it cannot be overridden. The default setting is "root".
2481 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2482 prefix string with which any file named in a -C command line option
2485 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2486 is retained for -C and -D only if the caller of Exim is root. Without
2487 it, the exim user may also use -C and -D and retain privilege.
2489 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2490 command line option is disabled.
2492 6. Macro names set by the -D option must start with an upper case letter, just
2493 like macro names defined in the configuration file.
2495 7. Added "dereference=" facility to LDAP.
2497 8. Two instances of the typo "uknown" in the source files are fixed.
2499 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2500 the Configure-Makefile script screwed up while processing it.
2502 10. Incorporated PCRE 4.4.
2504 11. The SMTP synchronization check was not operating right at the start of an
2505 SMTP session. For example, it could not catch a HELO sent before the client
2506 waited for the greeting. There is now a check for outstanding input at the
2507 point when the greeting is written. Because of the duplex, asynchronous
2508 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2509 way, but not yet received, when the check is performed.
2511 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2512 on TCP/IP sockets, because this apparently causes some broken clients to
2515 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2516 unchanged) from the Cygwin maintainer.
2518 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2519 was supported instead of testing for mbx. Effectively a typo.
2521 15. The spa authenticator server code was not checking that the input it
2522 received was valid base64.
2524 16. The debug output line for the "set" modifier in ACLs was not showing the
2525 name of the variable that was being set.
2527 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2529 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2530 lookup on its own. Something else was needed to trigger the lookup. For
2531 example, a match in host_lookup or the need for a host name in a host list.
2532 Now, if $sender_host_name is referenced and the host name has not yet been
2533 looked up, a lookup is performed. If the lookup fails, the variable remains
2534 empty, and $host_lookup_failed is set to "1".
2536 19. Added "eqi" as a case-independent comparison operator.
2538 20. The saslauthd authentication condition could segfault if neither service
2539 nor realm was specified.
2541 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2542 error message that was logged was misleading, and incoming SMTP
2543 connections were dropped. The message is now more accurate, and temporary
2544 errors are given to SMTP connections.
2546 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2547 (except RSET) with a 421 error, until QUIT is received. However, it was
2548 failing to send a response to QUIT.
2550 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2551 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2552 ACL is now run after the helo_try_verify_hosts code.
2554 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2555 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2556 case-independent, so other case variants are also recognized). Apparently
2557 some systems use these upper case variants.
2559 25. If more than two messages were waiting for the same host, and a transport
2560 filter was specified for the transport, Exim sent two messages over the
2561 same TCP/IP connection, and then failed with "socket operation on non-
2562 socket" when it tried to send the third.
2564 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2566 27. The extern definition of crypt16() in expand.c was not being excluded when
2567 the OS had its own crypt16() function.
2569 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2570 as a preferred synonym for return_size_limit, both as an option and as an
2573 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2575 30. Changed the default configuration ACL to relax the local part checking rule
2576 for addresses that are not in any local domains. For these addresses,
2577 slashes and pipe symbols are allowed within local parts, but the sequence
2578 /../ is explicitly forbidden.
2580 31. SPA server authentication was not clearing the challenge buffer before
2583 32. log_message in a "warn" ACL statement was writing to the reject log as
2584 well as to the main log, which contradicts the documentation and doesn't
2585 seem right (because no rejection is happening). So I have stopped it.
2587 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2588 However, I am unable to do any testing of this.
2590 34. Fixed an infelicity in the appendfile transport. When checking directories
2591 for a mailbox, to see if any needed to be created, it was accidentally
2592 using path names with one or more superfluous leading slashes; tracing
2593 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2595 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2596 contents are added to the log line that is written for every discarded
2597 recipient. (Previously a log_message setting was ignored.)
2599 36. The ${quote: operator now quotes the string if it is empty.
2601 37. The install script runs exim in order to find its version number. If for
2602 some reason other than non-existence or emptiness, which it checks, it
2603 could not run './exim', it was installing it with an empty version number,
2604 i.e. as "exim-". This error state is now caught, and the installation is
2607 38. An argument was missing from the function that creates an error message
2608 when Exim fails to connect to the socket for saslauthd authentication.
2609 This could cause Exim to crash, or give a corrupted message.
2611 39. Added isip, isip4, and isip6 to ${if conditions.
2613 40. The ACL variables $acl_xx are now saved with the message, and can be
2614 accessed later in routers, transports, and filters.
2616 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2617 strings in the file are not string-expanded.
2619 42. If a MAIL command specified a SIZE value that was too large to fit into an
2620 int variable, the check against message_size_limit failed. Such values are
2621 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2622 day this will have to be increased, but I don't think I want to be around
2623 when emails are that large.
2630 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2631 iconv() is not standard in FreeBSD.
2633 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2634 IPv6 enabled. The observed symptom was a segmentation fault on return from
2635 the function os_common_find_running_interfaces() in src/os.c.
2637 3. In the check_special_case() function in daemon.c I had used "errno" as an
2638 argument name, which causes warnings on some systems. This was basically a
2639 typo, since it was named "eno" in the comments!
2641 4. The code that waits for the clock to tick (at a resolution of some fraction
2642 of a second) so as to ensure message-id uniqueness was always waiting for
2643 at least one whole tick, when it could have waited for less. [This is
2644 almost certainly not relevant at current processor speeds, where it is
2645 unlikely to ever wait at all. But we try to future-proof.]
2647 5. The function that sleeps for a time interval that includes fractions of a
2648 second contained a race. It did not block SIGALRM between setting the
2649 timer, and suspending (a couple of lines later). If the interval was short
2650 and the sigsuspend() was delayed until after it had expired, the suspension
2651 never ended. On busy systems this could lead to processes getting stuck for
2654 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2655 process, before it forks any delivery processes. The open lookup caching
2656 mechanism meant that the open file or database connection was passed into
2657 the delivery process. The problem was that delivery processes always tidy
2658 up cached lookup data. This could cause a problem for the next delivery
2659 process started by the queue runner, because the external queue runner
2660 process does not know about the closure. So the next delivery process
2661 still has data in the lookup cache. In the case of a file lookup, there was
2662 no problem because closing a file descriptor in a subprocess doesn't affect
2663 the parent. However, if the lookup was caching a connection to a database,
2664 the connection was closed, and the second delivery process was likely to
2665 see errors such as "PGSQL: query failed: server closed the connection
2666 unexpectedly". The problem has been fixed by closing all cached lookups
2667 in a queue runner before running a delivery process.
2669 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2670 seem to have the "const" qualifier which it has on other OS. I've
2673 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2674 *of the same type* (client or server) with the same public name that an
2675 error should be diagnosed.
2677 9. When Exim looked up a host name for an IP address, but failed to find the
2678 original IP address when looking up the host name (a safety check), it
2679 output the message "<ip address> does not match any IP for NULL", which was
2680 confusing, to say the least. The bug was that the host name should have
2681 appeared instead of "NULL".
2683 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2684 user that is built into the binary, and the -C or -D options is used, root
2685 privilege is dropped before the configuration file is read. In addition,
2686 logging is switched to stderr instead of the normal log files. If the
2687 configuration then re-defines the Exim user, the unprivileged environment
2688 is probably not what is expected, so Exim logs a panic warning message (but
2691 However, if deliver_drop_privilege is set, the unprivileged state may well
2692 be exactly what is intended, so the warning has been cut out in that case,
2693 and Exim is allowed to try to write to its normal log files.
2699 1. smtp_return_error_details was not giving details for temporary sender
2700 or receiver verification errors.
2702 2. Diagnose a configuration error if two authenticators have the same public
2705 3. Exim used not to create the message log file for a message until the first
2706 delivery attempt. This could be confusing when incoming messages were held
2707 for policy or load reasons. The message log file is now created at the time
2708 the message is received, and an initial "Received" line is written to it.
2710 4. The automatically generated man page for command line options had a minor
2711 bug that caused no ill effects; however, a more serious problem was that
2712 the procedure for building the man page automatically didn't always
2713 operate. Consequently, release 4.20 contains an out-of-date version. This
2714 shouldn't happen again.
2716 5. When building Exim with embedded Perl support, the script that builds the
2717 Makefile was calling 'perl' to find its compile-time parameters, ignoring
2718 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
2720 6. The freeze_tell option was not being used for messages that were frozen on
2721 arrival, either by an ACL or by local_scan().
2723 7. Added the smtp_incomplete_transaction log selector.
2725 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
2726 was accepting AUTH without a new EHLO.
2728 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
2729 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
2732 10. Logging of TCP/IP connections (when configured) now happens in the main
2733 daemon process instead of the child process, so that the TCP/IP connection
2734 count is more accurate (but it can never be perfect).
2736 11. The use of "drop" in a nested ACL was not being handled correctly in the
2737 outer ACL. Now, if condition failure induced by the nested "drop" causes
2738 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
2739 or "require"), the connection is dropped.
2741 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
2742 that yield "discard" can now be used with an "accept" or a "discard" verb,
2743 but an error is generated for any others (because I can't see a useful way
2744 to define what should happen).
2746 13. When an ACL is read dynamically from a file (or anywhere else), the lines
2747 are now processed in the same way as lines in the Exim configuration file.
2748 In particular, continuation lines are supported.
2750 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
2752 15. Added -ti meaning -t -i.
2754 16. Check for letters, digits, hyphens, and dots in the names of dnslist
2755 domains, and warn by logging if others are found.
2757 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
2758 returned from GIFCONF when Exim is trying to find the list of interfaces on
2759 a host. The code in os.c has been modified to copy each ifreq to an aligned
2760 structure in all cases.
2762 Also, in some cases, the returned ifreq's were being copied to a 'struct
2763 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
2764 means the last couple of bytes of an IPv6 address could be chopped if the
2765 ifreq contained only a normal sockaddr (14 bytes storage).
2767 18. Named domain lists were not supported in the hosts_treat_as_local option.
2768 An entry such as +xxxx was not recognized, and was treated as a literal
2771 19. Ensure that header lines added by a DATA ACL are included in the reject log
2772 if the ACL subsequently rejects the message.
2774 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
2775 MD5 (which is deprecated).
2777 21. When testing a filter file using -bf, Exim was writing a message when it
2778 took the sender from a "From " line in the message, but it was not doing so
2779 when it took $return_path from a Return-Path: header line. It now does.
2781 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
2782 with a valid header line field name (a series of printing characters
2783 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
2785 23. Changed "disc" in the source to "disk" to conform to the documentation and
2786 the book and for uniformity.
2788 24. Ignore Sendmail's -Ooption=value command line item.
2790 25. When execve() failed while trying to run a command in a pipe transport,
2791 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
2792 could be confused with a return value of 69 from the command itself. This
2793 has been changed to 127, the value the shell returns if it is asked to run
2794 a non-existent command. The wording for the related log line suggests a
2795 non-existent command as the problem.
2797 26. If received_header_text expands to an empty string, do not add a Received:
2798 header line to the message. (Well, it adds a token one on the spool, but
2799 marks it "old" so that it doesn't get used or transmitted.)
2801 27. Installed eximstats 1.28 (addition of -nt option).
2803 28. There was no check for failure on the call to getsockname() in the daemon
2804 code. This can fail if there is a shortage of resources on the system, with
2805 ENOMEM, for example. A temporary error is now given on failure.
2807 29. Contrary to the C standard, it seems that in some environments, the
2808 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
2809 program. Exim now does this explicitly; it affects the formatting of
2810 timestamps using strftime().
2812 30. If exiqsumm was given junk data, it threw up some uninitialized variable
2813 complaints. I've now initialized all the variables, to avoid this.
2815 32. Header lines added by a system filter were not being "seen" during
2816 transport-time rewrites.
2818 33. The info_callback() function passed to OpenSSL is set up with type void
2819 (*)(SSL *, int, int), as described somewhere. However, when calling the
2820 function (actually a macro) that sets it up, the type void(*)() is
2821 expected. I've put in a cast to prevent warnings from picky compilers.
2823 34. If a DNS black list lookup found a CNAME record, but there were no A
2824 records associated with the domain it pointed at, Exim crashed.
2826 35. If a DNS black list lookup returned more than one A record, Exim ignored
2827 all but the first. It now scans all returned addresses if a particular IP
2828 value is being sought. In this situation, the contents of the
2829 $dnslist_value variable are a list of all the addresses, separated by a
2832 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
2833 to accept a host name and all its aliases if the forward lookup for any of
2834 them yielded the IP address of the incoming connection. Now it accepts only
2835 those names whose forward lookup yields the correct IP address. Any other
2836 names are discarded. This closes a loophole whereby a rogue DNS
2837 administrator could create reverse DNS records to break through a
2838 wildcarded host restriction in an ACL.
2840 37. If a user filter or a system filter that ran in a subprocess used any of
2841 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
2842 the wrong values were passed to the pipe command ($thisaddress had the
2843 value of $0, $0 had the value of $1, etc). This bug was introduced by
2844 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
2846 38. Improved the line breaking for long SMTP error messages from ACLs.
2847 Previously, if there was no break point between 40 and 75 characters, Exim
2848 left the rest of the message alone. Two changes have been made: (a) I've
2849 reduced the minimum length to 35 characters; (b) if it can't find a break
2850 point between 35 and 75 characters, it looks ahead and uses the first one
2851 that it finds. This may give the occasional overlong line, but at least the
2852 remaining text gets split now.
2854 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
2855 file descriptors might be low, and that setting 1000 would always raise it.
2856 It turns out that in some environments, the limit is already over 1000 and
2857 that lowering it causes trouble. So now Exim takes care not to decrease it.
2859 40. When delivering a message, the value of $return_path is set to $sender_
2860 address at the start of routing (routers may change the value). By an
2861 oversight, this default was not being set up when an address was tested by
2862 -bt or -bv, which affected the outcome if any router or filter referred to
2865 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
2866 log only when "message" or "log_message" are set. However, if one of the
2867 conditions was an address verification, or a call to a nested ACL, the
2868 messages generated by the underlying test were being passed through. This
2869 no longer happens. The underlying message is available in $acl_verify_
2870 message for both "message" and "log_message" expansions, so it can be
2871 passed through if needed.
2873 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
2874 new expansion $bh_ to give the encoded byte string without charset
2875 translation. Translation happens only if iconv() is available; HAVE_ICONV
2876 indicates this at build time. HEADERS_CHARSET gives the charset to
2877 translate to; headers_charset can change it in the configuration, and
2878 "headers charset" can change it in an individual filter file.
2880 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
2881 that creates RFC 2047 encoded "words" labels them as that charset instead
2882 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
2883 expansion operator; (ii) when Exim creates a From: line for a local
2884 message; (iii) when a header line is rewritten to include a "phrase" part.
2886 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
2887 buggy, causing it to skip the first lines of messages whose message ID
2888 ended in 'D'. This would not have bitten before Exim release 4.14, because
2889 message IDs were unlikely to end in 'D' before then. The effect was to have
2890 incorrect size information for certain domains.
2892 45. #include "config.h" was missing at the start of the crypt16.c module. This
2893 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
2896 46. If there was a timeout during a "random" callout check, Exim treated it as
2897 a failure of the random address, and carried on sending RSET and the real
2898 address. If the delay was just some slowness somewhere, the response to the
2899 original RCPT would be taken as a response to RSET and so on, causing
2900 mayhem of various kinds.
2902 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
2903 when I implemented it. It didn't allow for the fact that some option values
2904 may legitimatetly be negative (e.g. size_addition), and it didn't even do
2905 the right test for positive values.
2907 48. Domain names in DNS records are case-independent. Exim always looks them up
2908 in lower case. Some resolvers return domain names in exactly the case they
2909 appear in the zone file, that is, they may contain uppercase letters. Not
2910 all resolvers do this - some return always lower case. Exim was treating a
2911 change of case by a resolver as a change of domain, similar to a widening
2912 of a domain abbreviation. This triggered its re-routing code and so it was
2913 trying to route what was effectively the same domain again. This normally
2914 caused routing to fail (because the router wouldn't handle the domain
2915 twice). Now Exim checks for this case specially, and just changes the
2916 casing of the domain that it ultimately uses when it transmits the message
2919 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
2922 50. If a filter generated a file delivery with a non-absolute name (possible if
2923 no home directory exists for the router), the forbid_file option was not
2926 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
2927 the existing equality matching.
2929 52. Exim was using ints instead of ino_t variables in some places where it was
2930 dealing with inode numbers.
2932 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
2933 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
2934 called TMPDIR, and if it finds it is different, it changes its value.
2936 54. The smtp_printf() function is now made available to local_scan() so
2937 additional output lines can be written before returning. There is also an
2938 smtp_fflush() function to enable the detection of a dropped connection.
2939 The variables smtp_input and smtp_batched_input are exported to
2942 55. Changed the default runtime configuration: the message "Unknown user"
2943 has been removed from the ACL, and instead placed on the localuser router,
2944 using the cannot_route_message feature. This means that any verification
2945 failures that generate their own messages won't get overridden. Similarly,
2946 the "Unrouteable address" message that was in the ACL for unverifiable
2947 relay addresses has also been removed.
2949 56. Added hosts_avoid_esmtp to the smtp transport.
2951 57. The exicyclog script was not checking for the esoteric option
2952 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2953 will work only if exicyclog is run under the appropriate euid.
2955 58. Following a discussion on the list, the rules by which Exim recognises line
2956 endings on incoming messages have been changed. The -dropcr and drop_cr
2957 options are now no-ops, retained only for backwards compatibility. The
2958 following line terminators are recognized: LF CRLF CR. However, special
2959 processing applies to CR:
2961 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2962 nor a local message in the state where . is a terminator.
2964 (ii) If a bare CR is encountered in a header line, an extra space is added
2965 after the line terminator so as not to end the header. The reasoning
2966 behind this is that bare CRs in header lines are most likely either
2967 to be mistakes, or people trying to play silly games.
2969 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
2970 was being incorrectly given as 18 bytes larger than it should have been.
2971 This is a VOB (very old bug).
2973 60. This may never have affected anything current, but just in case it has:
2974 When the local host is found other than at the start of a list of hosts,
2975 the local host, those with the same MX, and any that follow, are discarded.
2976 When the list in question was part of a longer list of hosts, the following
2977 hosts (not currently being processed) were also being discarded. This no
2978 longer happens. I'm not sure if this situation could ever has previously
2981 61. Added the "/MX" feature to lists of hosts in the manualroute and query
2984 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
2985 header. This is something that is recommended in a new Internet Draft, and
2986 is something that is documented as being done by Sendmail. There are two
2987 possible values. For messages generated by the autoreply transport, Exim
2990 Auto-Submitted: auto-replied
2992 whereas for all other generated messages (e.g. bounces) it adds
2994 Auto-Submitted: auto-generated
2996 63. The "personal" condition in filters now includes a test for the
2997 Auto-Submitted: header. If it contains the string "auto-" the message it
2998 not considered personal.
3000 64. Added rcpt_include_affixes as a generic transport option.
3002 65. Added queue_only_override (default true).
3004 66. Added the syslog_duplication option.
3006 67. If what should have been the first header line of a message consisted of
3007 a space followed by a colon, Exim was mis-interpreting it as a header line.
3008 It isn't of course - it is syntactically invalid and should therefore be
3009 treated as the start of the message body. The misbehaviour could have
3010 caused a number of strange effects, including loss of data in subsequent
3011 header lines, and spool format errors.
3013 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3014 client host had authenticated. This control can now be exercised by an ACL
3015 for more flexibility.
3017 69. By default, callouts do not happen when testing with -bh. There is now a
3018 variant, -bhc, which does actually run the callout code, including
3019 consulting and updating the callout cache.
3021 70. Added support for saslauthd authentication, courtesy of Alexander
3024 71. If statvfs() failed on the spool or log directories while checking their
3025 size for availability, Exim confusingly gave the error "space shortage".
3026 Furthermore, in debugging mode it crashed with a floating point exception.
3027 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3028 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3029 serious problem, Exim now writes to the main and panic logs when this
3030 happens, with details of the failure. It then refuses to accept the
3031 incoming message, giving the message "spool directory problem" or "log
3032 directory problem" with a 421 code for SMTP messages.
3034 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3035 0, 1, and 2 exist, because some OS complain for execs without them (see
3036 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3037 descriptors. However, the code omitted to check that the open succeeded,
3038 causing mysterious errors if for some reason the permissions on /dev/null
3039 got screwed. Now Exim writes a message to the main and panic logs, and
3040 bombs out if it can't open /dev/null.
3042 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3043 interact so that it is all more flexible. It is supposed to remain
3044 backwards compatible. Also added extra_local_interfaces.
3046 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3047 to bomb out with an assertion failure - to the client this appears as a
3048 connection drop. This problem occurs in the part of the code that was taken
3049 from the Samba project. Fortunately, the assertion is in a very simple
3050 function, so I have fixed this by reproducing the function inline in the
3051 one place where it is called, and arranging for authentication to fail
3052 instead of killing the process with assert().
3054 75. The SPA client code was not working when the server requested OEM rather
3055 than Unicode encoding.
3057 76. Added code to make require_files with a specific uid setting more usable in
3058 the case where statting the file as root fails - usually a non-root-mounted
3059 NFS file system. When this happens and the failure is EACCES, Exim now
3060 forks a subprocess and does the per-uid checking as the relevant uid.
3062 77. Added process_log_path.
3064 78. If log_file_path was not explicitly set, a setting of check_log_space or
3065 check_log_inodes was ignored.
3067 79. If a space check for the spool or log partitions fails, the incident is now
3068 logged. Of course, in the latter case the data may get lost...
3070 80. Added the %p formatting code to string_format() so that it can be used to
3071 print addresses in debug_print(). Adjusted all the address printing in the
3072 debugging in store.c to use %p rather than %d.
3074 81. There was a concern that a line of code in smtp_in.c could overflow a
3075 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3076 initially expressed, the concern was not well-founded, because trailing
3077 spaces are removed early. However, if the trailing spaces were followed by
3078 a NULL, they did not get removed, so the overflow was possible. Two fixes
3081 (a) I re-wrote the offending code in a cleaner fashion.
3082 (b) If an incoming SMTP command contains a NULL character, it is rejected
3085 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3086 runs initgroups(), so that if the Exim user is in any additional groups,
3087 they will be used during message reception.
3093 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob