1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.154 2005/06/14 10:32:01 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.
121 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
122 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
124 TK/02 Fix ACL "control" statement not being available in MIME ACL.
126 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
128 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
129 to test Sieve filters that use "vacation".
131 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
132 that changes the way the GnuTLS parameters are stored in the cache file.
133 The new format can be generated externally. For backward compatibility,
134 if the data in the cache doesn't make sense, Exim assumes it has read an
135 old-format file, and it generates new data and writes a new file. This
136 means that you can't go back to an older release without removing the
139 PH/03 A redirect router that has both "unseen" and "one_time" set does not
140 work if there are any delivery delays because "one_time" forces the
141 parent to be marked "delivered", so its unseen clone is never tried
142 again. For this reason, Exim now forbids the simultaneous setting of
145 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
146 redirected to themselves ("homonym" addresses). Read the long ChangeLog
147 entry if you want to know the details. The fix, however, neglected to
148 consider the case when local delivery batching is involved. The test for
149 "previously delivered" was not happening when checking to see if an
150 address could be batched with a previous (undelivered) one; under
151 certain circumstances this could lead to multiple deliveries to the same
154 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
155 in its include files, and this causes problems building Exim.
157 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
158 header_syntax) but Exim was just ignoring anything given after a slash.
159 In particular, this caused confusion with an attempt to use "verify =
160 reverse_host_lookup/defer_ok". An error is now given when options are
161 supplied for verify items that do not have them. (Maybe reverse_host_
162 lookup should have a defer_ok option, but that's a different point.)
164 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
165 defined by RFC 821) to 2048, because there were problems with some AUTH
166 commands, and RFC 1869 says the size should be increased for extended
167 SMTP commands that take arguments.
169 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
172 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
173 "unknown" error; now it says that the functionality isn't in the binary.
175 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
176 an address' error message when a string expansion fails (syntax or
177 whatever). Otherwise the password may appear in the log. Following change
178 PH/42 below, there is no longer a chance of it appearing in a bounce
181 PH/11 Installed exipick version 20050225.0 from John Jetmore.
183 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
184 of its addresses was ever tried. (Bugzilla bug #2.)
186 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
187 the result incorrectly in the debug output. (It correctly added a newline
188 to what was transported.)
190 TF/01 Added $received_time.
192 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
193 commented out examples of how to interface to a virus scanner and to
194 SpamAssassin. Also added commented examples of av_scanner and
195 spamd_address settings.
197 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
198 and controls are allowed in which ACLs. There were a couple of minor
199 errors. Some of the entries in the conditions table (which is a table of
200 where they are NOT allowed) were getting very unwieldy; rewrote them as a
201 negation of where the condition IS allowed.
203 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
205 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
206 header file does not have a version number, so I've had to invent a new
207 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
208 API. The code is untested by me (my Linux distribution still has 0.3.2 of
209 radiusclient), but it was contributed by a Radius user.
211 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
212 files or directories, for OpenSSL.
214 PH/19 When an Exim process that is running as root has to create an Exim log
215 file, it does so in a subprocess that runs as exim:exim so as to get the
216 ownership right at creation (otherwise, other Exim processes might see
217 the file with the wrong ownership). There was no test for failure of this
218 fork() call, which would lead to the process getting stuck as it waited
219 for a non-existent subprocess. Forks do occasionally fail when resources
220 run out. I reviewed all the other calls to fork(); they all seem to check
223 PH/20 When checking for unexpected SMTP input at connect time (before writing
224 the banner), Exim was not dealing correctly with a non-positive return
225 from the read() function. If the client had disconnected by this time,
226 the result was a log entry for a synchronization error with an empty
227 string after "input=" when read() returned zero. If read() returned -1
228 (an event I could not check), uninitialized data bytes were printed.
229 There were reports of junk text (parts of files, etc) appearing after
232 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
234 PH/22 Added support for macro redefinition, and (re)definition in between
235 driver and ACL definitions.
237 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
238 forgetting to use the resulting value; it was using the unexpanded value.
240 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
241 hadn't been configured. The fix is from Juergen Kreileder, who
242 understands it better than I do:
244 "Here's what I see happening with three configured cyrus_sasl
245 authenticators configured (plain, login, cram-md5):
247 On startup auth_cyrus_sasl_init() gets called for each of these.
248 This means three calls to sasl_listmech() without a specified mech_list.
249 => SASL tests which mechs of all available mechs actually work
250 => three warnings about OTP not working
251 => the returned list contains: plain, login, cram-md5, digest-md5, ...
253 With the patch, sasl_listmech() also gets called three times. But now
254 SASL's mech_list option is set to the server_mech specified in the the
255 authenticator. Or in other words, the answer from sasl_listmech()
256 gets limited to just the mech you're testing for (which is different
258 => the return list contains just 'plain' or 'login', 'cram-md5' or
259 nothing depending on the value of ob->server_mech.
261 I've just tested the patch: Authentication still works fine,
262 unavailable mechs specified in the exim configuration are still
263 caught, and the auth.log warnings about OTP are gone."
265 PH/25 When debugging is enabled, the contents of the command line are added
266 to the debugging output, even when log_selector=+arguments is not
269 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
270 answer is "GNU", and only if the return is "GNU/something" is the answer
273 PH/27 $acl_verify_message is now set immediately after the failure of a
274 verification in an ACL, and so is available in subsequent modifiers. In
275 particular, the message can be preserved by coding like this:
277 warn !verify = sender
278 set acl_m0 = $acl_verify_message
280 Previously, $acl_verify_message was set only while expanding "message"
281 and "log_message" when a very denied access.
283 PH/28 Modified OS/os.c-Linux with
285 -#ifndef OS_LOAD_AVERAGE
286 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
288 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
289 nomenclature these days.)
291 PH/29 Installed patch from the Sieve maintainer that adds the options
292 sieve_useraddress and sieve_subaddress to the redirect router.
294 PH/30 In these circumstances:
295 . Two addresses routed to the same list of hosts;
296 . First host does not offer TLS;
297 . First host accepts first address;
298 . First host gives temporary error to second address;
299 . Second host offers TLS and a TLS session is established;
300 . Second host accepts second address.
301 Exim incorrectly logged both deliveries with the TLS parameters (cipher
302 and peerdn, if requested) that were in fact used only for the second
305 PH/31 When doing a callout as part of verifying an address, Exim was not paying
306 attention to any local part prefix or suffix that was matched by the
307 router that accepted the address. It now behaves in the same way as it
308 does for delivery: the affixes are removed from the local part unless
309 rcpt_include_affixes is set on the transport.
311 PH/32 Add the sender address, as F=<...>, to the log line when logging a
312 timeout during the DATA phase of an incoming message.
314 PH/33 Sieve envelope tests were broken for match types other than :is. I have
315 applied a patch sanctioned by the Sieve maintainer.
317 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
318 the uid or gid is negative. A case of a negative gid caused this to be
319 noticed. The fix allows for either to be negative.
321 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
322 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
325 PH/36 The change PH/12 above was broken. Fixed it.
327 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
328 the grounds that routing the same address twice would always produce the
329 same answer. This might have been true once, but it is certainly no
330 longer true now. Routing a child address may depend on the previous
331 routing that produced that child. Some complicated redirection strategies
332 went wrong when messages had multiple recipients, and made Exim's
333 behaviour dependent on the order in which the addresses were given.
335 I have moved the duplicate checking until after the routing is complete.
336 Exim scans the addresses that are assigned to local and remote
337 transports, and removes any duplicates. This means that more work will be
338 done, as duplicates will always all be routed, but duplicates are
339 presumably rare, so I don't expect this is of any significance.
341 For deliveries to pipes, files, and autoreplies, the duplicate checking
342 still happens during the routing process, since they are not going to be
345 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
346 It corrects a timeout issue with spamd. This is Ian's comment: "The
347 background is that sometimes spamd either never reads data from a
348 connection it has accepted, or it never writes response data. The exiscan
349 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
350 blindly assumes that writes won't block so it may never time out."
352 PH/39 Allow G after quota size as well as K and M.
354 PH/40 The value set for $authenticated_id in an authenticator may not contain
355 binary zeroes or newlines because the value is written to log lines and
356 to spool files. There was no check on this. Now the value is run through
357 the string_printing() function so that such characters are converted to
358 printable escape sequences.
360 PH/41 $message_linecount is a new variable that contains the total number of
361 lines in the message. Compare $body_linecount, which is the count for the
364 PH/42 Exim no longer gives details of delivery errors for specific addresses in
365 bounce and delay warning messages, except in certain special cases, which
368 (a) An SMTP error message from a remote host;
369 (b) A message specified in a :fail: redirection;
370 (c) A message specified in a "fail" command in a system filter;
371 (d) A message specified in a FAIL return from the queryprogram router;
372 (e) A message specified by the cannot_route_message router option.
374 In these cases only, Exim does include the error details in bounce and
375 warning messages. There are also a few cases where bland messages such
376 as "unrouteable address" or "local delivery error" are given.
378 PH/43 $value is now also set for the "else" part of a ${run expansion.
380 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
381 being worked on, but at least Exim now implements the latest version to
384 PH/45 In a pipe transport, although a timeout while waiting for the pipe
385 process to complete was treated as a delivery failure, a timeout while
386 writing the message to the pipe was logged, but erroneously treated as a
387 successful delivery. Such timeouts include transport filter timeouts. For
388 consistency with the overall process timeout, these timeouts are now
389 treated as errors, giving rise to delivery failures by default. However,
390 there is now a new Boolean option for the pipe transport called
391 timeout_defer, which, if set TRUE, converts the failures into defers for
392 both kinds of timeout. A transport filter timeout is now identified in
395 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
396 systems where "make" and "gmake" are different, calling "gmake" at top
397 level broke things. I've arranged for the value of $(MAKE) to be passed
398 from the Makefile to this script so that it can call the same version of
402 A note about Exim versions 4.44 and 4.50
403 ----------------------------------------
405 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
406 changes of various kinds. As a consequence, a big documentation update was
407 needed. This delayed the release for rather longer than seemed good, especially
408 in the light of a couple of (minor) security issues. Therefore, the changes
409 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
410 release. So 4.44 and 4.50 are in effect two different branches that both start
413 I have left the 4.50 change log unchanged; it contains all the changes since
414 4.43. The change log for 4.44 is below; many of its items are identical to
415 those for 4.50. This seems to be the most sensible way to preserve the
416 historical information.
422 1. Minor wording change to the doc/README.SIEVE file.
424 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
425 computation of the current number of files was incorrect.
427 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
428 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
431 4. Give more explanation in the error message when the command for a transport
432 filter fails to execute.
434 5. There are several places where Exim runs a non-Exim command in a
435 subprocess. The SIGUSR1 signal should be disabled for these processes. This
436 was being done only for the command run by the queryprogram router. It is
437 now done for all such subprocesses. The other cases are: ${run, transport
438 filters, and the commands run by the lmtp and pipe transports.
440 6. Added CONFIGURE_GROUP build-time option.
442 7. Some older OS have a limit of 256 on the maximum number of file
443 descriptors. Exim was using setrlimit() to set 1000 as a large value
444 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
445 systems. I've change it so that if it can't get 1000, it tries for 256.
447 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
448 was an oversight, and furthermore, ever since the addition of extra
449 controls (e.g. 4.43/32), the checks on when to allow different forms of
450 "control" were broken. There should now be diagnostics for all cases when a
451 control that does not make sense is encountered.
453 9. Added the /retain_sender option to "control=submission".
455 10. $recipients is now available in the predata ACL (oversight).
457 11. Tidy the search cache before the fork to do a delivery from a message
458 received from the command line. Otherwise the child will trigger a lookup
459 failure and thereby defer the delivery if it tries to use (for example) a
460 cached ldap connection that the parent has called unbind on.
462 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
463 of $address_data from the recipient verification was clobbered by the
466 13. The value of address_data from a sender verification is now available in
467 $sender_address_data in subsequent conditions in the ACL statement.
469 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
471 15. Added a new option "connect=<time>" to callout options, to set a different
474 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
475 was its contents. (It was OK if the option was not defined at all.)
477 17. A "Completed" log line is now written for messages that are removed from
478 the spool by the -Mrm option.
480 18. New variables $sender_verify_failure and $recipient_verify_failure contain
481 information about exactly what failed.
483 19. Added -dd to debug only the daemon process.
485 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
486 handles timeouts, both on the server side and network timeouts. Renamed the
487 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
489 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
490 It is now set to "smtps".
492 22. $host_address is now set to the target address during the checking of
495 23. When checking ignore_target_hosts for an ipliteral router, no host name was
496 being passed; this would have caused $sender_host_name to have been used if
497 matching the list had actually called for a host name (not very likely,
498 since this list is usually IP addresses). A host name is now passed as
501 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
502 code that specifies a non-restarting handler (typically sigaction() in
503 modern systems) in an attempt to fix a rare and obscure crash bug.
505 25. Narrowed the window for a race in the daemon that could cause it to ignore
506 SIGCHLD signals. This is not a major problem, because they are used only to
507 wake it up if nothing else does.
509 26. A malformed maildirsize file could cause Exim to calculate negative values
510 for the mailbox size or file count. Odd effects could occur as a result.
511 The maildirsize information is now recalculated if the size or filecount
514 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
515 support for a long time. Removed HAVE_SYS_VFS_H.
517 28. Installed the latest version of exipick from John Jetmore.
519 29. In an address list, if the pattern was not a regular expression, an empty
520 subject address (from a bounce message) matched only if the pattern was an
521 empty string. Non-empty patterns were not even tested. This was the wrong
522 because it is perfectly reasonable to use an empty address as part of a
523 database query. An empty address is now tested by patterns that are
524 lookups. However, all the other forms of pattern expect the subject to
525 contain a local part and a domain, and therefore, for them, an empty
526 address still always fails if the pattern is not itself empty.
528 30. Exim went into a mad DNS loop when attempting to do a callout where the
529 host was specified on an smtp transport, and looking it up yielded more
532 31. Re-factored the code for checking spool and log partition space into a
533 function that finds that data and another that does the check. The former
534 is then used to implement four new variables: $spool_space, $log_space,
535 $spool_inodes, and $log_inodes.
537 32. The RFC2047 encoding function was originally intended for short strings
538 such as real names; it was not keeping to the 75-character limit for
539 encoded words that the RFC imposes. It now respects the limit, and
540 generates multiple encoded words if necessary. To be on the safe side, I
541 have increased the buffer size for the ${rfc2047: expansion operator from
544 33. It is now permitted to omit both strings after an "if" condition; if the
545 condition is true, the result is "true". As before, when the second string
546 is omitted, a false condition yields an empty string. This makes it less
547 cumbersome to write custom ACL and router conditions.
549 34. Failure to deliver a bounce message always caused it to be frozen, even if
550 there was an errors_to setting on the router. The errors_to setting is now
553 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
554 canonical form (fully expanded) before being placed in
555 $sender_host_address.
557 36. The table in the code that translates DNS record types into text (T_A to
558 "A" for instance) was missing entries for NS and CNAME. It is just possible
559 that this could have caused confusion if both these types were looked up
560 for the same domain, because the text type is used as part of Exim's
561 per-process caching. But the chance of anyone hitting this buglet seems
564 37. The dnsdb lookup has been extended in a number of ways.
566 (1) There is a new type, "zns", which walks up the domain tree until it
567 finds some nameserver records. It should be used with care.
569 (2) There is a new type, "mxh", which is like "mx" except that it returns
570 just the host names, not the priorities.
572 (3) It is now possible to give a list of domains (or IP addresses) to be
573 looked up. The behaviour when one of the lookups defers can be
574 controlled by a keyword.
576 (4) It is now possible to specify the separator character for use when
577 multiple records are returned.
579 38. The dnslists ACL condition has been extended: it is now possible to supply
580 a list of IP addresses and/or domains to be looked up in a particular DNS
583 39. Added log_selector=+queue_time_overall.
585 40. When running the queue in the test harness, wait just a tad after forking a
586 delivery process, to get repeatability of debugging output.
588 41. Include certificate and key file names in error message when GnuTLS fails
589 to set them up, because the GnuTLS error message doesn't include the name
590 of the failing file when there is a problem reading it.
592 42. Allow both -bf and -bF in the same test run.
594 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
596 44. The "Exiscan patch" is now merged into the mainline Exim source.
598 45. Sometimes the final signoff response after QUIT could fail to get
599 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
600 < 0 before doing a fflush(). This bug looks as though it goes back to the
601 introduction of TLS in release 3.20, but "sometimes" must have been rare
602 because the tests only now provoked it.
604 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
605 (this can affect the format of dates).
607 47. exim_tidydb, when checking for the continued existence of a message for
608 which it has found a message-specific retry record, was not finding
609 messages that were in split spool directories. Consequently, it was
610 deleting retry records that should have stayed in existence.
612 48. Steve fixed some bugs in eximstats.
614 49. The SPA authentication driver was not abandoning authentication and moving
615 on to the next authenticator when an expansion was forced to fail,
616 contradicting the general specification for all authenticators. Instead it
617 was generating a temporary error. It now behaves as specified.
619 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
620 (the order specifies the preference for clients). The order is now AES256,
621 AES128, 3DES, ARCFOUR128.
623 51. Small patch to Sieve code - explicitly set From: when generating an
626 52. Exim crashed if a remote delivery caused a very long error message to be
627 recorded - for instance if somebody sent an entire SpamAssassin report back
628 as a large number of 550 error lines. This bug was coincidentally fixed by
629 increasing the size of one of Exim's internal buffers (big_buffer) that
630 happened as part of the Exiscan merge. However, to be on the safe side, I
631 have made the code more robust (and fixed the comments that describe what
634 53. Now that there can be additional text after "Completed" in log lines (if
635 the queue_time_overall log selector is set), a one-byte patch to exigrep
636 was needed to allow it to recognize "Completed" as not the last thing in
639 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
640 patch that reportedly fixes this has been added. I am not expert enough to
641 create a test for it. This is what the patch creator wrote:
643 "I found a little strange behaviour of ldap code when working with
644 Windows 2003 AD Domain, where users was placed in more than one
645 Organization Units. When I tried to give exim partial DN, the exit code
646 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
647 But simultaneously result of request was absolutely normal ldap result,
648 so I produce this patch..."
650 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
651 REFERENCE, so I have modified the code to exclude the patch when that macro
654 55. Some experimental protocols are using DNS PTR records for new purposes. The
655 keys for these records are domain names, not reversed IP addresses. The
656 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
657 leaves it alone. Component reversal etc. now happens only for IP addresses.
660 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
662 57. Double the size of the debug message buffer (to 2048) so that more of very
663 long debug lines gets shown.
665 58. The exicyclog utility now does better if the number of log files to keep
666 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
668 59. Two changes related to the smtp_active_hostname option:
670 (1) $smtp_active_hostname is now available as a variable.
671 (2) The default for smtp_banner uses $smtp_active_hostname instead
672 of $primary_hostname.
674 60. The host_aton() function is supposed to be passed a string that is known
675 to be a valid IP address. However, in the case of IPv6 addresses, it was
676 not checking this. This is a hostage to fortune. Exim now panics and dies
677 if the condition is not met. A case was found where this could be provoked
678 from a dnsdb PTR lookup with an IPv6 address that had more than 8
679 components; fortuitously, this particular loophole had already been fixed
680 by change 4.50/55 above.
682 If there are any other similar loopholes, the new check in host_aton()
683 itself should stop them being exploited. The report I received stated that
684 data on the command line could provoke the exploit when Exim was running as
685 exim, but did not say which command line option was involved. All I could
686 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
690 61. There was a buffer overflow vulnerability in the SPA authentication code
691 (which came originally from the Samba project). I have added a test to the
692 spa_base64_to_bits() function which I hope fixes it.
695 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
696 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
698 63. The daemon start-up calls getloadavg() while still root for those OS that
699 need the first call to be done as root, but it missed one case: when
700 deliver_queue_load_max is set with deliver_drop_privilege. This is
701 necessary for the benefit of the queue runner, because there is no re-exec
702 when deliver_drop_privilege is set.
704 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
707 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
708 was in use, was not putting the data itself into the right store pool;
709 consequently, it could be overwritten for a subsequent message in the same
710 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
713 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
715 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
716 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
717 the function were treating the return as a boolean value, which happened to
718 work because 0=false and not-0=true, but is not correct code.
720 68. The host_aton() function was not handling scoped IPv6 addresses (those
721 with, for example, "%eth0" on the end) correctly.
723 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
724 negated items (that is, ~something) in unsigned ints. Some compilers
725 apparently mutter when there is no cast.
727 70. If an address verification called from an ACL failed, and did not produce a
728 user-specific message (i.e. there was only a "system" message), nothing was
729 put in $acl_verify_message. In this situation, it now puts the system
732 71. Change 4.23/11 added synchronization checking at the start of an SMTP
733 session; change 4.31/43 added the unwanted input to the log line - except
734 that it did not do this in the start of session case. It now does.
736 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
737 This is wrong and can cause the other end to generate a synchronization
738 error if it is another Exim or anything else that does the synchronization
739 check. A QUIT command is no longer sent after a timeout.
741 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
744 74. The defer_ok option of callout verification was not working if it was used
745 when verifying addresses in header lines, that is, for this case:
747 verify = header_sender/callout=defer_ok
749 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
750 those file descriptors could be used for SMTP connections. If anything
751 wrote to stderr (the example that came up was "warn" in embedded Perl), it
752 could be sent to the SMTP client, causing chaos. The daemon now opens
753 stdin, stdout, and stderr to /dev/null when it puts itself into the
756 76. Arrange for output from Perl's "warn" command to be written to Exim's main
757 log by default. The user can override this with suitable Perl magic.
759 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
760 the log message when discard triggers, was not working for the DATA ACL or
761 for the non-SMTP ACL.
763 78. Error message wording change in sieve.c.
765 79. If smtp_accept_max_per_host was set, the number of connections could be
766 restricted to fewer than expected, because the daemon was trying to set up
767 a new connection before checking whether the processes handling previous
768 connections had finished. The check for completed processes is now done
769 earlier. On busy systems, this bug wouldn't be noticed because something
770 else would have woken the daemon, and it would have reaped the completed
773 80. If a message was submitted locally by a user whose login name contained one
774 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
775 It caused a spool format error. I have fixed the spool reading code. A
776 related problem was that the "from" clause in the Received: line became
777 illegal because of the space(s). It is now covered by ${quote_local_part.
779 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
782 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
784 83. Patch from Sieve maintainer to fix unterminated string problem in
787 84. Some minor changes to the Linux configuration files to help with other
788 OS variants using glibc.
790 85. One more patch for Sieve to update vacation handling to latest spec.
793 ----------------------------------------------------
794 See the note above about the 4.44 and 4.50 releases.
795 ----------------------------------------------------
801 1. Change 4.43/35 introduced a bug that caused file counts to be
802 incorrectly computed when quota_filecount was set in an appendfile
805 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
806 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
809 3. Give more explanation in the error message when the command for a transport
810 filter fails to execute.
812 4. There are several places where Exim runs a non-Exim command in a
813 subprocess. The SIGUSR1 signal should be disabled for these processes. This
814 was being done only for the command run by the queryprogram router. It is
815 now done for all such subprocesses. The other cases are: ${run, transport
816 filters, and the commands run by the lmtp and pipe transports.
818 5. Some older OS have a limit of 256 on the maximum number of file
819 descriptors. Exim was using setrlimit() to set 1000 as a large value
820 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
821 systems. I've change it so that if it can't get 1000, it tries for 256.
823 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
824 was an oversight, and furthermore, ever since the addition of extra
825 controls (e.g. 4.43/32), the checks on when to allow different forms of
826 "control" were broken. There should now be diagnostics for all cases when a
827 control that does not make sense is encountered.
829 7. $recipients is now available in the predata ACL (oversight).
831 8. Tidy the search cache before the fork to do a delivery from a message
832 received from the command line. Otherwise the child will trigger a lookup
833 failure and thereby defer the delivery if it tries to use (for example) a
834 cached ldap connection that the parent has called unbind on.
836 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
837 of $address_data from the recipient verification was clobbered by the
840 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
841 was its contents. (It was OK if the option was not defined at all.)
843 11. A "Completed" log line is now written for messages that are removed from
844 the spool by the -Mrm option.
846 12. $host_address is now set to the target address during the checking of
849 13. When checking ignore_target_hosts for an ipliteral router, no host name was
850 being passed; this would have caused $sender_host_name to have been used if
851 matching the list had actually called for a host name (not very likely,
852 since this list is usually IP addresses). A host name is now passed as
855 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
856 code that specifies a non-restarting handler (typically sigaction() in
857 modern systems) in an attempt to fix a rare and obscure crash bug.
859 15. Narrowed the window for a race in the daemon that could cause it to ignore
860 SIGCHLD signals. This is not a major problem, because they are used only to
861 wake it up if nothing else does.
863 16. A malformed maildirsize file could cause Exim to calculate negative values
864 for the mailbox size or file count. Odd effects could occur as a result.
865 The maildirsize information is now recalculated if the size or filecount
868 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
869 support for a long time. Removed HAVE_SYS_VFS_H.
871 18. Updated exipick to current release from John Jetmore.
873 19. Allow an empty sender to be matched against a lookup in an address list.
874 Previously the only cases considered were a regular expression, or an
877 20. Exim went into a mad DNS lookup loop when doing a callout where the
878 host was specified on the transport, if the DNS lookup yielded more than
881 21. The RFC2047 encoding function was originally intended for short strings
882 such as real names; it was not keeping to the 75-character limit for
883 encoded words that the RFC imposes. It now respects the limit, and
884 generates multiple encoded words if necessary. To be on the safe side, I
885 have increased the buffer size for the ${rfc2047: expansion operator from
888 22. Failure to deliver a bounce message always caused it to be frozen, even if
889 there was an errors_to setting on the router. The errors_to setting is now
892 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
893 canonical form (fully expanded) before being placed in
894 $sender_host_address.
896 24. Updated eximstats to version 1.33
898 25. Include certificate and key file names in error message when GnuTLS fails
899 to set them up, because the GnuTLS error message doesn't include the name
900 of the failing file when there is a problem reading it.
902 26. Expand error message when OpenSSL has problems setting up cert/key files.
905 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
906 (this can affect the format of dates).
908 28. exim_tidydb, when checking for the continued existence of a message for
909 which it has found a message-specific retry record, was not finding
910 messages that were in split spool directories. Consequently, it was
911 deleting retry records that should have stayed in existence.
913 29. eximstats updated to version 1.35
914 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
915 1.35 - bugfix such that pie charts by volume are generated correctly
917 30. The SPA authentication driver was not abandoning authentication and moving
918 on to the next authenticator when an expansion was forced to fail,
919 contradicting the general specification for all authenticators. Instead it
920 was generating a temporary error. It now behaves as specified.
922 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
923 (the order specifies the preference for clients). The order is now AES256,
924 AES128, 3DES, ARCFOUR128.
926 31. Small patch to Sieve code - explicitly set From: when generating an
929 32. Exim crashed if a remote delivery caused a very long error message to be
930 recorded - for instance if somebody sent an entire SpamAssassin report back
931 as a large number of 550 error lines. This bug was coincidentally fixed by
932 increasing the size of one of Exim's internal buffers (big_buffer) that
933 happened as part of the Exiscan merge. However, to be on the safe side, I
934 have made the code more robust (and fixed the comments that describe what
937 33. Some experimental protocols are using DNS PTR records for new purposes. The
938 keys for these records are domain names, not reversed IP addresses. The
939 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
940 leaves it alone. Component reversal etc. now happens only for IP addresses.
943 34. The host_aton() function is supposed to be passed a string that is known
944 to be a valid IP address. However, in the case of IPv6 addresses, it was
945 not checking this. This is a hostage to fortune. Exim now panics and dies
946 if the condition is not met. A case was found where this could be provoked
947 from a dnsdb PTR lookup with an IPv6 address that had more than 8
948 components; fortuitously, this particular loophole had already been fixed
949 by change 4.50/55 or 4.44/33 above.
951 If there are any other similar loopholes, the new check in host_aton()
952 itself should stop them being exploited. The report I received stated that
953 data on the command line could provoke the exploit when Exim was running as
954 exim, but did not say which command line option was involved. All I could
955 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
959 35. There was a buffer overflow vulnerability in the SPA authentication code
960 (which came originally from the Samba project). I have added a test to the
961 spa_base64_to_bits() function which I hope fixes it.
964 36. The daemon start-up calls getloadavg() while still root for those OS that
965 need the first call to be done as root, but it missed one case: when
966 deliver_queue_load_max is set with deliver_drop_privilege. This is
967 necessary for the benefit of the queue runner, because there is no re-exec
968 when deliver_drop_privilege is set.
970 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
971 was in use, was not putting the data itself into the right store pool;
972 consequently, it could be overwritten for a subsequent message in the same
973 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
976 38. Sometimes the final signoff response after QUIT could fail to get
977 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
978 < 0 before doing a fflush(). This bug looks as though it goes back to the
979 introduction of TLS in release 3.20, but "sometimes" must have been rare
980 because the tests only now provoked it.
986 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
987 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
988 Now it returns an integer. A number of calls were still expecting a T/F
989 return. Fortuitously, in all cases, the tests worked in OK situations,
990 which is the norm. However, things would have gone wrong on any write
991 failures on the smtp file descriptor. This function is used when sending
992 messages over SMTP and also when doing verify callouts.
994 2. When Exim is called to do synchronous delivery of a locally submitted
995 message (the -odf or -odi options), it no longer closes stderr before doing
998 3. Implemented the mua_wrapper option.
1000 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1002 5. Implemented the functions header_remove(), header_testname(),
1003 header_add_at_position(), and receive_remove_recipient(), and exported them
1006 6. If an ACL "warn" statement specified the addition of headers, Exim already
1007 inserted X-ACL-Warn: at the start if there was no header name. However, it
1008 was not making this test for the second and subsequent header lines if
1009 there were newlines in the string. This meant that an invalid header could
1010 be inserted if Exim was badly configured.
1012 7. Allow an ACL "warn" statement to add header lines at the start or after all
1013 the Received: headers, as well as at the end.
1015 8. Added the rcpt_4xx retry error code.
1017 9. Added postmaster_mailfrom=xxx to callout verification option.
1019 10. Added mailfrom=xxxx to the callout verification option, for verify=
1022 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1023 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1025 12. Inserted some casts to stop certain compilers warning when using pointer
1026 differences as field lengths or precisions in printf-type calls (mostly
1027 affecting debugging statements).
1029 13. Added optional readline() support for -be (dynamically loaded).
1031 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1032 same clock tick as a message's arrival, so that its received time was the
1033 same as the "first fail" time on the retry record, and that message
1034 remained on the queue past the ultimate address timeout, every queue runner
1035 would try a delivery (because it was past the ultimate address timeout) but
1036 after another failure, the ultimate address timeout, which should have then
1037 bounced the address, did not kick in. This was a "< instead of <=" error;
1038 in most cases the first failure would have been in the next clock tick
1039 after the received time, and all would be well.
1041 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1042 being recognized when the domain list was tested by the match_domain
1043 condition in an expansion string.
1045 16. Added the ${str2b64: operator.
1047 17. Exim was always calling setrlimit() to set a large limit for the number of
1048 processes, without checking whether the existing limit was already
1049 adequate. (It did check for the limit on file descriptors.) Furthermore,
1050 errors from getrlimit() and setrlimit() were being ignored. Now they are
1051 logged to the main and panic logs, but Exim does carry on, to try to do its
1052 job under whatever limits there are.
1054 18. Imported PCRE 5.0.
1056 19. Trivial typo in log message " temporarily refused connection" (the leading
1059 20. If the log selector return_path_on_delivery was set and an address was
1060 redirected to /dev/null, the delivery process crashed because it assumed
1061 that a return path would always be set for a "successful" delivery. In this
1062 case, the whole delivery is bypassed as an optimization, and therefore no
1065 21. Internal re-arrangement: the function for sending a challenge and reading
1066 a response while authentication was assuming a zero-terminated challenge
1067 string. It's now changed to take a pointer and a length, to allow for
1068 binary data in such strings.
1070 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1072 23. Exim was not respecting finduser_retries when seeking the login of the
1073 uid under which it was called; it was always trying 10 times. (The default
1074 setting of finduser_retries is zero.) Also, it was sleeping after the final
1075 failure, which is pointless.
1077 24. Implemented tls_on_connect_ports.
1079 25. Implemented acl_smtp_predata.
1081 26. If the domain in control=submission is set empty, Exim assumes that the
1082 authenticated id is a complete email address when it generates From: or
1083 Sender: header lines.
1085 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1086 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1087 chown and chgrp in /bin and hostname in /usr/bin.
1089 28. Exim was keeping the "process log" file open after each use, just as it
1090 does for the main log. This opens the possibility of it remaining open for
1091 long periods when the USR1 signal hits a daemon. Occasional processlog
1092 errors were reported, that could have been caused by this. Anyway, it seems
1093 much more sensible not to leave this file open at all, so that is what now
1096 29. The long-running daemon process does not normally write to the log once it
1097 has entered its main loop, and it closes the log before doing so. This is
1098 so that log files can straightforwardly be renamed and moved. However,
1099 there are a couple of unusual error situations where the daemon does write
1100 log entries, and I had neglected to close the log afterwards.
1102 30. The text of an SMTP error response that was received during a remote
1103 delivery was being truncated at 512 bytes. This is too short for some of
1104 the long messages that one sometimes sees. I've increased the limit to
1107 31. It is now possible to make retry rules that apply only when a message has a
1108 specific sender, in particular, an empty sender.
1110 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1111 it possible to be selective about when SMTP synchronization is enforced.
1113 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1115 32. Implemented hosts_connection_nolog.
1117 33. Added an ACL for QUIT.
1119 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1122 35. Added mailbox_size and mailbox_filecount to appendfile.
1124 36. Added control = no_multiline_responses to ACLs.
1126 37. There was a bug in the logic of the code that waits for the clock to tick
1127 in the case where the clock went backwards by a substantial amount such
1128 that the microsecond fraction of "now" was more than the microsecond
1129 fraction of "then" (but the whole seconds number was less).
1131 38. Added support for the libradius Radius client library this is found on
1132 FreeBSD (previously only the radiusclient library was supported).
1138 1. When certain lookups returned multiple values in the form name=value, the
1139 quoting of the values was not always being done properly. Specifically:
1140 (a) If the value started with a double quote, but contained no whitespace,
1142 (b) If the value contained whitespace other than a space character (i.e.
1143 tabs or newlines or carriage returns) it was not quoted.
1144 This fix has been applied to the mysql and pgsql lookups by writing a
1145 separate quoting function and calling it from the lookup code. The fix
1146 should probably also be applied to nisplus, ibase and oracle lookups, but
1147 since I cannot test any of those, I have not disturbed their existing code.
1149 2. A hit in the callout cache for a specific address caused a log line with no
1150 reason for rejecting RCPT. Now it says "Previous (cached) callout
1151 verification failure".
1153 3. There was an off-by-one bug in the queryprogram router. An over-long
1154 return line was truncated at 256 instead of 255 characters, thereby
1155 overflowing its buffer with the terminating zero. As well as fixing this, I
1156 have increased the buffer size to 1024 (and made a note to document this).
1158 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1159 when Exim is waiting for an SMTP response from a remote server, Exim
1160 restarts its select() call on the socket, thereby resetting its timeout.
1161 This is not a problem when such interrupts are rare. Somebody set up a cron
1162 job to run exiwhat every 2 minutes, which is less than the normal select()
1163 timeout (5 or 10 minutes). This meant that the select() timeout never
1164 kicked in because it was always reset. I have fixed this by comparing the
1165 time when an interrupt arrives with the time at the start of the first call
1166 to select(). If more time than the timeout has elapsed, the interrupt is
1167 treated as a timeout.
1169 5. Some internal re-factoring in preparation for the addition of Sieve
1170 extensions (by MH). In particular, the "personal" test is moved to a
1171 separate function, and given an option for scanning Cc: and Bcc: (which is
1172 not set for Exim filters).
1174 6. When Exim created an email address using the login of the caller as the
1175 local part (e.g. when creating a From: or Sender: header line), it was not
1176 quoting the local part when it contained special characters such as @.
1178 7. Installed new OpenBSD configuration files.
1180 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1181 try to make them clearer.
1183 9. Callout options, other than the timeout value, were being ignored when
1184 verifying sender addresses in header lines. For example, when using
1186 verify = header_sender/callout=no_cache
1188 the cache was (incorrectly) being used.
1190 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1191 only the Cygwin environment.
1193 11. When return_path_on_delivery was set as a log selector, if different remote
1194 addresses in the same message used different return paths and parallel
1195 remote delivery occurred, the wrong values would sometimes be logged.
1196 (Whenever a remote delivery process finished, the return path value from
1197 the most recently started remote delivery process was logged.)
1199 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1200 lines when AUTH and/or TLS are in use. This is the "received protocol"
1201 field. Exim used to use "asmtp" for authenticated SMTP, without any
1202 indication (in the protocol name) for TLS use. Now it follows the RFC and
1203 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1204 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1205 names appear in log lines as well as in Received: header lines.
1207 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1208 extensions, and comparison tests, and to fix some bugs.
1210 14. Changes to the "personal" filter test:
1212 (1) The test was buggy in that it was just doing the equivalent of
1213 "contains" tests on header lines. For example, if a user's address was
1214 anne@some.where, the "personal" test would incorrectly be true for
1216 To: susanne@some.where
1218 This test is now done by extracting each address from the header in turn,
1219 and checking the entire address. Other tests that are part of "personal"
1220 are now done using regular expressions (for example, to check local parts
1221 of addresses in From: header lines).
1223 (2) The list of non-personal local parts in From: addresses has been
1224 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1225 taken from the Sieve specification recommendations.
1227 (3) If the message contains any header line starting with "List-" it is
1228 treated as non-personal.
1230 (4) The test for "circular" in the Subject: header line has been removed
1231 because it now seems ill-conceived.
1233 15. Minor typos in src/EDITME comments corrected.
1235 16. Installed latest exipick from John Jetmore.
1237 17. If headers_add on a router specified a text string that was too long for
1238 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1239 of string_sprintf() is now avoided.
1241 18. $message_body_size was not set (it was always zero) when running the DATA
1242 ACL and the local_scan() function.
1244 19. For the "mail" command in an Exim filter, no default was being set for
1245 the once_repeat time, causing a random time value to be used if "once" was
1246 specified. (If the value happened to be <= 0, no repeat happened.) The
1247 default is now 0s, meaning "never repeat". The "vacation" command was OK
1248 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1249 (I found it when inspecting the code).
1251 20. There is now an overall timeout for performing a callout verification. It
1252 defaults to 4 times the callout timeout, which applies to individual SMTP
1253 commands during the callout. The overall timeout applies when there is more
1254 than one host that can be tried. The timeout is checked before trying the
1255 next host. This prevents very long delays if there are a large number of
1256 hosts and all are timing out (e.g. when the network connections are timing
1257 out). The value of the overall timeout can be changed by specifying an
1258 additional sub-option for "callout", called "maxwait". For example:
1260 verify = sender/callout=5s,maxwait=20s
1262 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1263 to the end before writing, but this should make it even safer).
1265 22. Exim was forgetting that it had advertised PIPELINING for the second and
1266 subsequent messages on an SMTP connection. It was also not resetting its
1267 memory on STARTTLS and an internal HELO.
1269 23. When Exim logs an SMTP synchronization error within a session, it now
1270 records whether PIPELINING has been advertised or not.
1272 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1273 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1274 rather than long int.
1276 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1278 26. Added the never_mail option to autoreply.
1284 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1285 crash if the getsockname() call failed; this can happen if a connection is
1286 closed very soon after it is established. The problem was simply in the
1287 order in which certain operations were done, causing Exim to try to write
1288 to the SMTP stream before it had set up the file descriptor. The bug has
1289 been fixed by making things happen in the correct order.
1295 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1296 before the connection was closed, thus losing the rejection response.
1298 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1299 some early Solaris releases, but causes trouble in current releases where
1300 socklen_t is defined.
1302 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1305 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1308 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1309 that releasing the top of it at the end releases what was used for sub-
1310 expansions (unless the block got too big). However, discard this block if
1311 the first thing is a variable or header, so that we can use its block when
1312 it is dynamic (useful for very large $message_headers, for example).
1314 6. Lookups now cache *every* query, not just the most recent. A new, separate
1315 store pool is used for this. It can be recovered when all lookup caches are
1316 flushed. Lookups now release memory at the end of their result strings.
1317 This has involved some general refactoring of the lookup sources.
1319 7. Some code has been added to the store_xxx() functions to reduce the amount
1320 of flapping under certain conditions.
1322 8. log_incoming_interface used to affect only the <= reception log lines. Now
1323 it causes the local interface and port to be added to several more SMTP log
1324 lines, for example "SMTP connection from", and rejection lines.
1326 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1328 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1330 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1331 could be overwritten at the end of the current message (or the start of a
1332 new message if it was set in a HELO ACL). The value is now preserved for
1333 the duration of the SMTP connection.
1335 12. If a transport had a headers_rewrite setting, and a matching header line
1336 contained an unqualified address, that address was qualified, even if it
1337 did not match any rewriting rules. The underlying bug was that the values
1338 of the flags that permit the existence of unqualified sender and recipient
1339 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1340 non-local messages, and by -bnq for local messages) were not being
1341 preserved with the message after it was received.
1343 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1344 "next input=" as part of the text comprising the host identity instead of
1345 the correct text. The code was using the same buffer for two different
1346 strings. However, depending on which order the printing function evaluated
1347 its arguments, the bug did not always show up. Under Linux, for example, my
1348 test suite worked just fine.
1350 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1351 doesn't work with some older versions of Perl. It has been changed to "my",
1352 which in any case is probably the better facility to use.
1354 15. A really picky compiler found some instances of statements for creating
1355 error messages that either had too many or two few arguments for the format
1358 16. The size of the buffer for calls to the DNS resolver has been increased
1359 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1360 for addresses that have a lot of PTR records. This alleviates a problem; it
1361 does not fully solve it.
1363 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1364 buffer now truncates the list and logs the incident, which is the same
1365 action as happens when Exim is looking up a host name and its aliases.
1366 Previously in this situation something unpredictable would happen;
1367 sometimes it was "internal error: store_reset failed".
1369 18. If a server dropped the connection unexpectedly when an Exim client was
1370 using GnuTLS and trying to read a response, the client delivery process
1371 crashed while trying to generate an error log message.
1373 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1374 string, for example:
1376 warn message = H1: something\nH2: something
1378 the text was added as a single header line from Exim's point of view
1379 though it ended up OK in the delivered message. However, searching for the
1380 second and subsequent header lines using $h_h2: did not work. This has been
1381 fixed. Similarly, if a system filter added multiple headers in this way,
1382 the routers could not see them.
1384 20. Expanded the error message when iplsearch is called with an invalid key to
1385 suggest using net-iplsearch in a host list.
1387 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1388 ACLs are no longer actually imposed (and a message to that effect is
1391 22. If a "gecos" field in a passwd entry contained escaped characters, in
1392 particular, if it contained a \" sequence, Exim got it wrong when building
1393 a From: or a Sender: header from that name. A second bug also caused
1394 incorrect handling when an unquoted " was present following a character
1395 that needed quoting.
1397 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1398 was not being matched caselessly.
1400 24. Arranged for all hyphens in the exim.8 source to be escaped with
1403 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1404 properly. Recipient callout cache records were still being keyed to include
1405 the sender, even when use_sender was set false. This led to far more
1406 callouts that were necessary. The sender is no longer included in the key
1407 when use_sender is false.
1409 26. Added "control = submission" modifier to ACLs.
1411 27. Added the ${base62d: operator to decode base 62 numbers.
1413 28. dnsdb lookups can now access SRV records.
1415 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1416 the configuration file.
1418 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1419 (-v) mode. This makes the output for a verbose queue run more intelligible.
1421 31. Added a use_postmaster feature to recipient callouts.
1423 32. Added the $body_zerocount variable, containing the number of binary zero
1424 bytes in the message body.
1426 33. The time of last modification of the "new" subdirectory is now used as the
1427 "mailbox time last read" when there is a quota error for a maildir
1430 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1432 35. Added +ignore_unknown as a special item in host lists.
1434 36. Code for decoding IPv6 addresses in host lists is now included, even if
1435 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1436 address was recognized as an IP address, but was then not correctly decoded
1437 into binary, causing unexpected and incorrect effects when compared with
1444 1. Very minor rewording of debugging text in manualroute to say "list of
1445 hosts" instead of "hostlist".
1447 2. If verify=header_syntax was set, and a header line with an unqualified
1448 address (no domain) and a large number of spaces between the end of the
1449 name and the colon was received, the reception process suffered a buffer
1450 overflow, and (when I tested it) crashed. This was caused by some obsolete
1451 code that should have been removed. The fix is to remove it!
1453 3. When running in the test harness, delay a bit after writing a bounce
1454 message to get a bit more predictability in the log output.
1456 4. Added a call to search_tidyup() just before forking a reception process. In
1457 theory, someone could use a lookup in the expansion of smtp_accept_max_
1458 per_host which, without the tidyup, could leave open a database connection.
1460 5. Added the variables $recipient_data and $sender_data which get set from a
1461 lookup success in an ACL "recipients" or "senders" condition, or a router
1462 "senders" option, similar to $domain_data and $local_part_data.
1464 6. Moved the writing of debug_print from before to after the "senders" test
1467 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1468 problems for message scanning, either using a data ACL, or using
1469 local_scan() because the Received: header was not generated till after they
1470 were called (in order to set the time as the time of reception completion).
1471 I have revised the way this works. The header is now generated after the
1472 body is received, but before the ACL or local_scan() are called. After they
1473 are run, the timestamp in the header is updated.
1479 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1480 before starting a queue runner without re-exec. This happened only when
1481 deliver_drop_privilege was set or when the Exim user was set to root. The
1482 effect of the bug was that timeouts during subsequent deliveries caused
1483 crashes instead of being properly handled. The handler is now left at its
1484 default (and expected) setting.
1486 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1487 message, again when deliver_drop_privilege is set or Exim is run as root.
1488 The bug described in (1) was not present in this case, but the tidying up
1489 of the other signals was missing. I have made the two cases consistent.
1491 3. The ignore_target_hosts setting on a manualroute router was being ignored
1492 for hosts that were looked up using the /MX notation.
1494 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1497 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1498 operated on the sender address. After changing the $sender_address to <>
1499 for the sender address verify, Exim was re-instated it as the original
1500 (before rewriting) address, but remembering that it had rewritten it, so it
1501 wasn't rewriting it again. This bug also had the effect of breaking the
1502 sender address verification caching when the sender address was rewritten.
1504 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1505 This has been changed so that if the ip literal address matches
1506 ignore_target_hosts, the router declines.
1508 7. Added expansion conditions match_domain, match_address, and match_local_
1509 part (NOT match_host).
1511 8. The placeholder for the Received: header didn't have a length field set.
1513 9. Added code to Exim itself and to exim_lock to test for a specific race
1514 condition that could lead to file corruption when using MBX delivery. The
1515 issue is with the lockfile that is created in /tmp. If this file is removed
1516 after a process has opened it but before that process has acquired a lock,
1517 there is the potential for a second process to recreate the file and also
1518 acquire a lock. This could lead to two Exim processes writing to the file
1519 at the same time. The added code performs the same test as UW imapd; it
1520 checks after acquiring the lock that its file descriptor still refers to
1521 the same named file.
1523 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1524 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1527 11. Added the smtp_active_hostname option. If used, this will typically be made
1528 to depend on the incoming interface address. Because $interface_address is
1529 not set up until the daemon has forked a reception process, error responses
1530 that can happen earlier (such as "too many connections") no longer contain
1533 12. If an expansion in a condition on a "warn" statement fails because a lookup
1534 defers, the "warn" statement is abandoned, and the next ACL statement is
1535 processed. Previously this caused the whole ACL to be aborted.
1537 13. Added the iplsearch lookup type.
1539 14. Added ident_timeout as a log selector.
1541 15. Added tls_certificate_verified as a log selector.
1543 16. Added a global option tls_require_ciphers (compare the smtp transport
1544 option of the same name). This controls incoming TLS connections.
1546 17. I finally figured out how to make tls_require_ciphers do a similar thing
1547 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1548 before starting the TLS session.
1550 18. Tabs are now shown as \t in -bP output.
1552 19. If the log selector return_path_on_delivery was set, Exim crashed when
1553 bouncing a message because it had too many Received: header lines.
1555 20. If two routers both had headers_remove settings, and the first one included
1556 a superfluous trailing colon, the final name in the first list and the
1557 first name in the second list were incorrectly joined into one item (with a
1558 colon in the middle).
1564 1. Added -C and -D options to the exinext utility, mainly to make it easier
1565 to include in the automated testing, but these could be helpful when
1566 multiple configurations are in use.
1568 2. The exinext utility was not formatting the output nicely when there was
1569 an alternate port involved in the retry record key, nor when there was a
1570 message id as well (for retries that were specific to a specific message
1571 and a specific host). It was also confused by IPv6 addresses, because of
1572 the additional colons they contain. I have fixed the IPv4 problem, and
1573 patched it up to do a reasonable job for IPv6.
1575 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1576 delivery, the log line now contains "pipelined" if PIPELINING was used.
1578 4. An SMTP transport process used to panic and die if the bind() call to set
1579 an explicit outgoing interface failed. This has been changed; it is now
1580 treated in the same way as a connect() failure.
1582 5. A reference to $sender_host_name in the part of a conditional expansion
1583 that was being skipped was still causing a DNS lookup. This no longer
1586 6. The def: expansion condition was not recognizing references to header lines
1587 that used bh_ and bheader_.
1589 7. Added the _cache feature to named lists.
1591 8. The code for checking quota_filecount in the appendfile transport was
1592 allowing one more file than it should have been.
1594 9. For compatibility with Sendmail, the command line option
1602 and sets the incoming protocol and host name (for trusted callers). The
1603 host name and its colon can be omitted when only the protocol is to be set.
1604 Note the Exim already has two private options, -pd and -ps, that refer to
1605 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1606 "s", but I don't think that's a major issue.
1608 10. A number of refactoring changes to the code, none of which should affect
1611 (a) The number of logging options was getting close to filling up the
1612 32-bit word that was used as a bit map. I have split them into two classes:
1613 those that are passed in the argument to log_write(), and those that are
1614 only ever tested independently outside of that function. These are now in
1615 separate 32-bit words, so there is plenty of room for expansion again.
1616 There is no change in the user interface or the logging behaviour.
1618 (b) When building, for example, log lines, the code previously used a
1619 macro that called string_cat() twice, in order to add two strings. This is
1620 not really sufficiently general. Furthermore, there was one instance where
1621 it was actually wrong because one of the argument was used twice, and in
1622 one call a function was used. (As it happened, calling the function twice
1623 did not affect the overall behaviour.) The macro has been replaced by a
1624 function that can join an arbitrary number of extra strings onto a growing
1627 (c) The code for expansion conditions now uses a table and a binary chop
1628 instead of a serial search (which was left over from when there were very
1629 few conditions). Also, it now recognizes conditions like "pam" even when
1630 the relevant support is not compiled in: a suitably worded error message is
1631 given if an attempt is made to use such a condition.
1633 11. Added ${time_interval:xxxxx}.
1635 12. A bug was causing one of the ddress fields not to be passed back correctly
1636 from remote delivery subprocesses. The field in question was not being
1637 subsequently used, so this caused to problems in practice.
1639 13. Added new log selectors queue_time and deliver_time.
1641 14. Might have fixed a bug in maildirsizefile handling that threw up
1642 "unexpected character" debug warnings, and recalculated the data
1643 unnecessarily. In any case, I expanded the warning message to give more
1646 15. Added the message "Restricted characters in address" to the statements in
1647 the default ACL that block characters like @ and % in local parts.
1649 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
1650 Three changes have been made:
1652 (a) There was a serious bug; a negative response to MAIL caused the whole
1653 recipient domain to be cached as invalid, thereby blocking all messages
1654 to all local parts at the same domain, from all senders. This bug has
1655 been fixed. The domain is no longer cached after a negative response to
1656 MAIL if the sender used is not empty.
1658 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
1661 (c) A new callout option, "use_sender" has been added for people who want
1662 the modified behaviour.
1668 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
1671 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
1672 indeed breaks things for older releases.
1674 3. Added additional logging to the case where there is a problem reading data
1675 from a filter that is running in a subprocess using a pipe, in order to
1676 try to track down a specific problem.
1678 4. Testing facility fudge: when running in the test harness and attempting
1679 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
1680 getting "No route to host". Convert this to a timeout.
1682 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
1685 6. Some OS don't have socklen_t but use size_t instead. This affects the
1686 fifth argument of getsockopt() amongst other things. This is now
1687 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
1688 can be set for individual OS. I have set it for SunOS5, OSF1, and
1689 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
1690 some earlier ones do not.
1692 7. Change 4.30/15 was not doing the test caselessly.
1694 8. The standard form for an IPv6 address literal was being rejected by address
1695 parsing in, for example, MAIL and RCPT commands. An example of this kind of
1696 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1697 this, as well as the form without the "IPv6" on the front (but only when
1698 address literals are enabled, of course).
1700 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
1702 10. Exim crashed if a message with an empty sender address specified by -f
1703 encountered a router with an errors_to setting. This could be provoked only
1704 by a command such as
1708 where an empty string was supplied; "<>" did not hit this bug.
1710 11. Installed PCRE release 4.5.
1712 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
1713 remained set. It is now erased.
1715 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
1716 times from message ids (which are base 36 rather than the normal 62).
1718 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
1719 were being counted as actual protocol errors, and logged if the log
1720 selector +smtp_protocol_error was set. One cannot be perfect in this test,
1721 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
1722 and DATA following a set of rejected RCPTs do not count as protocol errors.
1723 In other words, Exim assumes they were pipelined, though this may not
1724 actually be the case. Of course, in all cases the client gets an
1725 appropriate error code.
1727 15. If a lookup fails in an ACL condition, a message about the failure may
1728 be available; it is used if testing the ACL cannot continue, because most
1729 such messages specify what the cause of the deferral is. However, some
1730 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
1731 that caused an old message to be retained and used if a later statement
1732 caused a defer, replacing the real cause of the deferral.
1734 16. If an IP address had so many PTR records that the DNS lookup buffer
1735 was not large enough to hold them, Exim could crash while trying to process
1736 the truncated data. It now detects and logs this case.
1738 17. Further to 4.21/58, another change has been made: if (and only if) the
1739 first line of a message (the first header line) ends with CRLF, a bare LF
1740 in a subsequent header line has a space inserted after it, so as not to
1741 terminate the header.
1743 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
1744 unnecessarily, used atoi() instead of strtol(), and didn't check the
1745 termination when getting file sizes from file names by regex.
1747 19. Completely re-implemented the support for maildirsize files, in the light
1748 of a number of problems with the previous contributed implementation
1749 (4.30/29). In particular:
1751 . If the quota is zero, the maildirsize file is maintained, but no quota is
1754 . If the maildir directory does not exist, it is created before any attempt
1755 to write a maildirsize file.
1757 . The quota value in the file is just a cache; if the quota is changed in
1758 the transport, the new value overrides.
1760 . A regular expression is available for excluding directories from the
1763 20. The autoreply transport checks the characters in options that define the
1764 message's headers; it allows continued headers, but it was checking with
1765 isspace() after an embedded newline instead of explicitly looking for a
1768 21. If all the "regular" hosts to which an address was routed had passed their
1769 expiry times, and had not reached their retry times, the address was
1770 bounced, even if fallback hosts were defined. Now Exim should go on to try
1773 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
1774 equivalent code in the SMTP transport. Some hosts send humungous responses
1775 to HELO/EHLO, more than 1024 it seems.
1777 23. Refactoring: code in filter.c used (void *) for "any old type" but this
1778 gives compiler warnings in some environments. I've now done it "properly",
1781 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
1782 (because of problems with the built-in one) was declared to return uschar *
1783 instead of char *, causing compiler failure.
1785 25. Fixed a file descriptor leak when processing alias/forward files.
1787 26. Fixed a minor format string issue in dbfn.c.
1789 27. Typo in exim.c: ("dmbnz" for "dbmnz").
1791 28. If a filter file refered to $h_xxx or $message_headers, and the headers
1792 contained RFC 2047 "words", Exim's memory could, under certain conditions,
1795 29. When a sender address is verified, it is cached, to save repeating the test
1796 when there is more than one recipient in a message. However, when the
1797 verification involves a callout, it is possible for different callout
1798 options to be set for different recipients. It is too complicated to keep
1799 track of this in the cache, so now Exim always runs a verification when a
1800 callout is required, relying on the callout cache for the optimization.
1801 The overhead is duplication of the address routing, but this should not be
1804 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
1805 to be verified with callout=postmaster, and the main callout worked but the
1806 postmaster check failed, the verification correctly failed. However, if a
1807 subsequent RCPT command asked for sender verification *without* the
1808 postmaster check, incorrect caching caused this verification also to fail,
1811 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
1812 it was not caching the DNS options (qualify_single, search_parents) that
1813 were used when the lookup failed. A subsequent lookup with different
1814 options therefore always gave the same answer, though there were cases
1815 where it should not have. (Example: a "domains = !$mx_any" option on a
1816 dnslookup router: the "domains" option is always processed without any
1817 widening, but the router might have qualify_single set.) Now Exim uses the
1818 cached value only when the same options are set.
1820 32. Added John Jetmore's "exipick" utility to the distribution.
1822 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
1823 than a timeout (e.g. a certificate is required, and is not provided), an
1824 Exim server now closes the connection immediately. Previously it waited for
1825 the client to close - but if the client is SSL, it seems that they each
1826 wait for each other, leading to a delay before one of them times out.
1828 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
1829 maintained 0.8.x compatibility because I don't think many are using it, and
1830 it is clearly obsolete.
1832 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
1835 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
1836 client certificate was expired. A simple patch fixes this, though I don't
1837 understand the full logic of why the verify callback is called multiple
1840 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
1841 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
1842 which causes problems with some clients (such as the Certicom SSL Plus
1843 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
1844 disables the coutermeasure allowing Eudora to connect."
1846 38. Exim was not checking that a write() to a log file succeeded. This could
1847 lead to Bad Things if a log got too big, in particular if it hit a file
1848 size limit. Exim now panics and dies if it cannot write to a log file, just
1849 as it does if it cannot open a log file.
1851 39. Modified OS/Makefile-Linux so that it now contains
1853 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
1855 The two -D definitions ensure that Exim is compiled with large file
1856 support, which makes it possible to handle log files that are bigger than
1859 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
1860 instance) a domain was checked against a named list that involved a lookup,
1861 causing $domain_data to be set, then another domain was checked against the
1862 same list, then the first domain was re-checked, the value of $domain_data
1863 after the final check could be wrong. In particular, if the second check
1864 failed, it could be set empty. This bug probably also applied to
1867 41. The strip_trailing_dot option was not being applied to the address given
1868 with the -f command-line option.
1870 42. The code for reading a message's header from the spool was incrementing
1871 $received_count, but never initializing it. This meant that the value was
1872 incorrect (doubled) while delivering a message in the same process in which
1873 it was received. In the most common configuration of Exim, this never
1874 happens - a fresh exec is done - but it can happen when
1875 deliver_drop_privilege is set.
1877 43. When Exim logs an SMTP synchronization error - client data sent too soon -
1878 it now includes up to 150 characters of the unexpected data in the log
1881 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
1882 and building data strings. The size of both of these buffers was 10 000
1883 bytes - far larger than anybody would *ever* want, thought I. Needless to
1884 say, somebody hit the limit. I have increased the maximum line length to
1885 20 000 and the maximum data length of concatenated lines to 100 000. I have
1886 also fixed two bugs, because there was no checking on these buffers. Tsk,
1887 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1888 buffer is too small.
1890 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1891 lsearch lookups. Now it does.
1893 46. When parsing a route_list item in a manualroute router, a fixed-length
1894 buffer was used for the list of hosts. I made this 1024 bytes long,
1895 thinking that nobody would ever have a list of hosts that long. Wrong.
1896 Somebody had a whole pile of complicated expansion conditions, and the
1897 string was silently truncated, leading to an expansion error. It turns out
1898 that it is easier to change to an unlimited length (owing to other changes
1899 that have happened since this code was originally written) than to build
1900 structure for giving a limitation error. The length of the item that
1901 expands into the list of hosts is now unlimited.
1903 47. The lsearch lookup could not handle data where the length of text line was
1904 more than 4095 characters. Such lines were truncated, leading to shortened
1905 data being returned. It should now handle lines of any length.
1907 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
1908 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
1911 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
1912 replacing strings such as BIN_DIRECTORY in a source file: the replacement
1913 no longer happens in comment lines. A list of replacements is now placed
1914 at the head of all of the source files, except those whose only change is
1915 to replace PERL_COMMAND in the very first #! line.
1917 50. Replaced the slow insertion sort in queue.c, for sorting the list of
1918 messages on the queue, with a bottom-up merge sort, using code contributed
1919 by Michael Haardt. This should make operations like -bp somewhat faster on
1920 large queues. It won't affect queue runners, except when queue_run_in_order
1923 51. Installed eximstats 1.31 in the distribution.
1925 52. Added support for SRV lookups to the dnslookup router.
1927 53. If an ACL referred to $message_body or $message_body_end, the value was not
1928 reset for any messages that followed in the same SMTP session.
1930 54. The store-handling optimization for building very long strings was not
1931 differentiating between the different store pools. I don't think this
1932 actually made any difference in practice, but I've tidied it.
1934 55. While running the routers to verify a sender address, $sender_address
1935 was still set to the sender address. This is wrong, because when routing to
1936 send a bounce to the sender, it would be empty. Therefore, I have changed
1937 it so that, while verifying a sender address, $sender_address is set to <>.
1938 (There is no change to what happens when verifying a recipient address.)
1940 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
1941 target A or AAAA records (if not already returned) without resetting the
1942 qualify_single or search_parents options of the DNS resolver. These are
1943 inappropriate in this case because the targets of MX and SRV records must
1944 be FQDNs. A broken DNS record could cause trouble if it happened to have a
1945 target that, when qualified, matched something in the local domain. These
1946 two options are now turned off when doing these lookups.
1948 57. It seems that at least some releases of Reiserfs (which does not have the
1949 concept of a fixed number of inodes) returns zero and not -1 for the
1950 number of available inodes. This interacted badly with check_spool_inodes,
1951 which assumed that -1 was the "no such thing" setting. What I have done is
1952 to check that the total number of inodes is greater than zero before doing
1953 the test of how many are available.
1955 58. When a "warn" ACL statement has a log_message modifier, the message is
1956 remembered, and not repeated. This is to avoid a lot of repetition when a
1957 message has many recipients that cause the same warning to be written.
1958 Howewer, Exim was preserving the list of already written lines for an
1959 entire SMTP session, which doesn't seem right. The memory is now reset if a
1960 new message is started.
1962 59. The "rewrite" debugging flag was not showing the result of rewriting in the
1963 debugging output unless log_rewrite was also set.
1965 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
1966 of (int)(handle) when we know that handle contains (void *)(-1).
1968 61. The Exim daemon panic-logs an error return when it closes the incoming
1969 connection. However "connection reset by peer" seems to be common, and
1970 isn't really an error worthy of noting specially, so that particular error
1973 62. When Exim is trying to find all the local interfaces, it used to panic and
1974 die if the ioctl to get the interface flags failed. However, it seems that
1975 on at least one OS (Solaris 9) it is possible to have an interface that is
1976 included in the list of interfaces, but for which you get a failure error
1977 for this call. This happens when the interface is not "plumbed" into a
1978 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
1979 failure of the "get flags" call assumes that the interface is down.
1981 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
1982 makes life easier for people who are doing arithmetic on fields extracted
1983 from dates, where you often get leading zeros that should not be
1984 interpreted as octal.
1986 64. Added qualify_domain to the redirect router, to override the global
1989 65. If a pathologically long header line contained very many addresses (the
1990 report of this problem mentioned 10 000) and each of them was rewritten,
1991 Exim could use up a very large amount of memory. (It kept on making new
1992 copies of the header line as it rewrote, and never released the old ones.)
1993 At the expense of a bit more processing, the header rewriting function has
1994 been changed so that it no longer eats memory in this way.
1996 66. The generation of the Received: header has been moved from the time that a
1997 message starts to be received, to the time that it finishes. The timestamp
1998 in the Received: header should now be very close to that of the <= log
1999 line. There are two side-effects of this change:
2001 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2002 logged header lines no longer include the local Received: line, because
2003 it has not yet been created. The same applies to a copy of the message
2004 that is returned to a non-SMTP sender when a message is rejected.
2006 (b) When a filter file is tested using -bf, no additional Received: header
2007 is added to the test message. After some thought, I decided that this
2010 This change does not affect the value of $received_for. It is still set
2011 after address rewriting, but before local_scan() is called.
2013 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2015 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2016 gave an unhelpful panic error message, and a defer error. I have managed to
2017 change this behaviour so that it now rejects any supplied certificate,
2018 which seems right, as the list of acceptable certificates is empty.
2020 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2021 gave an unhelpful defer error. I have not managed to make this reject any
2022 supplied certificates, but the error message it gives is "no certificate
2023 supplied", which is not helpful.
2025 70. exigrep's output now also includes lines that are not associated with any
2026 message, but which match the given pattern. Implemented by a patch from
2027 Martin Sluka, which also tidied up the Perl a bit.
2029 71. Recipient callout verification, like sender verification, was using <> in
2030 the MAIL FROM command. This isn't really the right thing, since the actual
2031 sender may affect whether the remote host accepts the recipient or not. I
2032 have changed it to use the actual sender in the callout; this means that
2033 the cache record is now keyed on a recipient/sender pair, not just the
2034 recipient address. There doesn't seem to be a real danger of callout loops,
2035 since a callout by the remote host to check the sender would use <>.
2036 [SEE ABOVE: changed after hitting problems.]
2038 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2039 temporary errors. However, in the case of such a code being given after
2040 the end of a data transmission (i.e. after ".") Exim was failing to write
2041 a retry record for the message. (Yes, there was some broken host that was
2042 actually sending 8xx at this point.)
2044 73. An unknown lookup type in a host list could cause Exim to panic-die when
2045 the list was checked. (An example that provoked this was putting <; in the
2046 middle of a list instead of at the start.) If this happened during a DATA
2047 ACL check, a -D file could be left lying around. This kind of configuration
2048 error no longer causes Exim to die; instead it causes a defer errror. The
2049 incident is still logged to the main and panic logs.
2051 74. Buglet left over from Exim 3 conversion. The message "too many messages
2052 in one connection" was written to the rejectlog but not the mainlog, except
2053 when address rewriting (yes!) was being logged.
2055 75. Added write_rejectlog option.
2057 76. When a system filter was run not as root (that is, when system_filter_user
2058 was set), the values of the $n variables were not being returned to the
2059 main process; thus, they were not subsequently available in the $sn
2062 77. Added +return_path_on_delivery log selector.
2064 78. A connection timeout was being treated differently from recipients deferred
2065 when testing hosts_max_try with a message that was older than the host's
2066 retry timeout. (The host should not be counted, thus allowing all hosts to
2067 be tried at least once before bouncing.) This may have been the cause of an
2068 occasionally reported bug whereby a message would remain on the queue
2069 longer than the retry timeout, but would be bounced if a delivery was
2070 forced. I say "may" because I never totally pinned down the problem;
2071 setting up timeout/retry tests is difficult. See also the next item.
2073 79. The ultimate address timeout was not being applied to errors that involved
2074 a combination of host plus message (for example, a timeout on a MAIL
2075 command). When an address resolved to a number of possible hosts, and they
2076 were not all tried for each delivery (e.g. because of hosts_max_try), a
2077 message could remain on the queue longer than the retry timeout.
2079 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2082 81. Fixed an obscure SMTP outgoing bug which required at least the following
2083 conditions: (a) there was another message waiting for the same server;
2084 (b) the server returned 5xx to all RCPT commands in the first message so
2085 that the message was not completed; (c) the server dropped the connection
2086 or gave a negative response to the RSET that Exim sends to abort the
2087 transaction. The observed case was a dropped connection after DATA that had
2088 been sent in pipelining mode. That is, the server had advertised PIPELINING
2089 but was not implementing it correctly. The effect of the bug was incorrect
2090 behaviour, such as trying another host, and this could lead to a crash.
2096 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2097 and daemon.c were passed as pointers to ints; they should have been
2098 pointers to socklen_t variables (which are typically unsigned ints).
2100 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2103 3. Fixed a really odd bug that affected only the testing scheme; patching a
2104 certain fixed string in the binary changed the value of another string that
2105 happened to be identical to the end of the original first string.
2107 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2108 name", it returns that address as the IP address. On some operating
2109 systems (e.g. Solaris), it also passes back the IP address string as the
2110 "host name". However, on others (e.g. Linux), it passes back an empty
2111 string. Exim wasn't checking for this, and was changing the host name to an
2112 empty string, assuming it had been canonicized.
2114 5. Although rare, it is permitted to have more than one PTR record for a given
2115 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2116 all the names associated with an address, because they do in Solaris.
2117 However, it seems that they do not in Linux for data that comes from the
2118 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2119 I found this out when I moved to a new Linux workstation and tried to run
2120 the Exim test suite.
2122 To get round this problem I have changed the code so that it now does its
2123 own call to the DNS to look up PTR records when searching for a host name.
2124 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2125 addresses that are only in /etc/hosts are still found.
2127 This behaviour is, however, controlled by an option called host_lookup_
2128 order, which defaults to "bydns:byaddr". If people want to use the other
2129 order, or indeed, just use one or the other means of lookup, they can
2130 specify it in this variable.
2132 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2133 some operating systems, this comes back from gethostbyaddr() as an empty
2134 string, and this is what Exim used to test for. However, it seems that in
2135 other systems, "." is yielded. Exim now tests for this case too.
2137 7. The values of check_spool_space and check_log_space are now held internally
2138 as a number of kilobytes instead of an absolute number of bytes. If a
2139 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2140 kilobyte. This means that much larger values can be stored.
2142 8. Exim monitor: an attempt to get the action menu when not actually pointing
2143 at a message produces an empty menu entitled "No message selected". This
2144 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2145 no entries in it ("Shell widget menu has zero width and/or height"). So I
2146 have added a single, blank menu entry in this case.
2148 9. Added ${quote_local_part.
2150 10. MIME decoding is now applied to the contents of Subject: header lines when
2153 11. Now that a reference to $sender_host_address automatically causes a reverse
2154 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2155 host lookup before query-style lookups in lists that might use this
2156 variable. This has therefore been abolished, and the "net-" prefix is no
2157 longer necessary for query-style lookups.
2159 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2160 have been a typo for LFLAGS, so it has been changed.
2162 13. The install script calls Exim with "-C /dev/null" in order to find the
2163 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2164 to be output. Howeve, since Exim outputs its version number before the
2165 error, it didn't break the script. It just looked ugly. I fixed this by
2166 always allowing "-C /dev/null" if the caller is root.
2168 14. Ignore overlarge ACL variable number when reading spool file - insurance
2169 against a later release with more variables having written the file.
2171 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2172 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2173 this, as well as the form without the "IPv6" on the front.
2175 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2176 OS/Makefile-Darwin file.
2178 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2179 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2181 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2182 variable. However, some OS use ints for inodes. I've added cast to long int
2183 to get rid of the compiler warning.
2185 19. I had forgotten to lock out "/../" in configuration file names when
2186 ALT_CONFIG_PREFIX was set.
2188 20. Routers used for verification do not need to specify transports. However,
2189 if such a router generated a host list, and callout was configured, Exim
2190 crashed, because it could not find a port number from the (non-existent)
2191 transport. It now assumes port 25 in this circumstance.
2193 21. Added the -t option to exigrep.
2195 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2196 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2197 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2198 them is set, LOOKUP_LSEARCH is forced.
2200 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2202 24. Added sender and host information to the "rejected by local_scan()" log
2203 line; previously there was no indication of these.
2205 25. Added .include_if_exists.
2207 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2208 Linux. It turns out that not all file systems support this. Apparently some
2209 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2210 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2211 not supported on the file descriptor, is now ignored when Exim is trying to
2212 sync a directory. This applies only to Linux.
2214 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2216 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2217 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2220 29. Added support for maildirsize files from supplied patch (modified a bit).
2222 30. The use of :fail: followed by an empty string could lead Exim to respond to
2223 sender verification failures with (e.g.):
2225 550 Verification failed for <xxx>
2226 550 Sender verify failed
2228 where the first response line was missing the '-' that indicates it is not
2229 the final line of the response.
2231 31. The loop for finding the name of the user that called Exim had a hardwired
2232 limit of 10; it now uses the value of finduser_retries, which is used for
2233 all other user lookups.
2235 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2238 33. Exim was neglecting to zero errno before one call of strtol() when
2239 expanding a string and expecting an integer value. On some systems this
2240 resulted in spurious "integer overflow" errors. Also, it was casting the
2241 result into an int without checking.
2243 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2244 did not work. The code looks as if it has *never* worked, though it appears
2245 to have been documented since at least releast 1.62. I have made it work.
2247 35. The "timeout_DNS" error in retry rules, also documented since at least
2248 1.62, also never worked. As it isn't clear exactly what this means, and
2249 clearly it isn't a major issue, I have abolished the feature by treating it
2250 as "timeout", and writing a warning to the main and panic logs.
2252 36. The display of retry rules for -brt wasn't always showing the error code
2255 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2256 timeout_connect_A, timeout_connect_MX.
2258 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2259 to the empty sender.
2261 39. The daemon was not analysing the content of -oX till after it had closed
2262 stderr and disconnected from the controlling terminal. This meant that any
2263 syntax errors were only noted on the panic log, and the return code from
2264 the command was 0. By re-arranging the code a little, I've made the
2265 decoding happen first, so such errors now appear on stderr, and the return
2266 code is 1. However, the actual setting up of the sockets still happens in
2267 the disconnected process, so errors there are still only recorded on the
2270 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2271 connections (as happens on some IP stacks) was logged at start up time as
2272 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2273 it from "IPv6 and IPv4", which means that two separate sockets are being
2276 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2277 says whether AF_INET or AF_INET6 was passed as an argument.
2279 42. Exiwhat output was messed up when time zones were included in log
2282 43. Exiwhat now gives more information about the daemon's listening ports,
2283 and whether -tls-on-connect was used.
2285 44. The "port" option of the smtp transport is now expanded.
2287 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2288 silently ignored. Now an error message is written to the main and panic
2291 46. There's a new ACL modifier called "logwrite" which writes to a log file
2292 as soon as it is encountered.
2294 47. Added $local_user_uid and $local_user_gid at routing time.
2296 48. Exim crashed when trying to verify a sender address that was being
2299 49. Exim was recognizing only a space character after ".include". It now also
2300 recognizes a tab character.
2302 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2303 extracting the relevant information from the specification. The man page no
2304 longer contains scrambled data for the -d option, and I've added a section
2305 at the front about calling Exim under different names.
2307 51. Added "extra_headers" argument to the "mail" command in filter files.
2309 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2312 53. Installed eximstats 1.29.
2314 54. Added transport_filter_timeout as a generic transport option.
2316 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2317 Cc: header lines. This was required by RFC 822, but it not required by RFC
2320 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2321 incoming messages that did not have them. Now it does so only if the
2322 message originates locally, that is, if there is no associated remote host
2323 address. When Resent- header lines are present, this applies to the Resent-
2324 lines rather than the non-Resent- lines.
2326 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2327 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2329 58. Messages for configuration errors now include the name of the main
2330 configuration file - useful now that there may be more than one file in a
2331 list (.included file names were always shown).
2333 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2334 for those rare installations that do not start the daemon as root or run it
2335 setuid root. I've cut out the call to initgroups() if the daemon is not
2338 60. The Exim user and group can now be bound into the binary as text strings
2339 that are looked up at the start of Exim's processing.
2341 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2343 62. Added $mailstore_basename variable.
2345 63. Installed patch to sieve.c from Michael Haardt.
2347 64. When Exim failed to open the panic log after failing to open the main log,
2348 the original message it was trying to log was written to stderr and debug
2349 output, but if they were not available (the usual case in production), it
2350 was lost. Now it is written to syslog before the two lines that record the
2351 failures to open the logs.
2353 65. Users' Exim filters run in subprocesses under the user's uid. It is
2354 possible for a "deliver" command or an alias in a "personal" command to
2355 provoke an address rewrite. If logging of address rewriting is configured,
2356 this fails because the process is not running as root or exim. There may be
2357 a better way of dealing with this, but for the moment (because 4.30 needs
2358 to be released), I have disabled address rewrite logging when running a
2359 filter in a non-root, non-exim process.
2365 1. The buildconfig auxiliary program wasn't quoting the value set for
2366 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2367 not defined. This bug was masked in 4.22 by the effect that was fixed in
2370 2. Some messages that were rejected after a message id was allocated were
2371 shown as "incomplete" by exigrep. It no longer does this for messages that
2372 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2374 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2375 have allow_domain_literals set, the ID did not get logged in the <= line.
2376 Domain literals are now always recognized in Message-ID: header lines.
2378 4. The first argument for a ${extract expansion item is the key name or field
2379 number. Leading and trailing spaces in this item were not being ignored,
2380 causing some misleading effects.
2382 5. When deliver_drop_privilege was set, single queue runner processes started
2383 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2384 same command in the process it spins off) were not dropping privilege.
2386 6. When the daemon running as "exim" started a queue runner, it always
2387 re-executed Exim in the spun-off process. This is a waste of effort when
2388 deliver_drop_privilege is set. The new process now just calls the
2389 queue-runner function directly.
2395 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2398 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2399 ignored. Though the use of -oP was forcing the writing of a pid file, it
2400 was always written to the default place.
2402 3. If the message "no IP address found for host xxxx" is generated during
2403 incoming verification, it is now followed by identification of the incoming
2404 connection (so you can more easily find what provoked it).
2406 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2408 5. Added some features to "harden" Exim a bit more against certain attacks:
2410 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2411 be put in Local/Makefile. This is like the never_users runtime option,
2412 but it cannot be overridden. The default setting is "root".
2414 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2415 prefix string with which any file named in a -C command line option
2418 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2419 is retained for -C and -D only if the caller of Exim is root. Without
2420 it, the exim user may also use -C and -D and retain privilege.
2422 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2423 command line option is disabled.
2425 6. Macro names set by the -D option must start with an upper case letter, just
2426 like macro names defined in the configuration file.
2428 7. Added "dereference=" facility to LDAP.
2430 8. Two instances of the typo "uknown" in the source files are fixed.
2432 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2433 the Configure-Makefile script screwed up while processing it.
2435 10. Incorporated PCRE 4.4.
2437 11. The SMTP synchronization check was not operating right at the start of an
2438 SMTP session. For example, it could not catch a HELO sent before the client
2439 waited for the greeting. There is now a check for outstanding input at the
2440 point when the greeting is written. Because of the duplex, asynchronous
2441 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2442 way, but not yet received, when the check is performed.
2444 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2445 on TCP/IP sockets, because this apparently causes some broken clients to
2448 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2449 unchanged) from the Cygwin maintainer.
2451 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2452 was supported instead of testing for mbx. Effectively a typo.
2454 15. The spa authenticator server code was not checking that the input it
2455 received was valid base64.
2457 16. The debug output line for the "set" modifier in ACLs was not showing the
2458 name of the variable that was being set.
2460 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2462 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2463 lookup on its own. Something else was needed to trigger the lookup. For
2464 example, a match in host_lookup or the need for a host name in a host list.
2465 Now, if $sender_host_name is referenced and the host name has not yet been
2466 looked up, a lookup is performed. If the lookup fails, the variable remains
2467 empty, and $host_lookup_failed is set to "1".
2469 19. Added "eqi" as a case-independent comparison operator.
2471 20. The saslauthd authentication condition could segfault if neither service
2472 nor realm was specified.
2474 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2475 error message that was logged was misleading, and incoming SMTP
2476 connections were dropped. The message is now more accurate, and temporary
2477 errors are given to SMTP connections.
2479 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2480 (except RSET) with a 421 error, until QUIT is received. However, it was
2481 failing to send a response to QUIT.
2483 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2484 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2485 ACL is now run after the helo_try_verify_hosts code.
2487 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2488 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2489 case-independent, so other case variants are also recognized). Apparently
2490 some systems use these upper case variants.
2492 25. If more than two messages were waiting for the same host, and a transport
2493 filter was specified for the transport, Exim sent two messages over the
2494 same TCP/IP connection, and then failed with "socket operation on non-
2495 socket" when it tried to send the third.
2497 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2499 27. The extern definition of crypt16() in expand.c was not being excluded when
2500 the OS had its own crypt16() function.
2502 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2503 as a preferred synonym for return_size_limit, both as an option and as an
2506 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2508 30. Changed the default configuration ACL to relax the local part checking rule
2509 for addresses that are not in any local domains. For these addresses,
2510 slashes and pipe symbols are allowed within local parts, but the sequence
2511 /../ is explicitly forbidden.
2513 31. SPA server authentication was not clearing the challenge buffer before
2516 32. log_message in a "warn" ACL statement was writing to the reject log as
2517 well as to the main log, which contradicts the documentation and doesn't
2518 seem right (because no rejection is happening). So I have stopped it.
2520 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2521 However, I am unable to do any testing of this.
2523 34. Fixed an infelicity in the appendfile transport. When checking directories
2524 for a mailbox, to see if any needed to be created, it was accidentally
2525 using path names with one or more superfluous leading slashes; tracing
2526 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2528 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2529 contents are added to the log line that is written for every discarded
2530 recipient. (Previously a log_message setting was ignored.)
2532 36. The ${quote: operator now quotes the string if it is empty.
2534 37. The install script runs exim in order to find its version number. If for
2535 some reason other than non-existence or emptiness, which it checks, it
2536 could not run './exim', it was installing it with an empty version number,
2537 i.e. as "exim-". This error state is now caught, and the installation is
2540 38. An argument was missing from the function that creates an error message
2541 when Exim fails to connect to the socket for saslauthd authentication.
2542 This could cause Exim to crash, or give a corrupted message.
2544 39. Added isip, isip4, and isip6 to ${if conditions.
2546 40. The ACL variables $acl_xx are now saved with the message, and can be
2547 accessed later in routers, transports, and filters.
2549 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2550 strings in the file are not string-expanded.
2552 42. If a MAIL command specified a SIZE value that was too large to fit into an
2553 int variable, the check against message_size_limit failed. Such values are
2554 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2555 day this will have to be increased, but I don't think I want to be around
2556 when emails are that large.
2563 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2564 iconv() is not standard in FreeBSD.
2566 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2567 IPv6 enabled. The observed symptom was a segmentation fault on return from
2568 the function os_common_find_running_interfaces() in src/os.c.
2570 3. In the check_special_case() function in daemon.c I had used "errno" as an
2571 argument name, which causes warnings on some systems. This was basically a
2572 typo, since it was named "eno" in the comments!
2574 4. The code that waits for the clock to tick (at a resolution of some fraction
2575 of a second) so as to ensure message-id uniqueness was always waiting for
2576 at least one whole tick, when it could have waited for less. [This is
2577 almost certainly not relevant at current processor speeds, where it is
2578 unlikely to ever wait at all. But we try to future-proof.]
2580 5. The function that sleeps for a time interval that includes fractions of a
2581 second contained a race. It did not block SIGALRM between setting the
2582 timer, and suspending (a couple of lines later). If the interval was short
2583 and the sigsuspend() was delayed until after it had expired, the suspension
2584 never ended. On busy systems this could lead to processes getting stuck for
2587 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2588 process, before it forks any delivery processes. The open lookup caching
2589 mechanism meant that the open file or database connection was passed into
2590 the delivery process. The problem was that delivery processes always tidy
2591 up cached lookup data. This could cause a problem for the next delivery
2592 process started by the queue runner, because the external queue runner
2593 process does not know about the closure. So the next delivery process
2594 still has data in the lookup cache. In the case of a file lookup, there was
2595 no problem because closing a file descriptor in a subprocess doesn't affect
2596 the parent. However, if the lookup was caching a connection to a database,
2597 the connection was closed, and the second delivery process was likely to
2598 see errors such as "PGSQL: query failed: server closed the connection
2599 unexpectedly". The problem has been fixed by closing all cached lookups
2600 in a queue runner before running a delivery process.
2602 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2603 seem to have the "const" qualifier which it has on other OS. I've
2606 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2607 *of the same type* (client or server) with the same public name that an
2608 error should be diagnosed.
2610 9. When Exim looked up a host name for an IP address, but failed to find the
2611 original IP address when looking up the host name (a safety check), it
2612 output the message "<ip address> does not match any IP for NULL", which was
2613 confusing, to say the least. The bug was that the host name should have
2614 appeared instead of "NULL".
2616 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2617 user that is built into the binary, and the -C or -D options is used, root
2618 privilege is dropped before the configuration file is read. In addition,
2619 logging is switched to stderr instead of the normal log files. If the
2620 configuration then re-defines the Exim user, the unprivileged environment
2621 is probably not what is expected, so Exim logs a panic warning message (but
2624 However, if deliver_drop_privilege is set, the unprivileged state may well
2625 be exactly what is intended, so the warning has been cut out in that case,
2626 and Exim is allowed to try to write to its normal log files.
2632 1. smtp_return_error_details was not giving details for temporary sender
2633 or receiver verification errors.
2635 2. Diagnose a configuration error if two authenticators have the same public
2638 3. Exim used not to create the message log file for a message until the first
2639 delivery attempt. This could be confusing when incoming messages were held
2640 for policy or load reasons. The message log file is now created at the time
2641 the message is received, and an initial "Received" line is written to it.
2643 4. The automatically generated man page for command line options had a minor
2644 bug that caused no ill effects; however, a more serious problem was that
2645 the procedure for building the man page automatically didn't always
2646 operate. Consequently, release 4.20 contains an out-of-date version. This
2647 shouldn't happen again.
2649 5. When building Exim with embedded Perl support, the script that builds the
2650 Makefile was calling 'perl' to find its compile-time parameters, ignoring
2651 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
2653 6. The freeze_tell option was not being used for messages that were frozen on
2654 arrival, either by an ACL or by local_scan().
2656 7. Added the smtp_incomplete_transaction log selector.
2658 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
2659 was accepting AUTH without a new EHLO.
2661 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
2662 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
2665 10. Logging of TCP/IP connections (when configured) now happens in the main
2666 daemon process instead of the child process, so that the TCP/IP connection
2667 count is more accurate (but it can never be perfect).
2669 11. The use of "drop" in a nested ACL was not being handled correctly in the
2670 outer ACL. Now, if condition failure induced by the nested "drop" causes
2671 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
2672 or "require"), the connection is dropped.
2674 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
2675 that yield "discard" can now be used with an "accept" or a "discard" verb,
2676 but an error is generated for any others (because I can't see a useful way
2677 to define what should happen).
2679 13. When an ACL is read dynamically from a file (or anywhere else), the lines
2680 are now processed in the same way as lines in the Exim configuration file.
2681 In particular, continuation lines are supported.
2683 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
2685 15. Added -ti meaning -t -i.
2687 16. Check for letters, digits, hyphens, and dots in the names of dnslist
2688 domains, and warn by logging if others are found.
2690 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
2691 returned from GIFCONF when Exim is trying to find the list of interfaces on
2692 a host. The code in os.c has been modified to copy each ifreq to an aligned
2693 structure in all cases.
2695 Also, in some cases, the returned ifreq's were being copied to a 'struct
2696 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
2697 means the last couple of bytes of an IPv6 address could be chopped if the
2698 ifreq contained only a normal sockaddr (14 bytes storage).
2700 18. Named domain lists were not supported in the hosts_treat_as_local option.
2701 An entry such as +xxxx was not recognized, and was treated as a literal
2704 19. Ensure that header lines added by a DATA ACL are included in the reject log
2705 if the ACL subsequently rejects the message.
2707 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
2708 MD5 (which is deprecated).
2710 21. When testing a filter file using -bf, Exim was writing a message when it
2711 took the sender from a "From " line in the message, but it was not doing so
2712 when it took $return_path from a Return-Path: header line. It now does.
2714 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
2715 with a valid header line field name (a series of printing characters
2716 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
2718 23. Changed "disc" in the source to "disk" to conform to the documentation and
2719 the book and for uniformity.
2721 24. Ignore Sendmail's -Ooption=value command line item.
2723 25. When execve() failed while trying to run a command in a pipe transport,
2724 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
2725 could be confused with a return value of 69 from the command itself. This
2726 has been changed to 127, the value the shell returns if it is asked to run
2727 a non-existent command. The wording for the related log line suggests a
2728 non-existent command as the problem.
2730 26. If received_header_text expands to an empty string, do not add a Received:
2731 header line to the message. (Well, it adds a token one on the spool, but
2732 marks it "old" so that it doesn't get used or transmitted.)
2734 27. Installed eximstats 1.28 (addition of -nt option).
2736 28. There was no check for failure on the call to getsockname() in the daemon
2737 code. This can fail if there is a shortage of resources on the system, with
2738 ENOMEM, for example. A temporary error is now given on failure.
2740 29. Contrary to the C standard, it seems that in some environments, the
2741 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
2742 program. Exim now does this explicitly; it affects the formatting of
2743 timestamps using strftime().
2745 30. If exiqsumm was given junk data, it threw up some uninitialized variable
2746 complaints. I've now initialized all the variables, to avoid this.
2748 32. Header lines added by a system filter were not being "seen" during
2749 transport-time rewrites.
2751 33. The info_callback() function passed to OpenSSL is set up with type void
2752 (*)(SSL *, int, int), as described somewhere. However, when calling the
2753 function (actually a macro) that sets it up, the type void(*)() is
2754 expected. I've put in a cast to prevent warnings from picky compilers.
2756 34. If a DNS black list lookup found a CNAME record, but there were no A
2757 records associated with the domain it pointed at, Exim crashed.
2759 35. If a DNS black list lookup returned more than one A record, Exim ignored
2760 all but the first. It now scans all returned addresses if a particular IP
2761 value is being sought. In this situation, the contents of the
2762 $dnslist_value variable are a list of all the addresses, separated by a
2765 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
2766 to accept a host name and all its aliases if the forward lookup for any of
2767 them yielded the IP address of the incoming connection. Now it accepts only
2768 those names whose forward lookup yields the correct IP address. Any other
2769 names are discarded. This closes a loophole whereby a rogue DNS
2770 administrator could create reverse DNS records to break through a
2771 wildcarded host restriction in an ACL.
2773 37. If a user filter or a system filter that ran in a subprocess used any of
2774 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
2775 the wrong values were passed to the pipe command ($thisaddress had the
2776 value of $0, $0 had the value of $1, etc). This bug was introduced by
2777 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
2779 38. Improved the line breaking for long SMTP error messages from ACLs.
2780 Previously, if there was no break point between 40 and 75 characters, Exim
2781 left the rest of the message alone. Two changes have been made: (a) I've
2782 reduced the minimum length to 35 characters; (b) if it can't find a break
2783 point between 35 and 75 characters, it looks ahead and uses the first one
2784 that it finds. This may give the occasional overlong line, but at least the
2785 remaining text gets split now.
2787 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
2788 file descriptors might be low, and that setting 1000 would always raise it.
2789 It turns out that in some environments, the limit is already over 1000 and
2790 that lowering it causes trouble. So now Exim takes care not to decrease it.
2792 40. When delivering a message, the value of $return_path is set to $sender_
2793 address at the start of routing (routers may change the value). By an
2794 oversight, this default was not being set up when an address was tested by
2795 -bt or -bv, which affected the outcome if any router or filter referred to
2798 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
2799 log only when "message" or "log_message" are set. However, if one of the
2800 conditions was an address verification, or a call to a nested ACL, the
2801 messages generated by the underlying test were being passed through. This
2802 no longer happens. The underlying message is available in $acl_verify_
2803 message for both "message" and "log_message" expansions, so it can be
2804 passed through if needed.
2806 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
2807 new expansion $bh_ to give the encoded byte string without charset
2808 translation. Translation happens only if iconv() is available; HAVE_ICONV
2809 indicates this at build time. HEADERS_CHARSET gives the charset to
2810 translate to; headers_charset can change it in the configuration, and
2811 "headers charset" can change it in an individual filter file.
2813 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
2814 that creates RFC 2047 encoded "words" labels them as that charset instead
2815 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
2816 expansion operator; (ii) when Exim creates a From: line for a local
2817 message; (iii) when a header line is rewritten to include a "phrase" part.
2819 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
2820 buggy, causing it to skip the first lines of messages whose message ID
2821 ended in 'D'. This would not have bitten before Exim release 4.14, because
2822 message IDs were unlikely to end in 'D' before then. The effect was to have
2823 incorrect size information for certain domains.
2825 45. #include "config.h" was missing at the start of the crypt16.c module. This
2826 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
2829 46. If there was a timeout during a "random" callout check, Exim treated it as
2830 a failure of the random address, and carried on sending RSET and the real
2831 address. If the delay was just some slowness somewhere, the response to the
2832 original RCPT would be taken as a response to RSET and so on, causing
2833 mayhem of various kinds.
2835 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
2836 when I implemented it. It didn't allow for the fact that some option values
2837 may legitimatetly be negative (e.g. size_addition), and it didn't even do
2838 the right test for positive values.
2840 48. Domain names in DNS records are case-independent. Exim always looks them up
2841 in lower case. Some resolvers return domain names in exactly the case they
2842 appear in the zone file, that is, they may contain uppercase letters. Not
2843 all resolvers do this - some return always lower case. Exim was treating a
2844 change of case by a resolver as a change of domain, similar to a widening
2845 of a domain abbreviation. This triggered its re-routing code and so it was
2846 trying to route what was effectively the same domain again. This normally
2847 caused routing to fail (because the router wouldn't handle the domain
2848 twice). Now Exim checks for this case specially, and just changes the
2849 casing of the domain that it ultimately uses when it transmits the message
2852 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
2855 50. If a filter generated a file delivery with a non-absolute name (possible if
2856 no home directory exists for the router), the forbid_file option was not
2859 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
2860 the existing equality matching.
2862 52. Exim was using ints instead of ino_t variables in some places where it was
2863 dealing with inode numbers.
2865 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
2866 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
2867 called TMPDIR, and if it finds it is different, it changes its value.
2869 54. The smtp_printf() function is now made available to local_scan() so
2870 additional output lines can be written before returning. There is also an
2871 smtp_fflush() function to enable the detection of a dropped connection.
2872 The variables smtp_input and smtp_batched_input are exported to
2875 55. Changed the default runtime configuration: the message "Unknown user"
2876 has been removed from the ACL, and instead placed on the localuser router,
2877 using the cannot_route_message feature. This means that any verification
2878 failures that generate their own messages won't get overridden. Similarly,
2879 the "Unrouteable address" message that was in the ACL for unverifiable
2880 relay addresses has also been removed.
2882 56. Added hosts_avoid_esmtp to the smtp transport.
2884 57. The exicyclog script was not checking for the esoteric option
2885 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2886 will work only if exicyclog is run under the appropriate euid.
2888 58. Following a discussion on the list, the rules by which Exim recognises line
2889 endings on incoming messages have been changed. The -dropcr and drop_cr
2890 options are now no-ops, retained only for backwards compatibility. The
2891 following line terminators are recognized: LF CRLF CR. However, special
2892 processing applies to CR:
2894 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2895 nor a local message in the state where . is a terminator.
2897 (ii) If a bare CR is encountered in a header line, an extra space is added
2898 after the line terminator so as not to end the header. The reasoning
2899 behind this is that bare CRs in header lines are most likely either
2900 to be mistakes, or people trying to play silly games.
2902 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
2903 was being incorrectly given as 18 bytes larger than it should have been.
2904 This is a VOB (very old bug).
2906 60. This may never have affected anything current, but just in case it has:
2907 When the local host is found other than at the start of a list of hosts,
2908 the local host, those with the same MX, and any that follow, are discarded.
2909 When the list in question was part of a longer list of hosts, the following
2910 hosts (not currently being processed) were also being discarded. This no
2911 longer happens. I'm not sure if this situation could ever has previously
2914 61. Added the "/MX" feature to lists of hosts in the manualroute and query
2917 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
2918 header. This is something that is recommended in a new Internet Draft, and
2919 is something that is documented as being done by Sendmail. There are two
2920 possible values. For messages generated by the autoreply transport, Exim
2923 Auto-Submitted: auto-replied
2925 whereas for all other generated messages (e.g. bounces) it adds
2927 Auto-Submitted: auto-generated
2929 63. The "personal" condition in filters now includes a test for the
2930 Auto-Submitted: header. If it contains the string "auto-" the message it
2931 not considered personal.
2933 64. Added rcpt_include_affixes as a generic transport option.
2935 65. Added queue_only_override (default true).
2937 66. Added the syslog_duplication option.
2939 67. If what should have been the first header line of a message consisted of
2940 a space followed by a colon, Exim was mis-interpreting it as a header line.
2941 It isn't of course - it is syntactically invalid and should therefore be
2942 treated as the start of the message body. The misbehaviour could have
2943 caused a number of strange effects, including loss of data in subsequent
2944 header lines, and spool format errors.
2946 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
2947 client host had authenticated. This control can now be exercised by an ACL
2948 for more flexibility.
2950 69. By default, callouts do not happen when testing with -bh. There is now a
2951 variant, -bhc, which does actually run the callout code, including
2952 consulting and updating the callout cache.
2954 70. Added support for saslauthd authentication, courtesy of Alexander
2957 71. If statvfs() failed on the spool or log directories while checking their
2958 size for availability, Exim confusingly gave the error "space shortage".
2959 Furthermore, in debugging mode it crashed with a floating point exception.
2960 These checks are done if check_{spool,log}_{space,inodes} are set, and when
2961 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
2962 serious problem, Exim now writes to the main and panic logs when this
2963 happens, with details of the failure. It then refuses to accept the
2964 incoming message, giving the message "spool directory problem" or "log
2965 directory problem" with a 421 code for SMTP messages.
2967 72. When Exim is about to re-exec itself, it ensures that the file descriptors
2968 0, 1, and 2 exist, because some OS complain for execs without them (see
2969 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
2970 descriptors. However, the code omitted to check that the open succeeded,
2971 causing mysterious errors if for some reason the permissions on /dev/null
2972 got screwed. Now Exim writes a message to the main and panic logs, and
2973 bombs out if it can't open /dev/null.
2975 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
2976 interact so that it is all more flexible. It is supposed to remain
2977 backwards compatible. Also added extra_local_interfaces.
2979 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
2980 to bomb out with an assertion failure - to the client this appears as a
2981 connection drop. This problem occurs in the part of the code that was taken
2982 from the Samba project. Fortunately, the assertion is in a very simple
2983 function, so I have fixed this by reproducing the function inline in the
2984 one place where it is called, and arranging for authentication to fail
2985 instead of killing the process with assert().
2987 75. The SPA client code was not working when the server requested OEM rather
2988 than Unicode encoding.
2990 76. Added code to make require_files with a specific uid setting more usable in
2991 the case where statting the file as root fails - usually a non-root-mounted
2992 NFS file system. When this happens and the failure is EACCES, Exim now
2993 forks a subprocess and does the per-uid checking as the relevant uid.
2995 77. Added process_log_path.
2997 78. If log_file_path was not explicitly set, a setting of check_log_space or
2998 check_log_inodes was ignored.
3000 79. If a space check for the spool or log partitions fails, the incident is now
3001 logged. Of course, in the latter case the data may get lost...
3003 80. Added the %p formatting code to string_format() so that it can be used to
3004 print addresses in debug_print(). Adjusted all the address printing in the
3005 debugging in store.c to use %p rather than %d.
3007 81. There was a concern that a line of code in smtp_in.c could overflow a
3008 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3009 initially expressed, the concern was not well-founded, because trailing
3010 spaces are removed early. However, if the trailing spaces were followed by
3011 a NULL, they did not get removed, so the overflow was possible. Two fixes
3014 (a) I re-wrote the offending code in a cleaner fashion.
3015 (b) If an incoming SMTP command contains a NULL character, it is rejected
3018 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3019 runs initgroups(), so that if the Exim user is in any additional groups,
3020 they will be used during message reception.
3026 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob