1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.165 2005/06/20 13:58:22 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
10 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
12 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
13 command itself ended in error, the underlying message about the transport
14 filter timeout was being overwritten with the pipe command error. Now the
15 underlying error message should be appended to the second error message.
17 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
19 PH/02 Reduce the amount of output that "make" produces by default. Full output
20 can still be requested.
22 PH/03 The warning log line about a condition test deferring for a "warn" verb
23 was being output only once per connection, rather than after each
24 occurrence (because it was using the same function as for successful
25 "warn" verbs). This seems wrong, so I have changed it.
27 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
28 it should not have, which might have caused a crash in the right
29 circumstances, but probably never did.
31 PH/04 Installed a modified version of Tony Finch's patch to make submission
32 mode fix the return path as well as the Sender: header line, and to
33 add a /name= option so that you can make the user's friendly name appear
36 TF/03 Added the control = fakedefer ACL modifier.
38 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
39 Mark Lowes for thorough testing.
41 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
43 TK/03 Merged latest SRS patch from Miles Wilton.
45 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
46 with the definition in sysexits.h (which is #included earlier).
47 Fortunately, Exim does not actually use EX_OK. The code used to try to
48 preserve the sysexits.h value, by assumimg that macro definitions were
49 scanned for macro replacements. I have been disabused of this notion,
50 so now the code just undefines EX_OK before #including unistd.h.
52 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
53 in the smtp transport. When a block could not be written in a single
54 write() function, the timeout was being re-applied to each part-write.
55 This seems wrong - if the receiver was accepting one byte at a time it
56 would take for ever. The timeout is now adjusted when this happens. It
57 doesn't have to be particularly precise.
59 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
60 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
62 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
63 without a domain if the check to <postmaster@domain> fails.
65 SC/01 Eximstats: added -xls and the ability to specify output files
66 (patch written by Frank Heydlauf).
68 SC/02 Eximstats: use FileHandles for outputing results.
70 SC/03 Eximstats: allow any combination of xls, txt, and html output.
72 SC/04 Eximstats: fixed display of large numbers with -nvr option
74 SC/05 Eximstats: fixed merging of reports with empty tables.
76 SC/06 Eximstats: added the -include_original_destination flag
78 SC/07 Eximstats: removed tabs and trailing whitespace.
80 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
82 TK/06 MBOX spool code: Add real "From " MBOX separator line
83 so the .eml file is really in mbox format (even though
84 most programs do not really care). Patch from Alex Miller.
86 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
87 The latter is generated from $received_to and is only set if the
88 message has one envelope recipient. SA can use these headers,
89 obviously out-of-the-box. Patch from Alex Miller.
91 PH/08 The ${def test on a variable was returning false if the variable's
92 value was "0", contrary to what the specification has always said!
93 The result should be true unless the variable is empty.
95 PH/09 The syntax error of a character other than { following "${if
96 def:variable_name" (after optional whitespace) was not being diagnosed.
97 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
98 accidental colon was present, for example, could give incorrect results.
100 PH/10 Tidied the code in a number of places where the st_size field of a stat()
101 result is used (not including appendfile, where other changes are about
104 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
105 This involved changing a lot of size variables from int to off_t. It
106 should work with maildirs and everything.
108 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
109 spamd dying while we are connected to it.
111 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
114 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
115 function for BDB 4.3.
117 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
119 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
120 the sources for PCRE have been re-arranged and more files are now
123 PH/15 The code I had for printing potentially long long variables in PH/11
124 above was not the best (it lost precision). The length of off_t variables
125 is now inspected at build time, and an appropriate printing format (%ld
126 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
127 to be "long long int" or "long int". This is needed for the internal
128 formatting function string_vformat().
130 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
131 the configuration file to be ":syslog", then the script "guesses" where
132 the logs files are, rather than using the compiled in default. In our
133 case the guess is not the same as the compiled default, so the script
134 suddenly stopped working when I started to use syslog. The patch checks
135 to see if log_file_path is "". If so, it attempts to read it from exim
136 with no configuration file to get the compiled in version, before it
137 falls back to the previous guessing code."
139 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
140 implementing BATV in an Exim configuration. See NewStuff for the gory
143 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
144 Makefile that are specific to HP-UX.
146 PH/18 If the "use_postmaster" option was set for a recipient callout together
147 with the "random" option, the postmaster address was used as the MAIL
148 FROM address for the random test, but not for the subsequent recipient
149 test. It is now used for both.
151 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
152 patch removes a few documentation additions to RFC 3028, because the
153 latest draft now contains them. It adds the new en;ascii-case comparator
154 and a new error check for 8bit text in MIME parts. Comparator and
155 require names are now matched exactly. I enabled the subaddress
156 extension, but it is not well tested yet (read: it works for me)."
158 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
159 rework some of the code of TK/09 above to avoid the hardwired use of
160 "%lld" and "long long". Replaced the call to snprintf() with a call to
163 PH/21 Added another message to those in 4.51/PH/42, namely "All relevant MX
164 records point to non-existent hosts".
166 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
167 experimental DomainKeys support:
169 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
170 (2) On an error such as an illegally used "control", the wrong name for
171 the control was given.
173 These problems did NOT occur unless DomainKeys support was compiled.
179 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
180 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
182 TK/02 Fix ACL "control" statement not being available in MIME ACL.
184 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
186 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
187 to test Sieve filters that use "vacation".
189 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
190 that changes the way the GnuTLS parameters are stored in the cache file.
191 The new format can be generated externally. For backward compatibility,
192 if the data in the cache doesn't make sense, Exim assumes it has read an
193 old-format file, and it generates new data and writes a new file. This
194 means that you can't go back to an older release without removing the
197 PH/03 A redirect router that has both "unseen" and "one_time" set does not
198 work if there are any delivery delays because "one_time" forces the
199 parent to be marked "delivered", so its unseen clone is never tried
200 again. For this reason, Exim now forbids the simultaneous setting of
203 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
204 redirected to themselves ("homonym" addresses). Read the long ChangeLog
205 entry if you want to know the details. The fix, however, neglected to
206 consider the case when local delivery batching is involved. The test for
207 "previously delivered" was not happening when checking to see if an
208 address could be batched with a previous (undelivered) one; under
209 certain circumstances this could lead to multiple deliveries to the same
212 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
213 in its include files, and this causes problems building Exim.
215 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
216 header_syntax) but Exim was just ignoring anything given after a slash.
217 In particular, this caused confusion with an attempt to use "verify =
218 reverse_host_lookup/defer_ok". An error is now given when options are
219 supplied for verify items that do not have them. (Maybe reverse_host_
220 lookup should have a defer_ok option, but that's a different point.)
222 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
223 defined by RFC 821) to 2048, because there were problems with some AUTH
224 commands, and RFC 1869 says the size should be increased for extended
225 SMTP commands that take arguments.
227 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
230 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
231 "unknown" error; now it says that the functionality isn't in the binary.
233 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
234 an address' error message when a string expansion fails (syntax or
235 whatever). Otherwise the password may appear in the log. Following change
236 PH/42 below, there is no longer a chance of it appearing in a bounce
239 PH/11 Installed exipick version 20050225.0 from John Jetmore.
241 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
242 of its addresses was ever tried. (Bugzilla bug #2.)
244 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
245 the result incorrectly in the debug output. (It correctly added a newline
246 to what was transported.)
248 TF/01 Added $received_time.
250 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
251 commented out examples of how to interface to a virus scanner and to
252 SpamAssassin. Also added commented examples of av_scanner and
253 spamd_address settings.
255 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
256 and controls are allowed in which ACLs. There were a couple of minor
257 errors. Some of the entries in the conditions table (which is a table of
258 where they are NOT allowed) were getting very unwieldy; rewrote them as a
259 negation of where the condition IS allowed.
261 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
263 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
264 header file does not have a version number, so I've had to invent a new
265 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
266 API. The code is untested by me (my Linux distribution still has 0.3.2 of
267 radiusclient), but it was contributed by a Radius user.
269 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
270 files or directories, for OpenSSL.
272 PH/19 When an Exim process that is running as root has to create an Exim log
273 file, it does so in a subprocess that runs as exim:exim so as to get the
274 ownership right at creation (otherwise, other Exim processes might see
275 the file with the wrong ownership). There was no test for failure of this
276 fork() call, which would lead to the process getting stuck as it waited
277 for a non-existent subprocess. Forks do occasionally fail when resources
278 run out. I reviewed all the other calls to fork(); they all seem to check
281 PH/20 When checking for unexpected SMTP input at connect time (before writing
282 the banner), Exim was not dealing correctly with a non-positive return
283 from the read() function. If the client had disconnected by this time,
284 the result was a log entry for a synchronization error with an empty
285 string after "input=" when read() returned zero. If read() returned -1
286 (an event I could not check), uninitialized data bytes were printed.
287 There were reports of junk text (parts of files, etc) appearing after
290 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
292 PH/22 Added support for macro redefinition, and (re)definition in between
293 driver and ACL definitions.
295 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
296 forgetting to use the resulting value; it was using the unexpanded value.
298 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
299 hadn't been configured. The fix is from Juergen Kreileder, who
300 understands it better than I do:
302 "Here's what I see happening with three configured cyrus_sasl
303 authenticators configured (plain, login, cram-md5):
305 On startup auth_cyrus_sasl_init() gets called for each of these.
306 This means three calls to sasl_listmech() without a specified mech_list.
307 => SASL tests which mechs of all available mechs actually work
308 => three warnings about OTP not working
309 => the returned list contains: plain, login, cram-md5, digest-md5, ...
311 With the patch, sasl_listmech() also gets called three times. But now
312 SASL's mech_list option is set to the server_mech specified in the the
313 authenticator. Or in other words, the answer from sasl_listmech()
314 gets limited to just the mech you're testing for (which is different
316 => the return list contains just 'plain' or 'login', 'cram-md5' or
317 nothing depending on the value of ob->server_mech.
319 I've just tested the patch: Authentication still works fine,
320 unavailable mechs specified in the exim configuration are still
321 caught, and the auth.log warnings about OTP are gone."
323 PH/25 When debugging is enabled, the contents of the command line are added
324 to the debugging output, even when log_selector=+arguments is not
327 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
328 answer is "GNU", and only if the return is "GNU/something" is the answer
331 PH/27 $acl_verify_message is now set immediately after the failure of a
332 verification in an ACL, and so is available in subsequent modifiers. In
333 particular, the message can be preserved by coding like this:
335 warn !verify = sender
336 set acl_m0 = $acl_verify_message
338 Previously, $acl_verify_message was set only while expanding "message"
339 and "log_message" when a very denied access.
341 PH/28 Modified OS/os.c-Linux with
343 -#ifndef OS_LOAD_AVERAGE
344 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
346 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
347 nomenclature these days.)
349 PH/29 Installed patch from the Sieve maintainer that adds the options
350 sieve_useraddress and sieve_subaddress to the redirect router.
352 PH/30 In these circumstances:
353 . Two addresses routed to the same list of hosts;
354 . First host does not offer TLS;
355 . First host accepts first address;
356 . First host gives temporary error to second address;
357 . Second host offers TLS and a TLS session is established;
358 . Second host accepts second address.
359 Exim incorrectly logged both deliveries with the TLS parameters (cipher
360 and peerdn, if requested) that were in fact used only for the second
363 PH/31 When doing a callout as part of verifying an address, Exim was not paying
364 attention to any local part prefix or suffix that was matched by the
365 router that accepted the address. It now behaves in the same way as it
366 does for delivery: the affixes are removed from the local part unless
367 rcpt_include_affixes is set on the transport.
369 PH/32 Add the sender address, as F=<...>, to the log line when logging a
370 timeout during the DATA phase of an incoming message.
372 PH/33 Sieve envelope tests were broken for match types other than :is. I have
373 applied a patch sanctioned by the Sieve maintainer.
375 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
376 the uid or gid is negative. A case of a negative gid caused this to be
377 noticed. The fix allows for either to be negative.
379 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
380 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
383 PH/36 The change PH/12 above was broken. Fixed it.
385 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
386 the grounds that routing the same address twice would always produce the
387 same answer. This might have been true once, but it is certainly no
388 longer true now. Routing a child address may depend on the previous
389 routing that produced that child. Some complicated redirection strategies
390 went wrong when messages had multiple recipients, and made Exim's
391 behaviour dependent on the order in which the addresses were given.
393 I have moved the duplicate checking until after the routing is complete.
394 Exim scans the addresses that are assigned to local and remote
395 transports, and removes any duplicates. This means that more work will be
396 done, as duplicates will always all be routed, but duplicates are
397 presumably rare, so I don't expect this is of any significance.
399 For deliveries to pipes, files, and autoreplies, the duplicate checking
400 still happens during the routing process, since they are not going to be
403 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
404 It corrects a timeout issue with spamd. This is Ian's comment: "The
405 background is that sometimes spamd either never reads data from a
406 connection it has accepted, or it never writes response data. The exiscan
407 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
408 blindly assumes that writes won't block so it may never time out."
410 PH/39 Allow G after quota size as well as K and M.
412 PH/40 The value set for $authenticated_id in an authenticator may not contain
413 binary zeroes or newlines because the value is written to log lines and
414 to spool files. There was no check on this. Now the value is run through
415 the string_printing() function so that such characters are converted to
416 printable escape sequences.
418 PH/41 $message_linecount is a new variable that contains the total number of
419 lines in the message. Compare $body_linecount, which is the count for the
422 PH/42 Exim no longer gives details of delivery errors for specific addresses in
423 bounce and delay warning messages, except in certain special cases, which
426 (a) An SMTP error message from a remote host;
427 (b) A message specified in a :fail: redirection;
428 (c) A message specified in a "fail" command in a system filter;
429 (d) A message specified in a FAIL return from the queryprogram router;
430 (e) A message specified by the cannot_route_message router option.
432 In these cases only, Exim does include the error details in bounce and
433 warning messages. There are also a few cases where bland messages such
434 as "unrouteable address" or "local delivery error" are given.
436 PH/43 $value is now also set for the "else" part of a ${run expansion.
438 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
439 being worked on, but at least Exim now implements the latest version to
442 PH/45 In a pipe transport, although a timeout while waiting for the pipe
443 process to complete was treated as a delivery failure, a timeout while
444 writing the message to the pipe was logged, but erroneously treated as a
445 successful delivery. Such timeouts include transport filter timeouts. For
446 consistency with the overall process timeout, these timeouts are now
447 treated as errors, giving rise to delivery failures by default. However,
448 there is now a new Boolean option for the pipe transport called
449 timeout_defer, which, if set TRUE, converts the failures into defers for
450 both kinds of timeout. A transport filter timeout is now identified in
453 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
454 systems where "make" and "gmake" are different, calling "gmake" at top
455 level broke things. I've arranged for the value of $(MAKE) to be passed
456 from the Makefile to this script so that it can call the same version of
460 A note about Exim versions 4.44 and 4.50
461 ----------------------------------------
463 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
464 changes of various kinds. As a consequence, a big documentation update was
465 needed. This delayed the release for rather longer than seemed good, especially
466 in the light of a couple of (minor) security issues. Therefore, the changes
467 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
468 release. So 4.44 and 4.50 are in effect two different branches that both start
471 I have left the 4.50 change log unchanged; it contains all the changes since
472 4.43. The change log for 4.44 is below; many of its items are identical to
473 those for 4.50. This seems to be the most sensible way to preserve the
474 historical information.
480 1. Minor wording change to the doc/README.SIEVE file.
482 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
483 computation of the current number of files was incorrect.
485 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
486 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
489 4. Give more explanation in the error message when the command for a transport
490 filter fails to execute.
492 5. There are several places where Exim runs a non-Exim command in a
493 subprocess. The SIGUSR1 signal should be disabled for these processes. This
494 was being done only for the command run by the queryprogram router. It is
495 now done for all such subprocesses. The other cases are: ${run, transport
496 filters, and the commands run by the lmtp and pipe transports.
498 6. Added CONFIGURE_GROUP build-time option.
500 7. Some older OS have a limit of 256 on the maximum number of file
501 descriptors. Exim was using setrlimit() to set 1000 as a large value
502 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
503 systems. I've change it so that if it can't get 1000, it tries for 256.
505 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
506 was an oversight, and furthermore, ever since the addition of extra
507 controls (e.g. 4.43/32), the checks on when to allow different forms of
508 "control" were broken. There should now be diagnostics for all cases when a
509 control that does not make sense is encountered.
511 9. Added the /retain_sender option to "control=submission".
513 10. $recipients is now available in the predata ACL (oversight).
515 11. Tidy the search cache before the fork to do a delivery from a message
516 received from the command line. Otherwise the child will trigger a lookup
517 failure and thereby defer the delivery if it tries to use (for example) a
518 cached ldap connection that the parent has called unbind on.
520 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
521 of $address_data from the recipient verification was clobbered by the
524 13. The value of address_data from a sender verification is now available in
525 $sender_address_data in subsequent conditions in the ACL statement.
527 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
529 15. Added a new option "connect=<time>" to callout options, to set a different
532 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
533 was its contents. (It was OK if the option was not defined at all.)
535 17. A "Completed" log line is now written for messages that are removed from
536 the spool by the -Mrm option.
538 18. New variables $sender_verify_failure and $recipient_verify_failure contain
539 information about exactly what failed.
541 19. Added -dd to debug only the daemon process.
543 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
544 handles timeouts, both on the server side and network timeouts. Renamed the
545 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
547 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
548 It is now set to "smtps".
550 22. $host_address is now set to the target address during the checking of
553 23. When checking ignore_target_hosts for an ipliteral router, no host name was
554 being passed; this would have caused $sender_host_name to have been used if
555 matching the list had actually called for a host name (not very likely,
556 since this list is usually IP addresses). A host name is now passed as
559 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
560 code that specifies a non-restarting handler (typically sigaction() in
561 modern systems) in an attempt to fix a rare and obscure crash bug.
563 25. Narrowed the window for a race in the daemon that could cause it to ignore
564 SIGCHLD signals. This is not a major problem, because they are used only to
565 wake it up if nothing else does.
567 26. A malformed maildirsize file could cause Exim to calculate negative values
568 for the mailbox size or file count. Odd effects could occur as a result.
569 The maildirsize information is now recalculated if the size or filecount
572 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
573 support for a long time. Removed HAVE_SYS_VFS_H.
575 28. Installed the latest version of exipick from John Jetmore.
577 29. In an address list, if the pattern was not a regular expression, an empty
578 subject address (from a bounce message) matched only if the pattern was an
579 empty string. Non-empty patterns were not even tested. This was the wrong
580 because it is perfectly reasonable to use an empty address as part of a
581 database query. An empty address is now tested by patterns that are
582 lookups. However, all the other forms of pattern expect the subject to
583 contain a local part and a domain, and therefore, for them, an empty
584 address still always fails if the pattern is not itself empty.
586 30. Exim went into a mad DNS loop when attempting to do a callout where the
587 host was specified on an smtp transport, and looking it up yielded more
590 31. Re-factored the code for checking spool and log partition space into a
591 function that finds that data and another that does the check. The former
592 is then used to implement four new variables: $spool_space, $log_space,
593 $spool_inodes, and $log_inodes.
595 32. The RFC2047 encoding function was originally intended for short strings
596 such as real names; it was not keeping to the 75-character limit for
597 encoded words that the RFC imposes. It now respects the limit, and
598 generates multiple encoded words if necessary. To be on the safe side, I
599 have increased the buffer size for the ${rfc2047: expansion operator from
602 33. It is now permitted to omit both strings after an "if" condition; if the
603 condition is true, the result is "true". As before, when the second string
604 is omitted, a false condition yields an empty string. This makes it less
605 cumbersome to write custom ACL and router conditions.
607 34. Failure to deliver a bounce message always caused it to be frozen, even if
608 there was an errors_to setting on the router. The errors_to setting is now
611 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
612 canonical form (fully expanded) before being placed in
613 $sender_host_address.
615 36. The table in the code that translates DNS record types into text (T_A to
616 "A" for instance) was missing entries for NS and CNAME. It is just possible
617 that this could have caused confusion if both these types were looked up
618 for the same domain, because the text type is used as part of Exim's
619 per-process caching. But the chance of anyone hitting this buglet seems
622 37. The dnsdb lookup has been extended in a number of ways.
624 (1) There is a new type, "zns", which walks up the domain tree until it
625 finds some nameserver records. It should be used with care.
627 (2) There is a new type, "mxh", which is like "mx" except that it returns
628 just the host names, not the priorities.
630 (3) It is now possible to give a list of domains (or IP addresses) to be
631 looked up. The behaviour when one of the lookups defers can be
632 controlled by a keyword.
634 (4) It is now possible to specify the separator character for use when
635 multiple records are returned.
637 38. The dnslists ACL condition has been extended: it is now possible to supply
638 a list of IP addresses and/or domains to be looked up in a particular DNS
641 39. Added log_selector=+queue_time_overall.
643 40. When running the queue in the test harness, wait just a tad after forking a
644 delivery process, to get repeatability of debugging output.
646 41. Include certificate and key file names in error message when GnuTLS fails
647 to set them up, because the GnuTLS error message doesn't include the name
648 of the failing file when there is a problem reading it.
650 42. Allow both -bf and -bF in the same test run.
652 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
654 44. The "Exiscan patch" is now merged into the mainline Exim source.
656 45. Sometimes the final signoff response after QUIT could fail to get
657 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
658 < 0 before doing a fflush(). This bug looks as though it goes back to the
659 introduction of TLS in release 3.20, but "sometimes" must have been rare
660 because the tests only now provoked it.
662 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
663 (this can affect the format of dates).
665 47. exim_tidydb, when checking for the continued existence of a message for
666 which it has found a message-specific retry record, was not finding
667 messages that were in split spool directories. Consequently, it was
668 deleting retry records that should have stayed in existence.
670 48. Steve fixed some bugs in eximstats.
672 49. The SPA authentication driver was not abandoning authentication and moving
673 on to the next authenticator when an expansion was forced to fail,
674 contradicting the general specification for all authenticators. Instead it
675 was generating a temporary error. It now behaves as specified.
677 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
678 (the order specifies the preference for clients). The order is now AES256,
679 AES128, 3DES, ARCFOUR128.
681 51. Small patch to Sieve code - explicitly set From: when generating an
684 52. Exim crashed if a remote delivery caused a very long error message to be
685 recorded - for instance if somebody sent an entire SpamAssassin report back
686 as a large number of 550 error lines. This bug was coincidentally fixed by
687 increasing the size of one of Exim's internal buffers (big_buffer) that
688 happened as part of the Exiscan merge. However, to be on the safe side, I
689 have made the code more robust (and fixed the comments that describe what
692 53. Now that there can be additional text after "Completed" in log lines (if
693 the queue_time_overall log selector is set), a one-byte patch to exigrep
694 was needed to allow it to recognize "Completed" as not the last thing in
697 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
698 patch that reportedly fixes this has been added. I am not expert enough to
699 create a test for it. This is what the patch creator wrote:
701 "I found a little strange behaviour of ldap code when working with
702 Windows 2003 AD Domain, where users was placed in more than one
703 Organization Units. When I tried to give exim partial DN, the exit code
704 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
705 But simultaneously result of request was absolutely normal ldap result,
706 so I produce this patch..."
708 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
709 REFERENCE, so I have modified the code to exclude the patch when that macro
712 55. Some experimental protocols are using DNS PTR records for new purposes. The
713 keys for these records are domain names, not reversed IP addresses. The
714 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
715 leaves it alone. Component reversal etc. now happens only for IP addresses.
718 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
720 57. Double the size of the debug message buffer (to 2048) so that more of very
721 long debug lines gets shown.
723 58. The exicyclog utility now does better if the number of log files to keep
724 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
726 59. Two changes related to the smtp_active_hostname option:
728 (1) $smtp_active_hostname is now available as a variable.
729 (2) The default for smtp_banner uses $smtp_active_hostname instead
730 of $primary_hostname.
732 60. The host_aton() function is supposed to be passed a string that is known
733 to be a valid IP address. However, in the case of IPv6 addresses, it was
734 not checking this. This is a hostage to fortune. Exim now panics and dies
735 if the condition is not met. A case was found where this could be provoked
736 from a dnsdb PTR lookup with an IPv6 address that had more than 8
737 components; fortuitously, this particular loophole had already been fixed
738 by change 4.50/55 above.
740 If there are any other similar loopholes, the new check in host_aton()
741 itself should stop them being exploited. The report I received stated that
742 data on the command line could provoke the exploit when Exim was running as
743 exim, but did not say which command line option was involved. All I could
744 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
748 61. There was a buffer overflow vulnerability in the SPA authentication code
749 (which came originally from the Samba project). I have added a test to the
750 spa_base64_to_bits() function which I hope fixes it.
753 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
754 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
756 63. The daemon start-up calls getloadavg() while still root for those OS that
757 need the first call to be done as root, but it missed one case: when
758 deliver_queue_load_max is set with deliver_drop_privilege. This is
759 necessary for the benefit of the queue runner, because there is no re-exec
760 when deliver_drop_privilege is set.
762 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
765 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
766 was in use, was not putting the data itself into the right store pool;
767 consequently, it could be overwritten for a subsequent message in the same
768 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
771 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
773 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
774 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
775 the function were treating the return as a boolean value, which happened to
776 work because 0=false and not-0=true, but is not correct code.
778 68. The host_aton() function was not handling scoped IPv6 addresses (those
779 with, for example, "%eth0" on the end) correctly.
781 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
782 negated items (that is, ~something) in unsigned ints. Some compilers
783 apparently mutter when there is no cast.
785 70. If an address verification called from an ACL failed, and did not produce a
786 user-specific message (i.e. there was only a "system" message), nothing was
787 put in $acl_verify_message. In this situation, it now puts the system
790 71. Change 4.23/11 added synchronization checking at the start of an SMTP
791 session; change 4.31/43 added the unwanted input to the log line - except
792 that it did not do this in the start of session case. It now does.
794 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
795 This is wrong and can cause the other end to generate a synchronization
796 error if it is another Exim or anything else that does the synchronization
797 check. A QUIT command is no longer sent after a timeout.
799 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
802 74. The defer_ok option of callout verification was not working if it was used
803 when verifying addresses in header lines, that is, for this case:
805 verify = header_sender/callout=defer_ok
807 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
808 those file descriptors could be used for SMTP connections. If anything
809 wrote to stderr (the example that came up was "warn" in embedded Perl), it
810 could be sent to the SMTP client, causing chaos. The daemon now opens
811 stdin, stdout, and stderr to /dev/null when it puts itself into the
814 76. Arrange for output from Perl's "warn" command to be written to Exim's main
815 log by default. The user can override this with suitable Perl magic.
817 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
818 the log message when discard triggers, was not working for the DATA ACL or
819 for the non-SMTP ACL.
821 78. Error message wording change in sieve.c.
823 79. If smtp_accept_max_per_host was set, the number of connections could be
824 restricted to fewer than expected, because the daemon was trying to set up
825 a new connection before checking whether the processes handling previous
826 connections had finished. The check for completed processes is now done
827 earlier. On busy systems, this bug wouldn't be noticed because something
828 else would have woken the daemon, and it would have reaped the completed
831 80. If a message was submitted locally by a user whose login name contained one
832 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
833 It caused a spool format error. I have fixed the spool reading code. A
834 related problem was that the "from" clause in the Received: line became
835 illegal because of the space(s). It is now covered by ${quote_local_part.
837 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
840 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
842 83. Patch from Sieve maintainer to fix unterminated string problem in
845 84. Some minor changes to the Linux configuration files to help with other
846 OS variants using glibc.
848 85. One more patch for Sieve to update vacation handling to latest spec.
851 ----------------------------------------------------
852 See the note above about the 4.44 and 4.50 releases.
853 ----------------------------------------------------
859 1. Change 4.43/35 introduced a bug that caused file counts to be
860 incorrectly computed when quota_filecount was set in an appendfile
863 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
864 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
867 3. Give more explanation in the error message when the command for a transport
868 filter fails to execute.
870 4. There are several places where Exim runs a non-Exim command in a
871 subprocess. The SIGUSR1 signal should be disabled for these processes. This
872 was being done only for the command run by the queryprogram router. It is
873 now done for all such subprocesses. The other cases are: ${run, transport
874 filters, and the commands run by the lmtp and pipe transports.
876 5. Some older OS have a limit of 256 on the maximum number of file
877 descriptors. Exim was using setrlimit() to set 1000 as a large value
878 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
879 systems. I've change it so that if it can't get 1000, it tries for 256.
881 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
882 was an oversight, and furthermore, ever since the addition of extra
883 controls (e.g. 4.43/32), the checks on when to allow different forms of
884 "control" were broken. There should now be diagnostics for all cases when a
885 control that does not make sense is encountered.
887 7. $recipients is now available in the predata ACL (oversight).
889 8. Tidy the search cache before the fork to do a delivery from a message
890 received from the command line. Otherwise the child will trigger a lookup
891 failure and thereby defer the delivery if it tries to use (for example) a
892 cached ldap connection that the parent has called unbind on.
894 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
895 of $address_data from the recipient verification was clobbered by the
898 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
899 was its contents. (It was OK if the option was not defined at all.)
901 11. A "Completed" log line is now written for messages that are removed from
902 the spool by the -Mrm option.
904 12. $host_address is now set to the target address during the checking of
907 13. When checking ignore_target_hosts for an ipliteral router, no host name was
908 being passed; this would have caused $sender_host_name to have been used if
909 matching the list had actually called for a host name (not very likely,
910 since this list is usually IP addresses). A host name is now passed as
913 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
914 code that specifies a non-restarting handler (typically sigaction() in
915 modern systems) in an attempt to fix a rare and obscure crash bug.
917 15. Narrowed the window for a race in the daemon that could cause it to ignore
918 SIGCHLD signals. This is not a major problem, because they are used only to
919 wake it up if nothing else does.
921 16. A malformed maildirsize file could cause Exim to calculate negative values
922 for the mailbox size or file count. Odd effects could occur as a result.
923 The maildirsize information is now recalculated if the size or filecount
926 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
927 support for a long time. Removed HAVE_SYS_VFS_H.
929 18. Updated exipick to current release from John Jetmore.
931 19. Allow an empty sender to be matched against a lookup in an address list.
932 Previously the only cases considered were a regular expression, or an
935 20. Exim went into a mad DNS lookup loop when doing a callout where the
936 host was specified on the transport, if the DNS lookup yielded more than
939 21. The RFC2047 encoding function was originally intended for short strings
940 such as real names; it was not keeping to the 75-character limit for
941 encoded words that the RFC imposes. It now respects the limit, and
942 generates multiple encoded words if necessary. To be on the safe side, I
943 have increased the buffer size for the ${rfc2047: expansion operator from
946 22. Failure to deliver a bounce message always caused it to be frozen, even if
947 there was an errors_to setting on the router. The errors_to setting is now
950 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
951 canonical form (fully expanded) before being placed in
952 $sender_host_address.
954 24. Updated eximstats to version 1.33
956 25. Include certificate and key file names in error message when GnuTLS fails
957 to set them up, because the GnuTLS error message doesn't include the name
958 of the failing file when there is a problem reading it.
960 26. Expand error message when OpenSSL has problems setting up cert/key files.
963 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
964 (this can affect the format of dates).
966 28. exim_tidydb, when checking for the continued existence of a message for
967 which it has found a message-specific retry record, was not finding
968 messages that were in split spool directories. Consequently, it was
969 deleting retry records that should have stayed in existence.
971 29. eximstats updated to version 1.35
972 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
973 1.35 - bugfix such that pie charts by volume are generated correctly
975 30. The SPA authentication driver was not abandoning authentication and moving
976 on to the next authenticator when an expansion was forced to fail,
977 contradicting the general specification for all authenticators. Instead it
978 was generating a temporary error. It now behaves as specified.
980 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
981 (the order specifies the preference for clients). The order is now AES256,
982 AES128, 3DES, ARCFOUR128.
984 31. Small patch to Sieve code - explicitly set From: when generating an
987 32. Exim crashed if a remote delivery caused a very long error message to be
988 recorded - for instance if somebody sent an entire SpamAssassin report back
989 as a large number of 550 error lines. This bug was coincidentally fixed by
990 increasing the size of one of Exim's internal buffers (big_buffer) that
991 happened as part of the Exiscan merge. However, to be on the safe side, I
992 have made the code more robust (and fixed the comments that describe what
995 33. Some experimental protocols are using DNS PTR records for new purposes. The
996 keys for these records are domain names, not reversed IP addresses. The
997 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
998 leaves it alone. Component reversal etc. now happens only for IP addresses.
1001 34. The host_aton() function is supposed to be passed a string that is known
1002 to be a valid IP address. However, in the case of IPv6 addresses, it was
1003 not checking this. This is a hostage to fortune. Exim now panics and dies
1004 if the condition is not met. A case was found where this could be provoked
1005 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1006 components; fortuitously, this particular loophole had already been fixed
1007 by change 4.50/55 or 4.44/33 above.
1009 If there are any other similar loopholes, the new check in host_aton()
1010 itself should stop them being exploited. The report I received stated that
1011 data on the command line could provoke the exploit when Exim was running as
1012 exim, but did not say which command line option was involved. All I could
1013 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1014 running as the user.
1017 35. There was a buffer overflow vulnerability in the SPA authentication code
1018 (which came originally from the Samba project). I have added a test to the
1019 spa_base64_to_bits() function which I hope fixes it.
1022 36. The daemon start-up calls getloadavg() while still root for those OS that
1023 need the first call to be done as root, but it missed one case: when
1024 deliver_queue_load_max is set with deliver_drop_privilege. This is
1025 necessary for the benefit of the queue runner, because there is no re-exec
1026 when deliver_drop_privilege is set.
1028 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1029 was in use, was not putting the data itself into the right store pool;
1030 consequently, it could be overwritten for a subsequent message in the same
1031 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1034 38. Sometimes the final signoff response after QUIT could fail to get
1035 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1036 < 0 before doing a fflush(). This bug looks as though it goes back to the
1037 introduction of TLS in release 3.20, but "sometimes" must have been rare
1038 because the tests only now provoked it.
1044 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1045 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1046 Now it returns an integer. A number of calls were still expecting a T/F
1047 return. Fortuitously, in all cases, the tests worked in OK situations,
1048 which is the norm. However, things would have gone wrong on any write
1049 failures on the smtp file descriptor. This function is used when sending
1050 messages over SMTP and also when doing verify callouts.
1052 2. When Exim is called to do synchronous delivery of a locally submitted
1053 message (the -odf or -odi options), it no longer closes stderr before doing
1056 3. Implemented the mua_wrapper option.
1058 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1060 5. Implemented the functions header_remove(), header_testname(),
1061 header_add_at_position(), and receive_remove_recipient(), and exported them
1064 6. If an ACL "warn" statement specified the addition of headers, Exim already
1065 inserted X-ACL-Warn: at the start if there was no header name. However, it
1066 was not making this test for the second and subsequent header lines if
1067 there were newlines in the string. This meant that an invalid header could
1068 be inserted if Exim was badly configured.
1070 7. Allow an ACL "warn" statement to add header lines at the start or after all
1071 the Received: headers, as well as at the end.
1073 8. Added the rcpt_4xx retry error code.
1075 9. Added postmaster_mailfrom=xxx to callout verification option.
1077 10. Added mailfrom=xxxx to the callout verification option, for verify=
1080 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1081 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1083 12. Inserted some casts to stop certain compilers warning when using pointer
1084 differences as field lengths or precisions in printf-type calls (mostly
1085 affecting debugging statements).
1087 13. Added optional readline() support for -be (dynamically loaded).
1089 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1090 same clock tick as a message's arrival, so that its received time was the
1091 same as the "first fail" time on the retry record, and that message
1092 remained on the queue past the ultimate address timeout, every queue runner
1093 would try a delivery (because it was past the ultimate address timeout) but
1094 after another failure, the ultimate address timeout, which should have then
1095 bounced the address, did not kick in. This was a "< instead of <=" error;
1096 in most cases the first failure would have been in the next clock tick
1097 after the received time, and all would be well.
1099 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1100 being recognized when the domain list was tested by the match_domain
1101 condition in an expansion string.
1103 16. Added the ${str2b64: operator.
1105 17. Exim was always calling setrlimit() to set a large limit for the number of
1106 processes, without checking whether the existing limit was already
1107 adequate. (It did check for the limit on file descriptors.) Furthermore,
1108 errors from getrlimit() and setrlimit() were being ignored. Now they are
1109 logged to the main and panic logs, but Exim does carry on, to try to do its
1110 job under whatever limits there are.
1112 18. Imported PCRE 5.0.
1114 19. Trivial typo in log message " temporarily refused connection" (the leading
1117 20. If the log selector return_path_on_delivery was set and an address was
1118 redirected to /dev/null, the delivery process crashed because it assumed
1119 that a return path would always be set for a "successful" delivery. In this
1120 case, the whole delivery is bypassed as an optimization, and therefore no
1123 21. Internal re-arrangement: the function for sending a challenge and reading
1124 a response while authentication was assuming a zero-terminated challenge
1125 string. It's now changed to take a pointer and a length, to allow for
1126 binary data in such strings.
1128 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1130 23. Exim was not respecting finduser_retries when seeking the login of the
1131 uid under which it was called; it was always trying 10 times. (The default
1132 setting of finduser_retries is zero.) Also, it was sleeping after the final
1133 failure, which is pointless.
1135 24. Implemented tls_on_connect_ports.
1137 25. Implemented acl_smtp_predata.
1139 26. If the domain in control=submission is set empty, Exim assumes that the
1140 authenticated id is a complete email address when it generates From: or
1141 Sender: header lines.
1143 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1144 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1145 chown and chgrp in /bin and hostname in /usr/bin.
1147 28. Exim was keeping the "process log" file open after each use, just as it
1148 does for the main log. This opens the possibility of it remaining open for
1149 long periods when the USR1 signal hits a daemon. Occasional processlog
1150 errors were reported, that could have been caused by this. Anyway, it seems
1151 much more sensible not to leave this file open at all, so that is what now
1154 29. The long-running daemon process does not normally write to the log once it
1155 has entered its main loop, and it closes the log before doing so. This is
1156 so that log files can straightforwardly be renamed and moved. However,
1157 there are a couple of unusual error situations where the daemon does write
1158 log entries, and I had neglected to close the log afterwards.
1160 30. The text of an SMTP error response that was received during a remote
1161 delivery was being truncated at 512 bytes. This is too short for some of
1162 the long messages that one sometimes sees. I've increased the limit to
1165 31. It is now possible to make retry rules that apply only when a message has a
1166 specific sender, in particular, an empty sender.
1168 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1169 it possible to be selective about when SMTP synchronization is enforced.
1171 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1173 32. Implemented hosts_connection_nolog.
1175 33. Added an ACL for QUIT.
1177 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1180 35. Added mailbox_size and mailbox_filecount to appendfile.
1182 36. Added control = no_multiline_responses to ACLs.
1184 37. There was a bug in the logic of the code that waits for the clock to tick
1185 in the case where the clock went backwards by a substantial amount such
1186 that the microsecond fraction of "now" was more than the microsecond
1187 fraction of "then" (but the whole seconds number was less).
1189 38. Added support for the libradius Radius client library this is found on
1190 FreeBSD (previously only the radiusclient library was supported).
1196 1. When certain lookups returned multiple values in the form name=value, the
1197 quoting of the values was not always being done properly. Specifically:
1198 (a) If the value started with a double quote, but contained no whitespace,
1200 (b) If the value contained whitespace other than a space character (i.e.
1201 tabs or newlines or carriage returns) it was not quoted.
1202 This fix has been applied to the mysql and pgsql lookups by writing a
1203 separate quoting function and calling it from the lookup code. The fix
1204 should probably also be applied to nisplus, ibase and oracle lookups, but
1205 since I cannot test any of those, I have not disturbed their existing code.
1207 2. A hit in the callout cache for a specific address caused a log line with no
1208 reason for rejecting RCPT. Now it says "Previous (cached) callout
1209 verification failure".
1211 3. There was an off-by-one bug in the queryprogram router. An over-long
1212 return line was truncated at 256 instead of 255 characters, thereby
1213 overflowing its buffer with the terminating zero. As well as fixing this, I
1214 have increased the buffer size to 1024 (and made a note to document this).
1216 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1217 when Exim is waiting for an SMTP response from a remote server, Exim
1218 restarts its select() call on the socket, thereby resetting its timeout.
1219 This is not a problem when such interrupts are rare. Somebody set up a cron
1220 job to run exiwhat every 2 minutes, which is less than the normal select()
1221 timeout (5 or 10 minutes). This meant that the select() timeout never
1222 kicked in because it was always reset. I have fixed this by comparing the
1223 time when an interrupt arrives with the time at the start of the first call
1224 to select(). If more time than the timeout has elapsed, the interrupt is
1225 treated as a timeout.
1227 5. Some internal re-factoring in preparation for the addition of Sieve
1228 extensions (by MH). In particular, the "personal" test is moved to a
1229 separate function, and given an option for scanning Cc: and Bcc: (which is
1230 not set for Exim filters).
1232 6. When Exim created an email address using the login of the caller as the
1233 local part (e.g. when creating a From: or Sender: header line), it was not
1234 quoting the local part when it contained special characters such as @.
1236 7. Installed new OpenBSD configuration files.
1238 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1239 try to make them clearer.
1241 9. Callout options, other than the timeout value, were being ignored when
1242 verifying sender addresses in header lines. For example, when using
1244 verify = header_sender/callout=no_cache
1246 the cache was (incorrectly) being used.
1248 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1249 only the Cygwin environment.
1251 11. When return_path_on_delivery was set as a log selector, if different remote
1252 addresses in the same message used different return paths and parallel
1253 remote delivery occurred, the wrong values would sometimes be logged.
1254 (Whenever a remote delivery process finished, the return path value from
1255 the most recently started remote delivery process was logged.)
1257 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1258 lines when AUTH and/or TLS are in use. This is the "received protocol"
1259 field. Exim used to use "asmtp" for authenticated SMTP, without any
1260 indication (in the protocol name) for TLS use. Now it follows the RFC and
1261 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1262 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1263 names appear in log lines as well as in Received: header lines.
1265 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1266 extensions, and comparison tests, and to fix some bugs.
1268 14. Changes to the "personal" filter test:
1270 (1) The test was buggy in that it was just doing the equivalent of
1271 "contains" tests on header lines. For example, if a user's address was
1272 anne@some.where, the "personal" test would incorrectly be true for
1274 To: susanne@some.where
1276 This test is now done by extracting each address from the header in turn,
1277 and checking the entire address. Other tests that are part of "personal"
1278 are now done using regular expressions (for example, to check local parts
1279 of addresses in From: header lines).
1281 (2) The list of non-personal local parts in From: addresses has been
1282 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1283 taken from the Sieve specification recommendations.
1285 (3) If the message contains any header line starting with "List-" it is
1286 treated as non-personal.
1288 (4) The test for "circular" in the Subject: header line has been removed
1289 because it now seems ill-conceived.
1291 15. Minor typos in src/EDITME comments corrected.
1293 16. Installed latest exipick from John Jetmore.
1295 17. If headers_add on a router specified a text string that was too long for
1296 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1297 of string_sprintf() is now avoided.
1299 18. $message_body_size was not set (it was always zero) when running the DATA
1300 ACL and the local_scan() function.
1302 19. For the "mail" command in an Exim filter, no default was being set for
1303 the once_repeat time, causing a random time value to be used if "once" was
1304 specified. (If the value happened to be <= 0, no repeat happened.) The
1305 default is now 0s, meaning "never repeat". The "vacation" command was OK
1306 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1307 (I found it when inspecting the code).
1309 20. There is now an overall timeout for performing a callout verification. It
1310 defaults to 4 times the callout timeout, which applies to individual SMTP
1311 commands during the callout. The overall timeout applies when there is more
1312 than one host that can be tried. The timeout is checked before trying the
1313 next host. This prevents very long delays if there are a large number of
1314 hosts and all are timing out (e.g. when the network connections are timing
1315 out). The value of the overall timeout can be changed by specifying an
1316 additional sub-option for "callout", called "maxwait". For example:
1318 verify = sender/callout=5s,maxwait=20s
1320 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1321 to the end before writing, but this should make it even safer).
1323 22. Exim was forgetting that it had advertised PIPELINING for the second and
1324 subsequent messages on an SMTP connection. It was also not resetting its
1325 memory on STARTTLS and an internal HELO.
1327 23. When Exim logs an SMTP synchronization error within a session, it now
1328 records whether PIPELINING has been advertised or not.
1330 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1331 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1332 rather than long int.
1334 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1336 26. Added the never_mail option to autoreply.
1342 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1343 crash if the getsockname() call failed; this can happen if a connection is
1344 closed very soon after it is established. The problem was simply in the
1345 order in which certain operations were done, causing Exim to try to write
1346 to the SMTP stream before it had set up the file descriptor. The bug has
1347 been fixed by making things happen in the correct order.
1353 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1354 before the connection was closed, thus losing the rejection response.
1356 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1357 some early Solaris releases, but causes trouble in current releases where
1358 socklen_t is defined.
1360 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1363 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1366 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1367 that releasing the top of it at the end releases what was used for sub-
1368 expansions (unless the block got too big). However, discard this block if
1369 the first thing is a variable or header, so that we can use its block when
1370 it is dynamic (useful for very large $message_headers, for example).
1372 6. Lookups now cache *every* query, not just the most recent. A new, separate
1373 store pool is used for this. It can be recovered when all lookup caches are
1374 flushed. Lookups now release memory at the end of their result strings.
1375 This has involved some general refactoring of the lookup sources.
1377 7. Some code has been added to the store_xxx() functions to reduce the amount
1378 of flapping under certain conditions.
1380 8. log_incoming_interface used to affect only the <= reception log lines. Now
1381 it causes the local interface and port to be added to several more SMTP log
1382 lines, for example "SMTP connection from", and rejection lines.
1384 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1386 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1388 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1389 could be overwritten at the end of the current message (or the start of a
1390 new message if it was set in a HELO ACL). The value is now preserved for
1391 the duration of the SMTP connection.
1393 12. If a transport had a headers_rewrite setting, and a matching header line
1394 contained an unqualified address, that address was qualified, even if it
1395 did not match any rewriting rules. The underlying bug was that the values
1396 of the flags that permit the existence of unqualified sender and recipient
1397 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1398 non-local messages, and by -bnq for local messages) were not being
1399 preserved with the message after it was received.
1401 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1402 "next input=" as part of the text comprising the host identity instead of
1403 the correct text. The code was using the same buffer for two different
1404 strings. However, depending on which order the printing function evaluated
1405 its arguments, the bug did not always show up. Under Linux, for example, my
1406 test suite worked just fine.
1408 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1409 doesn't work with some older versions of Perl. It has been changed to "my",
1410 which in any case is probably the better facility to use.
1412 15. A really picky compiler found some instances of statements for creating
1413 error messages that either had too many or two few arguments for the format
1416 16. The size of the buffer for calls to the DNS resolver has been increased
1417 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1418 for addresses that have a lot of PTR records. This alleviates a problem; it
1419 does not fully solve it.
1421 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1422 buffer now truncates the list and logs the incident, which is the same
1423 action as happens when Exim is looking up a host name and its aliases.
1424 Previously in this situation something unpredictable would happen;
1425 sometimes it was "internal error: store_reset failed".
1427 18. If a server dropped the connection unexpectedly when an Exim client was
1428 using GnuTLS and trying to read a response, the client delivery process
1429 crashed while trying to generate an error log message.
1431 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1432 string, for example:
1434 warn message = H1: something\nH2: something
1436 the text was added as a single header line from Exim's point of view
1437 though it ended up OK in the delivered message. However, searching for the
1438 second and subsequent header lines using $h_h2: did not work. This has been
1439 fixed. Similarly, if a system filter added multiple headers in this way,
1440 the routers could not see them.
1442 20. Expanded the error message when iplsearch is called with an invalid key to
1443 suggest using net-iplsearch in a host list.
1445 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1446 ACLs are no longer actually imposed (and a message to that effect is
1449 22. If a "gecos" field in a passwd entry contained escaped characters, in
1450 particular, if it contained a \" sequence, Exim got it wrong when building
1451 a From: or a Sender: header from that name. A second bug also caused
1452 incorrect handling when an unquoted " was present following a character
1453 that needed quoting.
1455 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1456 was not being matched caselessly.
1458 24. Arranged for all hyphens in the exim.8 source to be escaped with
1461 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1462 properly. Recipient callout cache records were still being keyed to include
1463 the sender, even when use_sender was set false. This led to far more
1464 callouts that were necessary. The sender is no longer included in the key
1465 when use_sender is false.
1467 26. Added "control = submission" modifier to ACLs.
1469 27. Added the ${base62d: operator to decode base 62 numbers.
1471 28. dnsdb lookups can now access SRV records.
1473 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1474 the configuration file.
1476 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1477 (-v) mode. This makes the output for a verbose queue run more intelligible.
1479 31. Added a use_postmaster feature to recipient callouts.
1481 32. Added the $body_zerocount variable, containing the number of binary zero
1482 bytes in the message body.
1484 33. The time of last modification of the "new" subdirectory is now used as the
1485 "mailbox time last read" when there is a quota error for a maildir
1488 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1490 35. Added +ignore_unknown as a special item in host lists.
1492 36. Code for decoding IPv6 addresses in host lists is now included, even if
1493 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1494 address was recognized as an IP address, but was then not correctly decoded
1495 into binary, causing unexpected and incorrect effects when compared with
1502 1. Very minor rewording of debugging text in manualroute to say "list of
1503 hosts" instead of "hostlist".
1505 2. If verify=header_syntax was set, and a header line with an unqualified
1506 address (no domain) and a large number of spaces between the end of the
1507 name and the colon was received, the reception process suffered a buffer
1508 overflow, and (when I tested it) crashed. This was caused by some obsolete
1509 code that should have been removed. The fix is to remove it!
1511 3. When running in the test harness, delay a bit after writing a bounce
1512 message to get a bit more predictability in the log output.
1514 4. Added a call to search_tidyup() just before forking a reception process. In
1515 theory, someone could use a lookup in the expansion of smtp_accept_max_
1516 per_host which, without the tidyup, could leave open a database connection.
1518 5. Added the variables $recipient_data and $sender_data which get set from a
1519 lookup success in an ACL "recipients" or "senders" condition, or a router
1520 "senders" option, similar to $domain_data and $local_part_data.
1522 6. Moved the writing of debug_print from before to after the "senders" test
1525 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1526 problems for message scanning, either using a data ACL, or using
1527 local_scan() because the Received: header was not generated till after they
1528 were called (in order to set the time as the time of reception completion).
1529 I have revised the way this works. The header is now generated after the
1530 body is received, but before the ACL or local_scan() are called. After they
1531 are run, the timestamp in the header is updated.
1537 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1538 before starting a queue runner without re-exec. This happened only when
1539 deliver_drop_privilege was set or when the Exim user was set to root. The
1540 effect of the bug was that timeouts during subsequent deliveries caused
1541 crashes instead of being properly handled. The handler is now left at its
1542 default (and expected) setting.
1544 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1545 message, again when deliver_drop_privilege is set or Exim is run as root.
1546 The bug described in (1) was not present in this case, but the tidying up
1547 of the other signals was missing. I have made the two cases consistent.
1549 3. The ignore_target_hosts setting on a manualroute router was being ignored
1550 for hosts that were looked up using the /MX notation.
1552 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1555 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1556 operated on the sender address. After changing the $sender_address to <>
1557 for the sender address verify, Exim was re-instated it as the original
1558 (before rewriting) address, but remembering that it had rewritten it, so it
1559 wasn't rewriting it again. This bug also had the effect of breaking the
1560 sender address verification caching when the sender address was rewritten.
1562 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1563 This has been changed so that if the ip literal address matches
1564 ignore_target_hosts, the router declines.
1566 7. Added expansion conditions match_domain, match_address, and match_local_
1567 part (NOT match_host).
1569 8. The placeholder for the Received: header didn't have a length field set.
1571 9. Added code to Exim itself and to exim_lock to test for a specific race
1572 condition that could lead to file corruption when using MBX delivery. The
1573 issue is with the lockfile that is created in /tmp. If this file is removed
1574 after a process has opened it but before that process has acquired a lock,
1575 there is the potential for a second process to recreate the file and also
1576 acquire a lock. This could lead to two Exim processes writing to the file
1577 at the same time. The added code performs the same test as UW imapd; it
1578 checks after acquiring the lock that its file descriptor still refers to
1579 the same named file.
1581 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1582 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1585 11. Added the smtp_active_hostname option. If used, this will typically be made
1586 to depend on the incoming interface address. Because $interface_address is
1587 not set up until the daemon has forked a reception process, error responses
1588 that can happen earlier (such as "too many connections") no longer contain
1591 12. If an expansion in a condition on a "warn" statement fails because a lookup
1592 defers, the "warn" statement is abandoned, and the next ACL statement is
1593 processed. Previously this caused the whole ACL to be aborted.
1595 13. Added the iplsearch lookup type.
1597 14. Added ident_timeout as a log selector.
1599 15. Added tls_certificate_verified as a log selector.
1601 16. Added a global option tls_require_ciphers (compare the smtp transport
1602 option of the same name). This controls incoming TLS connections.
1604 17. I finally figured out how to make tls_require_ciphers do a similar thing
1605 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1606 before starting the TLS session.
1608 18. Tabs are now shown as \t in -bP output.
1610 19. If the log selector return_path_on_delivery was set, Exim crashed when
1611 bouncing a message because it had too many Received: header lines.
1613 20. If two routers both had headers_remove settings, and the first one included
1614 a superfluous trailing colon, the final name in the first list and the
1615 first name in the second list were incorrectly joined into one item (with a
1616 colon in the middle).
1622 1. Added -C and -D options to the exinext utility, mainly to make it easier
1623 to include in the automated testing, but these could be helpful when
1624 multiple configurations are in use.
1626 2. The exinext utility was not formatting the output nicely when there was
1627 an alternate port involved in the retry record key, nor when there was a
1628 message id as well (for retries that were specific to a specific message
1629 and a specific host). It was also confused by IPv6 addresses, because of
1630 the additional colons they contain. I have fixed the IPv4 problem, and
1631 patched it up to do a reasonable job for IPv6.
1633 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1634 delivery, the log line now contains "pipelined" if PIPELINING was used.
1636 4. An SMTP transport process used to panic and die if the bind() call to set
1637 an explicit outgoing interface failed. This has been changed; it is now
1638 treated in the same way as a connect() failure.
1640 5. A reference to $sender_host_name in the part of a conditional expansion
1641 that was being skipped was still causing a DNS lookup. This no longer
1644 6. The def: expansion condition was not recognizing references to header lines
1645 that used bh_ and bheader_.
1647 7. Added the _cache feature to named lists.
1649 8. The code for checking quota_filecount in the appendfile transport was
1650 allowing one more file than it should have been.
1652 9. For compatibility with Sendmail, the command line option
1660 and sets the incoming protocol and host name (for trusted callers). The
1661 host name and its colon can be omitted when only the protocol is to be set.
1662 Note the Exim already has two private options, -pd and -ps, that refer to
1663 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1664 "s", but I don't think that's a major issue.
1666 10. A number of refactoring changes to the code, none of which should affect
1669 (a) The number of logging options was getting close to filling up the
1670 32-bit word that was used as a bit map. I have split them into two classes:
1671 those that are passed in the argument to log_write(), and those that are
1672 only ever tested independently outside of that function. These are now in
1673 separate 32-bit words, so there is plenty of room for expansion again.
1674 There is no change in the user interface or the logging behaviour.
1676 (b) When building, for example, log lines, the code previously used a
1677 macro that called string_cat() twice, in order to add two strings. This is
1678 not really sufficiently general. Furthermore, there was one instance where
1679 it was actually wrong because one of the argument was used twice, and in
1680 one call a function was used. (As it happened, calling the function twice
1681 did not affect the overall behaviour.) The macro has been replaced by a
1682 function that can join an arbitrary number of extra strings onto a growing
1685 (c) The code for expansion conditions now uses a table and a binary chop
1686 instead of a serial search (which was left over from when there were very
1687 few conditions). Also, it now recognizes conditions like "pam" even when
1688 the relevant support is not compiled in: a suitably worded error message is
1689 given if an attempt is made to use such a condition.
1691 11. Added ${time_interval:xxxxx}.
1693 12. A bug was causing one of the ddress fields not to be passed back correctly
1694 from remote delivery subprocesses. The field in question was not being
1695 subsequently used, so this caused to problems in practice.
1697 13. Added new log selectors queue_time and deliver_time.
1699 14. Might have fixed a bug in maildirsizefile handling that threw up
1700 "unexpected character" debug warnings, and recalculated the data
1701 unnecessarily. In any case, I expanded the warning message to give more
1704 15. Added the message "Restricted characters in address" to the statements in
1705 the default ACL that block characters like @ and % in local parts.
1707 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
1708 Three changes have been made:
1710 (a) There was a serious bug; a negative response to MAIL caused the whole
1711 recipient domain to be cached as invalid, thereby blocking all messages
1712 to all local parts at the same domain, from all senders. This bug has
1713 been fixed. The domain is no longer cached after a negative response to
1714 MAIL if the sender used is not empty.
1716 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
1719 (c) A new callout option, "use_sender" has been added for people who want
1720 the modified behaviour.
1726 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
1729 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
1730 indeed breaks things for older releases.
1732 3. Added additional logging to the case where there is a problem reading data
1733 from a filter that is running in a subprocess using a pipe, in order to
1734 try to track down a specific problem.
1736 4. Testing facility fudge: when running in the test harness and attempting
1737 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
1738 getting "No route to host". Convert this to a timeout.
1740 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
1743 6. Some OS don't have socklen_t but use size_t instead. This affects the
1744 fifth argument of getsockopt() amongst other things. This is now
1745 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
1746 can be set for individual OS. I have set it for SunOS5, OSF1, and
1747 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
1748 some earlier ones do not.
1750 7. Change 4.30/15 was not doing the test caselessly.
1752 8. The standard form for an IPv6 address literal was being rejected by address
1753 parsing in, for example, MAIL and RCPT commands. An example of this kind of
1754 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1755 this, as well as the form without the "IPv6" on the front (but only when
1756 address literals are enabled, of course).
1758 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
1760 10. Exim crashed if a message with an empty sender address specified by -f
1761 encountered a router with an errors_to setting. This could be provoked only
1762 by a command such as
1766 where an empty string was supplied; "<>" did not hit this bug.
1768 11. Installed PCRE release 4.5.
1770 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
1771 remained set. It is now erased.
1773 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
1774 times from message ids (which are base 36 rather than the normal 62).
1776 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
1777 were being counted as actual protocol errors, and logged if the log
1778 selector +smtp_protocol_error was set. One cannot be perfect in this test,
1779 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
1780 and DATA following a set of rejected RCPTs do not count as protocol errors.
1781 In other words, Exim assumes they were pipelined, though this may not
1782 actually be the case. Of course, in all cases the client gets an
1783 appropriate error code.
1785 15. If a lookup fails in an ACL condition, a message about the failure may
1786 be available; it is used if testing the ACL cannot continue, because most
1787 such messages specify what the cause of the deferral is. However, some
1788 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
1789 that caused an old message to be retained and used if a later statement
1790 caused a defer, replacing the real cause of the deferral.
1792 16. If an IP address had so many PTR records that the DNS lookup buffer
1793 was not large enough to hold them, Exim could crash while trying to process
1794 the truncated data. It now detects and logs this case.
1796 17. Further to 4.21/58, another change has been made: if (and only if) the
1797 first line of a message (the first header line) ends with CRLF, a bare LF
1798 in a subsequent header line has a space inserted after it, so as not to
1799 terminate the header.
1801 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
1802 unnecessarily, used atoi() instead of strtol(), and didn't check the
1803 termination when getting file sizes from file names by regex.
1805 19. Completely re-implemented the support for maildirsize files, in the light
1806 of a number of problems with the previous contributed implementation
1807 (4.30/29). In particular:
1809 . If the quota is zero, the maildirsize file is maintained, but no quota is
1812 . If the maildir directory does not exist, it is created before any attempt
1813 to write a maildirsize file.
1815 . The quota value in the file is just a cache; if the quota is changed in
1816 the transport, the new value overrides.
1818 . A regular expression is available for excluding directories from the
1821 20. The autoreply transport checks the characters in options that define the
1822 message's headers; it allows continued headers, but it was checking with
1823 isspace() after an embedded newline instead of explicitly looking for a
1826 21. If all the "regular" hosts to which an address was routed had passed their
1827 expiry times, and had not reached their retry times, the address was
1828 bounced, even if fallback hosts were defined. Now Exim should go on to try
1831 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
1832 equivalent code in the SMTP transport. Some hosts send humungous responses
1833 to HELO/EHLO, more than 1024 it seems.
1835 23. Refactoring: code in filter.c used (void *) for "any old type" but this
1836 gives compiler warnings in some environments. I've now done it "properly",
1839 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
1840 (because of problems with the built-in one) was declared to return uschar *
1841 instead of char *, causing compiler failure.
1843 25. Fixed a file descriptor leak when processing alias/forward files.
1845 26. Fixed a minor format string issue in dbfn.c.
1847 27. Typo in exim.c: ("dmbnz" for "dbmnz").
1849 28. If a filter file refered to $h_xxx or $message_headers, and the headers
1850 contained RFC 2047 "words", Exim's memory could, under certain conditions,
1853 29. When a sender address is verified, it is cached, to save repeating the test
1854 when there is more than one recipient in a message. However, when the
1855 verification involves a callout, it is possible for different callout
1856 options to be set for different recipients. It is too complicated to keep
1857 track of this in the cache, so now Exim always runs a verification when a
1858 callout is required, relying on the callout cache for the optimization.
1859 The overhead is duplication of the address routing, but this should not be
1862 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
1863 to be verified with callout=postmaster, and the main callout worked but the
1864 postmaster check failed, the verification correctly failed. However, if a
1865 subsequent RCPT command asked for sender verification *without* the
1866 postmaster check, incorrect caching caused this verification also to fail,
1869 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
1870 it was not caching the DNS options (qualify_single, search_parents) that
1871 were used when the lookup failed. A subsequent lookup with different
1872 options therefore always gave the same answer, though there were cases
1873 where it should not have. (Example: a "domains = !$mx_any" option on a
1874 dnslookup router: the "domains" option is always processed without any
1875 widening, but the router might have qualify_single set.) Now Exim uses the
1876 cached value only when the same options are set.
1878 32. Added John Jetmore's "exipick" utility to the distribution.
1880 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
1881 than a timeout (e.g. a certificate is required, and is not provided), an
1882 Exim server now closes the connection immediately. Previously it waited for
1883 the client to close - but if the client is SSL, it seems that they each
1884 wait for each other, leading to a delay before one of them times out.
1886 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
1887 maintained 0.8.x compatibility because I don't think many are using it, and
1888 it is clearly obsolete.
1890 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
1893 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
1894 client certificate was expired. A simple patch fixes this, though I don't
1895 understand the full logic of why the verify callback is called multiple
1898 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
1899 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
1900 which causes problems with some clients (such as the Certicom SSL Plus
1901 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
1902 disables the coutermeasure allowing Eudora to connect."
1904 38. Exim was not checking that a write() to a log file succeeded. This could
1905 lead to Bad Things if a log got too big, in particular if it hit a file
1906 size limit. Exim now panics and dies if it cannot write to a log file, just
1907 as it does if it cannot open a log file.
1909 39. Modified OS/Makefile-Linux so that it now contains
1911 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
1913 The two -D definitions ensure that Exim is compiled with large file
1914 support, which makes it possible to handle log files that are bigger than
1917 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
1918 instance) a domain was checked against a named list that involved a lookup,
1919 causing $domain_data to be set, then another domain was checked against the
1920 same list, then the first domain was re-checked, the value of $domain_data
1921 after the final check could be wrong. In particular, if the second check
1922 failed, it could be set empty. This bug probably also applied to
1925 41. The strip_trailing_dot option was not being applied to the address given
1926 with the -f command-line option.
1928 42. The code for reading a message's header from the spool was incrementing
1929 $received_count, but never initializing it. This meant that the value was
1930 incorrect (doubled) while delivering a message in the same process in which
1931 it was received. In the most common configuration of Exim, this never
1932 happens - a fresh exec is done - but it can happen when
1933 deliver_drop_privilege is set.
1935 43. When Exim logs an SMTP synchronization error - client data sent too soon -
1936 it now includes up to 150 characters of the unexpected data in the log
1939 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
1940 and building data strings. The size of both of these buffers was 10 000
1941 bytes - far larger than anybody would *ever* want, thought I. Needless to
1942 say, somebody hit the limit. I have increased the maximum line length to
1943 20 000 and the maximum data length of concatenated lines to 100 000. I have
1944 also fixed two bugs, because there was no checking on these buffers. Tsk,
1945 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1946 buffer is too small.
1948 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1949 lsearch lookups. Now it does.
1951 46. When parsing a route_list item in a manualroute router, a fixed-length
1952 buffer was used for the list of hosts. I made this 1024 bytes long,
1953 thinking that nobody would ever have a list of hosts that long. Wrong.
1954 Somebody had a whole pile of complicated expansion conditions, and the
1955 string was silently truncated, leading to an expansion error. It turns out
1956 that it is easier to change to an unlimited length (owing to other changes
1957 that have happened since this code was originally written) than to build
1958 structure for giving a limitation error. The length of the item that
1959 expands into the list of hosts is now unlimited.
1961 47. The lsearch lookup could not handle data where the length of text line was
1962 more than 4095 characters. Such lines were truncated, leading to shortened
1963 data being returned. It should now handle lines of any length.
1965 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
1966 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
1969 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
1970 replacing strings such as BIN_DIRECTORY in a source file: the replacement
1971 no longer happens in comment lines. A list of replacements is now placed
1972 at the head of all of the source files, except those whose only change is
1973 to replace PERL_COMMAND in the very first #! line.
1975 50. Replaced the slow insertion sort in queue.c, for sorting the list of
1976 messages on the queue, with a bottom-up merge sort, using code contributed
1977 by Michael Haardt. This should make operations like -bp somewhat faster on
1978 large queues. It won't affect queue runners, except when queue_run_in_order
1981 51. Installed eximstats 1.31 in the distribution.
1983 52. Added support for SRV lookups to the dnslookup router.
1985 53. If an ACL referred to $message_body or $message_body_end, the value was not
1986 reset for any messages that followed in the same SMTP session.
1988 54. The store-handling optimization for building very long strings was not
1989 differentiating between the different store pools. I don't think this
1990 actually made any difference in practice, but I've tidied it.
1992 55. While running the routers to verify a sender address, $sender_address
1993 was still set to the sender address. This is wrong, because when routing to
1994 send a bounce to the sender, it would be empty. Therefore, I have changed
1995 it so that, while verifying a sender address, $sender_address is set to <>.
1996 (There is no change to what happens when verifying a recipient address.)
1998 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
1999 target A or AAAA records (if not already returned) without resetting the
2000 qualify_single or search_parents options of the DNS resolver. These are
2001 inappropriate in this case because the targets of MX and SRV records must
2002 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2003 target that, when qualified, matched something in the local domain. These
2004 two options are now turned off when doing these lookups.
2006 57. It seems that at least some releases of Reiserfs (which does not have the
2007 concept of a fixed number of inodes) returns zero and not -1 for the
2008 number of available inodes. This interacted badly with check_spool_inodes,
2009 which assumed that -1 was the "no such thing" setting. What I have done is
2010 to check that the total number of inodes is greater than zero before doing
2011 the test of how many are available.
2013 58. When a "warn" ACL statement has a log_message modifier, the message is
2014 remembered, and not repeated. This is to avoid a lot of repetition when a
2015 message has many recipients that cause the same warning to be written.
2016 Howewer, Exim was preserving the list of already written lines for an
2017 entire SMTP session, which doesn't seem right. The memory is now reset if a
2018 new message is started.
2020 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2021 debugging output unless log_rewrite was also set.
2023 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2024 of (int)(handle) when we know that handle contains (void *)(-1).
2026 61. The Exim daemon panic-logs an error return when it closes the incoming
2027 connection. However "connection reset by peer" seems to be common, and
2028 isn't really an error worthy of noting specially, so that particular error
2031 62. When Exim is trying to find all the local interfaces, it used to panic and
2032 die if the ioctl to get the interface flags failed. However, it seems that
2033 on at least one OS (Solaris 9) it is possible to have an interface that is
2034 included in the list of interfaces, but for which you get a failure error
2035 for this call. This happens when the interface is not "plumbed" into a
2036 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2037 failure of the "get flags" call assumes that the interface is down.
2039 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2040 makes life easier for people who are doing arithmetic on fields extracted
2041 from dates, where you often get leading zeros that should not be
2042 interpreted as octal.
2044 64. Added qualify_domain to the redirect router, to override the global
2047 65. If a pathologically long header line contained very many addresses (the
2048 report of this problem mentioned 10 000) and each of them was rewritten,
2049 Exim could use up a very large amount of memory. (It kept on making new
2050 copies of the header line as it rewrote, and never released the old ones.)
2051 At the expense of a bit more processing, the header rewriting function has
2052 been changed so that it no longer eats memory in this way.
2054 66. The generation of the Received: header has been moved from the time that a
2055 message starts to be received, to the time that it finishes. The timestamp
2056 in the Received: header should now be very close to that of the <= log
2057 line. There are two side-effects of this change:
2059 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2060 logged header lines no longer include the local Received: line, because
2061 it has not yet been created. The same applies to a copy of the message
2062 that is returned to a non-SMTP sender when a message is rejected.
2064 (b) When a filter file is tested using -bf, no additional Received: header
2065 is added to the test message. After some thought, I decided that this
2068 This change does not affect the value of $received_for. It is still set
2069 after address rewriting, but before local_scan() is called.
2071 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2073 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2074 gave an unhelpful panic error message, and a defer error. I have managed to
2075 change this behaviour so that it now rejects any supplied certificate,
2076 which seems right, as the list of acceptable certificates is empty.
2078 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2079 gave an unhelpful defer error. I have not managed to make this reject any
2080 supplied certificates, but the error message it gives is "no certificate
2081 supplied", which is not helpful.
2083 70. exigrep's output now also includes lines that are not associated with any
2084 message, but which match the given pattern. Implemented by a patch from
2085 Martin Sluka, which also tidied up the Perl a bit.
2087 71. Recipient callout verification, like sender verification, was using <> in
2088 the MAIL FROM command. This isn't really the right thing, since the actual
2089 sender may affect whether the remote host accepts the recipient or not. I
2090 have changed it to use the actual sender in the callout; this means that
2091 the cache record is now keyed on a recipient/sender pair, not just the
2092 recipient address. There doesn't seem to be a real danger of callout loops,
2093 since a callout by the remote host to check the sender would use <>.
2094 [SEE ABOVE: changed after hitting problems.]
2096 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2097 temporary errors. However, in the case of such a code being given after
2098 the end of a data transmission (i.e. after ".") Exim was failing to write
2099 a retry record for the message. (Yes, there was some broken host that was
2100 actually sending 8xx at this point.)
2102 73. An unknown lookup type in a host list could cause Exim to panic-die when
2103 the list was checked. (An example that provoked this was putting <; in the
2104 middle of a list instead of at the start.) If this happened during a DATA
2105 ACL check, a -D file could be left lying around. This kind of configuration
2106 error no longer causes Exim to die; instead it causes a defer errror. The
2107 incident is still logged to the main and panic logs.
2109 74. Buglet left over from Exim 3 conversion. The message "too many messages
2110 in one connection" was written to the rejectlog but not the mainlog, except
2111 when address rewriting (yes!) was being logged.
2113 75. Added write_rejectlog option.
2115 76. When a system filter was run not as root (that is, when system_filter_user
2116 was set), the values of the $n variables were not being returned to the
2117 main process; thus, they were not subsequently available in the $sn
2120 77. Added +return_path_on_delivery log selector.
2122 78. A connection timeout was being treated differently from recipients deferred
2123 when testing hosts_max_try with a message that was older than the host's
2124 retry timeout. (The host should not be counted, thus allowing all hosts to
2125 be tried at least once before bouncing.) This may have been the cause of an
2126 occasionally reported bug whereby a message would remain on the queue
2127 longer than the retry timeout, but would be bounced if a delivery was
2128 forced. I say "may" because I never totally pinned down the problem;
2129 setting up timeout/retry tests is difficult. See also the next item.
2131 79. The ultimate address timeout was not being applied to errors that involved
2132 a combination of host plus message (for example, a timeout on a MAIL
2133 command). When an address resolved to a number of possible hosts, and they
2134 were not all tried for each delivery (e.g. because of hosts_max_try), a
2135 message could remain on the queue longer than the retry timeout.
2137 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2140 81. Fixed an obscure SMTP outgoing bug which required at least the following
2141 conditions: (a) there was another message waiting for the same server;
2142 (b) the server returned 5xx to all RCPT commands in the first message so
2143 that the message was not completed; (c) the server dropped the connection
2144 or gave a negative response to the RSET that Exim sends to abort the
2145 transaction. The observed case was a dropped connection after DATA that had
2146 been sent in pipelining mode. That is, the server had advertised PIPELINING
2147 but was not implementing it correctly. The effect of the bug was incorrect
2148 behaviour, such as trying another host, and this could lead to a crash.
2154 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2155 and daemon.c were passed as pointers to ints; they should have been
2156 pointers to socklen_t variables (which are typically unsigned ints).
2158 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2161 3. Fixed a really odd bug that affected only the testing scheme; patching a
2162 certain fixed string in the binary changed the value of another string that
2163 happened to be identical to the end of the original first string.
2165 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2166 name", it returns that address as the IP address. On some operating
2167 systems (e.g. Solaris), it also passes back the IP address string as the
2168 "host name". However, on others (e.g. Linux), it passes back an empty
2169 string. Exim wasn't checking for this, and was changing the host name to an
2170 empty string, assuming it had been canonicized.
2172 5. Although rare, it is permitted to have more than one PTR record for a given
2173 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2174 all the names associated with an address, because they do in Solaris.
2175 However, it seems that they do not in Linux for data that comes from the
2176 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2177 I found this out when I moved to a new Linux workstation and tried to run
2178 the Exim test suite.
2180 To get round this problem I have changed the code so that it now does its
2181 own call to the DNS to look up PTR records when searching for a host name.
2182 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2183 addresses that are only in /etc/hosts are still found.
2185 This behaviour is, however, controlled by an option called host_lookup_
2186 order, which defaults to "bydns:byaddr". If people want to use the other
2187 order, or indeed, just use one or the other means of lookup, they can
2188 specify it in this variable.
2190 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2191 some operating systems, this comes back from gethostbyaddr() as an empty
2192 string, and this is what Exim used to test for. However, it seems that in
2193 other systems, "." is yielded. Exim now tests for this case too.
2195 7. The values of check_spool_space and check_log_space are now held internally
2196 as a number of kilobytes instead of an absolute number of bytes. If a
2197 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2198 kilobyte. This means that much larger values can be stored.
2200 8. Exim monitor: an attempt to get the action menu when not actually pointing
2201 at a message produces an empty menu entitled "No message selected". This
2202 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2203 no entries in it ("Shell widget menu has zero width and/or height"). So I
2204 have added a single, blank menu entry in this case.
2206 9. Added ${quote_local_part.
2208 10. MIME decoding is now applied to the contents of Subject: header lines when
2211 11. Now that a reference to $sender_host_address automatically causes a reverse
2212 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2213 host lookup before query-style lookups in lists that might use this
2214 variable. This has therefore been abolished, and the "net-" prefix is no
2215 longer necessary for query-style lookups.
2217 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2218 have been a typo for LFLAGS, so it has been changed.
2220 13. The install script calls Exim with "-C /dev/null" in order to find the
2221 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2222 to be output. Howeve, since Exim outputs its version number before the
2223 error, it didn't break the script. It just looked ugly. I fixed this by
2224 always allowing "-C /dev/null" if the caller is root.
2226 14. Ignore overlarge ACL variable number when reading spool file - insurance
2227 against a later release with more variables having written the file.
2229 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2230 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2231 this, as well as the form without the "IPv6" on the front.
2233 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2234 OS/Makefile-Darwin file.
2236 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2237 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2239 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2240 variable. However, some OS use ints for inodes. I've added cast to long int
2241 to get rid of the compiler warning.
2243 19. I had forgotten to lock out "/../" in configuration file names when
2244 ALT_CONFIG_PREFIX was set.
2246 20. Routers used for verification do not need to specify transports. However,
2247 if such a router generated a host list, and callout was configured, Exim
2248 crashed, because it could not find a port number from the (non-existent)
2249 transport. It now assumes port 25 in this circumstance.
2251 21. Added the -t option to exigrep.
2253 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2254 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2255 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2256 them is set, LOOKUP_LSEARCH is forced.
2258 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2260 24. Added sender and host information to the "rejected by local_scan()" log
2261 line; previously there was no indication of these.
2263 25. Added .include_if_exists.
2265 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2266 Linux. It turns out that not all file systems support this. Apparently some
2267 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2268 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2269 not supported on the file descriptor, is now ignored when Exim is trying to
2270 sync a directory. This applies only to Linux.
2272 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2274 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2275 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2278 29. Added support for maildirsize files from supplied patch (modified a bit).
2280 30. The use of :fail: followed by an empty string could lead Exim to respond to
2281 sender verification failures with (e.g.):
2283 550 Verification failed for <xxx>
2284 550 Sender verify failed
2286 where the first response line was missing the '-' that indicates it is not
2287 the final line of the response.
2289 31. The loop for finding the name of the user that called Exim had a hardwired
2290 limit of 10; it now uses the value of finduser_retries, which is used for
2291 all other user lookups.
2293 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2296 33. Exim was neglecting to zero errno before one call of strtol() when
2297 expanding a string and expecting an integer value. On some systems this
2298 resulted in spurious "integer overflow" errors. Also, it was casting the
2299 result into an int without checking.
2301 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2302 did not work. The code looks as if it has *never* worked, though it appears
2303 to have been documented since at least releast 1.62. I have made it work.
2305 35. The "timeout_DNS" error in retry rules, also documented since at least
2306 1.62, also never worked. As it isn't clear exactly what this means, and
2307 clearly it isn't a major issue, I have abolished the feature by treating it
2308 as "timeout", and writing a warning to the main and panic logs.
2310 36. The display of retry rules for -brt wasn't always showing the error code
2313 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2314 timeout_connect_A, timeout_connect_MX.
2316 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2317 to the empty sender.
2319 39. The daemon was not analysing the content of -oX till after it had closed
2320 stderr and disconnected from the controlling terminal. This meant that any
2321 syntax errors were only noted on the panic log, and the return code from
2322 the command was 0. By re-arranging the code a little, I've made the
2323 decoding happen first, so such errors now appear on stderr, and the return
2324 code is 1. However, the actual setting up of the sockets still happens in
2325 the disconnected process, so errors there are still only recorded on the
2328 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2329 connections (as happens on some IP stacks) was logged at start up time as
2330 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2331 it from "IPv6 and IPv4", which means that two separate sockets are being
2334 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2335 says whether AF_INET or AF_INET6 was passed as an argument.
2337 42. Exiwhat output was messed up when time zones were included in log
2340 43. Exiwhat now gives more information about the daemon's listening ports,
2341 and whether -tls-on-connect was used.
2343 44. The "port" option of the smtp transport is now expanded.
2345 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2346 silently ignored. Now an error message is written to the main and panic
2349 46. There's a new ACL modifier called "logwrite" which writes to a log file
2350 as soon as it is encountered.
2352 47. Added $local_user_uid and $local_user_gid at routing time.
2354 48. Exim crashed when trying to verify a sender address that was being
2357 49. Exim was recognizing only a space character after ".include". It now also
2358 recognizes a tab character.
2360 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2361 extracting the relevant information from the specification. The man page no
2362 longer contains scrambled data for the -d option, and I've added a section
2363 at the front about calling Exim under different names.
2365 51. Added "extra_headers" argument to the "mail" command in filter files.
2367 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2370 53. Installed eximstats 1.29.
2372 54. Added transport_filter_timeout as a generic transport option.
2374 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2375 Cc: header lines. This was required by RFC 822, but it not required by RFC
2378 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2379 incoming messages that did not have them. Now it does so only if the
2380 message originates locally, that is, if there is no associated remote host
2381 address. When Resent- header lines are present, this applies to the Resent-
2382 lines rather than the non-Resent- lines.
2384 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2385 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2387 58. Messages for configuration errors now include the name of the main
2388 configuration file - useful now that there may be more than one file in a
2389 list (.included file names were always shown).
2391 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2392 for those rare installations that do not start the daemon as root or run it
2393 setuid root. I've cut out the call to initgroups() if the daemon is not
2396 60. The Exim user and group can now be bound into the binary as text strings
2397 that are looked up at the start of Exim's processing.
2399 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2401 62. Added $mailstore_basename variable.
2403 63. Installed patch to sieve.c from Michael Haardt.
2405 64. When Exim failed to open the panic log after failing to open the main log,
2406 the original message it was trying to log was written to stderr and debug
2407 output, but if they were not available (the usual case in production), it
2408 was lost. Now it is written to syslog before the two lines that record the
2409 failures to open the logs.
2411 65. Users' Exim filters run in subprocesses under the user's uid. It is
2412 possible for a "deliver" command or an alias in a "personal" command to
2413 provoke an address rewrite. If logging of address rewriting is configured,
2414 this fails because the process is not running as root or exim. There may be
2415 a better way of dealing with this, but for the moment (because 4.30 needs
2416 to be released), I have disabled address rewrite logging when running a
2417 filter in a non-root, non-exim process.
2423 1. The buildconfig auxiliary program wasn't quoting the value set for
2424 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2425 not defined. This bug was masked in 4.22 by the effect that was fixed in
2428 2. Some messages that were rejected after a message id was allocated were
2429 shown as "incomplete" by exigrep. It no longer does this for messages that
2430 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2432 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2433 have allow_domain_literals set, the ID did not get logged in the <= line.
2434 Domain literals are now always recognized in Message-ID: header lines.
2436 4. The first argument for a ${extract expansion item is the key name or field
2437 number. Leading and trailing spaces in this item were not being ignored,
2438 causing some misleading effects.
2440 5. When deliver_drop_privilege was set, single queue runner processes started
2441 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2442 same command in the process it spins off) were not dropping privilege.
2444 6. When the daemon running as "exim" started a queue runner, it always
2445 re-executed Exim in the spun-off process. This is a waste of effort when
2446 deliver_drop_privilege is set. The new process now just calls the
2447 queue-runner function directly.
2453 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2456 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2457 ignored. Though the use of -oP was forcing the writing of a pid file, it
2458 was always written to the default place.
2460 3. If the message "no IP address found for host xxxx" is generated during
2461 incoming verification, it is now followed by identification of the incoming
2462 connection (so you can more easily find what provoked it).
2464 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2466 5. Added some features to "harden" Exim a bit more against certain attacks:
2468 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2469 be put in Local/Makefile. This is like the never_users runtime option,
2470 but it cannot be overridden. The default setting is "root".
2472 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2473 prefix string with which any file named in a -C command line option
2476 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2477 is retained for -C and -D only if the caller of Exim is root. Without
2478 it, the exim user may also use -C and -D and retain privilege.
2480 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2481 command line option is disabled.
2483 6. Macro names set by the -D option must start with an upper case letter, just
2484 like macro names defined in the configuration file.
2486 7. Added "dereference=" facility to LDAP.
2488 8. Two instances of the typo "uknown" in the source files are fixed.
2490 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2491 the Configure-Makefile script screwed up while processing it.
2493 10. Incorporated PCRE 4.4.
2495 11. The SMTP synchronization check was not operating right at the start of an
2496 SMTP session. For example, it could not catch a HELO sent before the client
2497 waited for the greeting. There is now a check for outstanding input at the
2498 point when the greeting is written. Because of the duplex, asynchronous
2499 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2500 way, but not yet received, when the check is performed.
2502 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2503 on TCP/IP sockets, because this apparently causes some broken clients to
2506 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2507 unchanged) from the Cygwin maintainer.
2509 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2510 was supported instead of testing for mbx. Effectively a typo.
2512 15. The spa authenticator server code was not checking that the input it
2513 received was valid base64.
2515 16. The debug output line for the "set" modifier in ACLs was not showing the
2516 name of the variable that was being set.
2518 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2520 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2521 lookup on its own. Something else was needed to trigger the lookup. For
2522 example, a match in host_lookup or the need for a host name in a host list.
2523 Now, if $sender_host_name is referenced and the host name has not yet been
2524 looked up, a lookup is performed. If the lookup fails, the variable remains
2525 empty, and $host_lookup_failed is set to "1".
2527 19. Added "eqi" as a case-independent comparison operator.
2529 20. The saslauthd authentication condition could segfault if neither service
2530 nor realm was specified.
2532 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2533 error message that was logged was misleading, and incoming SMTP
2534 connections were dropped. The message is now more accurate, and temporary
2535 errors are given to SMTP connections.
2537 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2538 (except RSET) with a 421 error, until QUIT is received. However, it was
2539 failing to send a response to QUIT.
2541 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2542 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2543 ACL is now run after the helo_try_verify_hosts code.
2545 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2546 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2547 case-independent, so other case variants are also recognized). Apparently
2548 some systems use these upper case variants.
2550 25. If more than two messages were waiting for the same host, and a transport
2551 filter was specified for the transport, Exim sent two messages over the
2552 same TCP/IP connection, and then failed with "socket operation on non-
2553 socket" when it tried to send the third.
2555 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2557 27. The extern definition of crypt16() in expand.c was not being excluded when
2558 the OS had its own crypt16() function.
2560 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2561 as a preferred synonym for return_size_limit, both as an option and as an
2564 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2566 30. Changed the default configuration ACL to relax the local part checking rule
2567 for addresses that are not in any local domains. For these addresses,
2568 slashes and pipe symbols are allowed within local parts, but the sequence
2569 /../ is explicitly forbidden.
2571 31. SPA server authentication was not clearing the challenge buffer before
2574 32. log_message in a "warn" ACL statement was writing to the reject log as
2575 well as to the main log, which contradicts the documentation and doesn't
2576 seem right (because no rejection is happening). So I have stopped it.
2578 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2579 However, I am unable to do any testing of this.
2581 34. Fixed an infelicity in the appendfile transport. When checking directories
2582 for a mailbox, to see if any needed to be created, it was accidentally
2583 using path names with one or more superfluous leading slashes; tracing
2584 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2586 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2587 contents are added to the log line that is written for every discarded
2588 recipient. (Previously a log_message setting was ignored.)
2590 36. The ${quote: operator now quotes the string if it is empty.
2592 37. The install script runs exim in order to find its version number. If for
2593 some reason other than non-existence or emptiness, which it checks, it
2594 could not run './exim', it was installing it with an empty version number,
2595 i.e. as "exim-". This error state is now caught, and the installation is
2598 38. An argument was missing from the function that creates an error message
2599 when Exim fails to connect to the socket for saslauthd authentication.
2600 This could cause Exim to crash, or give a corrupted message.
2602 39. Added isip, isip4, and isip6 to ${if conditions.
2604 40. The ACL variables $acl_xx are now saved with the message, and can be
2605 accessed later in routers, transports, and filters.
2607 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2608 strings in the file are not string-expanded.
2610 42. If a MAIL command specified a SIZE value that was too large to fit into an
2611 int variable, the check against message_size_limit failed. Such values are
2612 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2613 day this will have to be increased, but I don't think I want to be around
2614 when emails are that large.
2621 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2622 iconv() is not standard in FreeBSD.
2624 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2625 IPv6 enabled. The observed symptom was a segmentation fault on return from
2626 the function os_common_find_running_interfaces() in src/os.c.
2628 3. In the check_special_case() function in daemon.c I had used "errno" as an
2629 argument name, which causes warnings on some systems. This was basically a
2630 typo, since it was named "eno" in the comments!
2632 4. The code that waits for the clock to tick (at a resolution of some fraction
2633 of a second) so as to ensure message-id uniqueness was always waiting for
2634 at least one whole tick, when it could have waited for less. [This is
2635 almost certainly not relevant at current processor speeds, where it is
2636 unlikely to ever wait at all. But we try to future-proof.]
2638 5. The function that sleeps for a time interval that includes fractions of a
2639 second contained a race. It did not block SIGALRM between setting the
2640 timer, and suspending (a couple of lines later). If the interval was short
2641 and the sigsuspend() was delayed until after it had expired, the suspension
2642 never ended. On busy systems this could lead to processes getting stuck for
2645 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2646 process, before it forks any delivery processes. The open lookup caching
2647 mechanism meant that the open file or database connection was passed into
2648 the delivery process. The problem was that delivery processes always tidy
2649 up cached lookup data. This could cause a problem for the next delivery
2650 process started by the queue runner, because the external queue runner
2651 process does not know about the closure. So the next delivery process
2652 still has data in the lookup cache. In the case of a file lookup, there was
2653 no problem because closing a file descriptor in a subprocess doesn't affect
2654 the parent. However, if the lookup was caching a connection to a database,
2655 the connection was closed, and the second delivery process was likely to
2656 see errors such as "PGSQL: query failed: server closed the connection
2657 unexpectedly". The problem has been fixed by closing all cached lookups
2658 in a queue runner before running a delivery process.
2660 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2661 seem to have the "const" qualifier which it has on other OS. I've
2664 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2665 *of the same type* (client or server) with the same public name that an
2666 error should be diagnosed.
2668 9. When Exim looked up a host name for an IP address, but failed to find the
2669 original IP address when looking up the host name (a safety check), it
2670 output the message "<ip address> does not match any IP for NULL", which was
2671 confusing, to say the least. The bug was that the host name should have
2672 appeared instead of "NULL".
2674 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2675 user that is built into the binary, and the -C or -D options is used, root
2676 privilege is dropped before the configuration file is read. In addition,
2677 logging is switched to stderr instead of the normal log files. If the
2678 configuration then re-defines the Exim user, the unprivileged environment
2679 is probably not what is expected, so Exim logs a panic warning message (but
2682 However, if deliver_drop_privilege is set, the unprivileged state may well
2683 be exactly what is intended, so the warning has been cut out in that case,
2684 and Exim is allowed to try to write to its normal log files.
2690 1. smtp_return_error_details was not giving details for temporary sender
2691 or receiver verification errors.
2693 2. Diagnose a configuration error if two authenticators have the same public
2696 3. Exim used not to create the message log file for a message until the first
2697 delivery attempt. This could be confusing when incoming messages were held
2698 for policy or load reasons. The message log file is now created at the time
2699 the message is received, and an initial "Received" line is written to it.
2701 4. The automatically generated man page for command line options had a minor
2702 bug that caused no ill effects; however, a more serious problem was that
2703 the procedure for building the man page automatically didn't always
2704 operate. Consequently, release 4.20 contains an out-of-date version. This
2705 shouldn't happen again.
2707 5. When building Exim with embedded Perl support, the script that builds the
2708 Makefile was calling 'perl' to find its compile-time parameters, ignoring
2709 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
2711 6. The freeze_tell option was not being used for messages that were frozen on
2712 arrival, either by an ACL or by local_scan().
2714 7. Added the smtp_incomplete_transaction log selector.
2716 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
2717 was accepting AUTH without a new EHLO.
2719 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
2720 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
2723 10. Logging of TCP/IP connections (when configured) now happens in the main
2724 daemon process instead of the child process, so that the TCP/IP connection
2725 count is more accurate (but it can never be perfect).
2727 11. The use of "drop" in a nested ACL was not being handled correctly in the
2728 outer ACL. Now, if condition failure induced by the nested "drop" causes
2729 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
2730 or "require"), the connection is dropped.
2732 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
2733 that yield "discard" can now be used with an "accept" or a "discard" verb,
2734 but an error is generated for any others (because I can't see a useful way
2735 to define what should happen).
2737 13. When an ACL is read dynamically from a file (or anywhere else), the lines
2738 are now processed in the same way as lines in the Exim configuration file.
2739 In particular, continuation lines are supported.
2741 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
2743 15. Added -ti meaning -t -i.
2745 16. Check for letters, digits, hyphens, and dots in the names of dnslist
2746 domains, and warn by logging if others are found.
2748 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
2749 returned from GIFCONF when Exim is trying to find the list of interfaces on
2750 a host. The code in os.c has been modified to copy each ifreq to an aligned
2751 structure in all cases.
2753 Also, in some cases, the returned ifreq's were being copied to a 'struct
2754 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
2755 means the last couple of bytes of an IPv6 address could be chopped if the
2756 ifreq contained only a normal sockaddr (14 bytes storage).
2758 18. Named domain lists were not supported in the hosts_treat_as_local option.
2759 An entry such as +xxxx was not recognized, and was treated as a literal
2762 19. Ensure that header lines added by a DATA ACL are included in the reject log
2763 if the ACL subsequently rejects the message.
2765 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
2766 MD5 (which is deprecated).
2768 21. When testing a filter file using -bf, Exim was writing a message when it
2769 took the sender from a "From " line in the message, but it was not doing so
2770 when it took $return_path from a Return-Path: header line. It now does.
2772 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
2773 with a valid header line field name (a series of printing characters
2774 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
2776 23. Changed "disc" in the source to "disk" to conform to the documentation and
2777 the book and for uniformity.
2779 24. Ignore Sendmail's -Ooption=value command line item.
2781 25. When execve() failed while trying to run a command in a pipe transport,
2782 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
2783 could be confused with a return value of 69 from the command itself. This
2784 has been changed to 127, the value the shell returns if it is asked to run
2785 a non-existent command. The wording for the related log line suggests a
2786 non-existent command as the problem.
2788 26. If received_header_text expands to an empty string, do not add a Received:
2789 header line to the message. (Well, it adds a token one on the spool, but
2790 marks it "old" so that it doesn't get used or transmitted.)
2792 27. Installed eximstats 1.28 (addition of -nt option).
2794 28. There was no check for failure on the call to getsockname() in the daemon
2795 code. This can fail if there is a shortage of resources on the system, with
2796 ENOMEM, for example. A temporary error is now given on failure.
2798 29. Contrary to the C standard, it seems that in some environments, the
2799 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
2800 program. Exim now does this explicitly; it affects the formatting of
2801 timestamps using strftime().
2803 30. If exiqsumm was given junk data, it threw up some uninitialized variable
2804 complaints. I've now initialized all the variables, to avoid this.
2806 32. Header lines added by a system filter were not being "seen" during
2807 transport-time rewrites.
2809 33. The info_callback() function passed to OpenSSL is set up with type void
2810 (*)(SSL *, int, int), as described somewhere. However, when calling the
2811 function (actually a macro) that sets it up, the type void(*)() is
2812 expected. I've put in a cast to prevent warnings from picky compilers.
2814 34. If a DNS black list lookup found a CNAME record, but there were no A
2815 records associated with the domain it pointed at, Exim crashed.
2817 35. If a DNS black list lookup returned more than one A record, Exim ignored
2818 all but the first. It now scans all returned addresses if a particular IP
2819 value is being sought. In this situation, the contents of the
2820 $dnslist_value variable are a list of all the addresses, separated by a
2823 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
2824 to accept a host name and all its aliases if the forward lookup for any of
2825 them yielded the IP address of the incoming connection. Now it accepts only
2826 those names whose forward lookup yields the correct IP address. Any other
2827 names are discarded. This closes a loophole whereby a rogue DNS
2828 administrator could create reverse DNS records to break through a
2829 wildcarded host restriction in an ACL.
2831 37. If a user filter or a system filter that ran in a subprocess used any of
2832 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
2833 the wrong values were passed to the pipe command ($thisaddress had the
2834 value of $0, $0 had the value of $1, etc). This bug was introduced by
2835 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
2837 38. Improved the line breaking for long SMTP error messages from ACLs.
2838 Previously, if there was no break point between 40 and 75 characters, Exim
2839 left the rest of the message alone. Two changes have been made: (a) I've
2840 reduced the minimum length to 35 characters; (b) if it can't find a break
2841 point between 35 and 75 characters, it looks ahead and uses the first one
2842 that it finds. This may give the occasional overlong line, but at least the
2843 remaining text gets split now.
2845 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
2846 file descriptors might be low, and that setting 1000 would always raise it.
2847 It turns out that in some environments, the limit is already over 1000 and
2848 that lowering it causes trouble. So now Exim takes care not to decrease it.
2850 40. When delivering a message, the value of $return_path is set to $sender_
2851 address at the start of routing (routers may change the value). By an
2852 oversight, this default was not being set up when an address was tested by
2853 -bt or -bv, which affected the outcome if any router or filter referred to
2856 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
2857 log only when "message" or "log_message" are set. However, if one of the
2858 conditions was an address verification, or a call to a nested ACL, the
2859 messages generated by the underlying test were being passed through. This
2860 no longer happens. The underlying message is available in $acl_verify_
2861 message for both "message" and "log_message" expansions, so it can be
2862 passed through if needed.
2864 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
2865 new expansion $bh_ to give the encoded byte string without charset
2866 translation. Translation happens only if iconv() is available; HAVE_ICONV
2867 indicates this at build time. HEADERS_CHARSET gives the charset to
2868 translate to; headers_charset can change it in the configuration, and
2869 "headers charset" can change it in an individual filter file.
2871 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
2872 that creates RFC 2047 encoded "words" labels them as that charset instead
2873 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
2874 expansion operator; (ii) when Exim creates a From: line for a local
2875 message; (iii) when a header line is rewritten to include a "phrase" part.
2877 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
2878 buggy, causing it to skip the first lines of messages whose message ID
2879 ended in 'D'. This would not have bitten before Exim release 4.14, because
2880 message IDs were unlikely to end in 'D' before then. The effect was to have
2881 incorrect size information for certain domains.
2883 45. #include "config.h" was missing at the start of the crypt16.c module. This
2884 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
2887 46. If there was a timeout during a "random" callout check, Exim treated it as
2888 a failure of the random address, and carried on sending RSET and the real
2889 address. If the delay was just some slowness somewhere, the response to the
2890 original RCPT would be taken as a response to RSET and so on, causing
2891 mayhem of various kinds.
2893 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
2894 when I implemented it. It didn't allow for the fact that some option values
2895 may legitimatetly be negative (e.g. size_addition), and it didn't even do
2896 the right test for positive values.
2898 48. Domain names in DNS records are case-independent. Exim always looks them up
2899 in lower case. Some resolvers return domain names in exactly the case they
2900 appear in the zone file, that is, they may contain uppercase letters. Not
2901 all resolvers do this - some return always lower case. Exim was treating a
2902 change of case by a resolver as a change of domain, similar to a widening
2903 of a domain abbreviation. This triggered its re-routing code and so it was
2904 trying to route what was effectively the same domain again. This normally
2905 caused routing to fail (because the router wouldn't handle the domain
2906 twice). Now Exim checks for this case specially, and just changes the
2907 casing of the domain that it ultimately uses when it transmits the message
2910 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
2913 50. If a filter generated a file delivery with a non-absolute name (possible if
2914 no home directory exists for the router), the forbid_file option was not
2917 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
2918 the existing equality matching.
2920 52. Exim was using ints instead of ino_t variables in some places where it was
2921 dealing with inode numbers.
2923 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
2924 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
2925 called TMPDIR, and if it finds it is different, it changes its value.
2927 54. The smtp_printf() function is now made available to local_scan() so
2928 additional output lines can be written before returning. There is also an
2929 smtp_fflush() function to enable the detection of a dropped connection.
2930 The variables smtp_input and smtp_batched_input are exported to
2933 55. Changed the default runtime configuration: the message "Unknown user"
2934 has been removed from the ACL, and instead placed on the localuser router,
2935 using the cannot_route_message feature. This means that any verification
2936 failures that generate their own messages won't get overridden. Similarly,
2937 the "Unrouteable address" message that was in the ACL for unverifiable
2938 relay addresses has also been removed.
2940 56. Added hosts_avoid_esmtp to the smtp transport.
2942 57. The exicyclog script was not checking for the esoteric option
2943 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2944 will work only if exicyclog is run under the appropriate euid.
2946 58. Following a discussion on the list, the rules by which Exim recognises line
2947 endings on incoming messages have been changed. The -dropcr and drop_cr
2948 options are now no-ops, retained only for backwards compatibility. The
2949 following line terminators are recognized: LF CRLF CR. However, special
2950 processing applies to CR:
2952 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2953 nor a local message in the state where . is a terminator.
2955 (ii) If a bare CR is encountered in a header line, an extra space is added
2956 after the line terminator so as not to end the header. The reasoning
2957 behind this is that bare CRs in header lines are most likely either
2958 to be mistakes, or people trying to play silly games.
2960 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
2961 was being incorrectly given as 18 bytes larger than it should have been.
2962 This is a VOB (very old bug).
2964 60. This may never have affected anything current, but just in case it has:
2965 When the local host is found other than at the start of a list of hosts,
2966 the local host, those with the same MX, and any that follow, are discarded.
2967 When the list in question was part of a longer list of hosts, the following
2968 hosts (not currently being processed) were also being discarded. This no
2969 longer happens. I'm not sure if this situation could ever has previously
2972 61. Added the "/MX" feature to lists of hosts in the manualroute and query
2975 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
2976 header. This is something that is recommended in a new Internet Draft, and
2977 is something that is documented as being done by Sendmail. There are two
2978 possible values. For messages generated by the autoreply transport, Exim
2981 Auto-Submitted: auto-replied
2983 whereas for all other generated messages (e.g. bounces) it adds
2985 Auto-Submitted: auto-generated
2987 63. The "personal" condition in filters now includes a test for the
2988 Auto-Submitted: header. If it contains the string "auto-" the message it
2989 not considered personal.
2991 64. Added rcpt_include_affixes as a generic transport option.
2993 65. Added queue_only_override (default true).
2995 66. Added the syslog_duplication option.
2997 67. If what should have been the first header line of a message consisted of
2998 a space followed by a colon, Exim was mis-interpreting it as a header line.
2999 It isn't of course - it is syntactically invalid and should therefore be
3000 treated as the start of the message body. The misbehaviour could have
3001 caused a number of strange effects, including loss of data in subsequent
3002 header lines, and spool format errors.
3004 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3005 client host had authenticated. This control can now be exercised by an ACL
3006 for more flexibility.
3008 69. By default, callouts do not happen when testing with -bh. There is now a
3009 variant, -bhc, which does actually run the callout code, including
3010 consulting and updating the callout cache.
3012 70. Added support for saslauthd authentication, courtesy of Alexander
3015 71. If statvfs() failed on the spool or log directories while checking their
3016 size for availability, Exim confusingly gave the error "space shortage".
3017 Furthermore, in debugging mode it crashed with a floating point exception.
3018 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3019 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3020 serious problem, Exim now writes to the main and panic logs when this
3021 happens, with details of the failure. It then refuses to accept the
3022 incoming message, giving the message "spool directory problem" or "log
3023 directory problem" with a 421 code for SMTP messages.
3025 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3026 0, 1, and 2 exist, because some OS complain for execs without them (see
3027 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3028 descriptors. However, the code omitted to check that the open succeeded,
3029 causing mysterious errors if for some reason the permissions on /dev/null
3030 got screwed. Now Exim writes a message to the main and panic logs, and
3031 bombs out if it can't open /dev/null.
3033 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3034 interact so that it is all more flexible. It is supposed to remain
3035 backwards compatible. Also added extra_local_interfaces.
3037 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3038 to bomb out with an assertion failure - to the client this appears as a
3039 connection drop. This problem occurs in the part of the code that was taken
3040 from the Samba project. Fortunately, the assertion is in a very simple
3041 function, so I have fixed this by reproducing the function inline in the
3042 one place where it is called, and arranging for authentication to fail
3043 instead of killing the process with assert().
3045 75. The SPA client code was not working when the server requested OEM rather
3046 than Unicode encoding.
3048 76. Added code to make require_files with a specific uid setting more usable in
3049 the case where statting the file as root fails - usually a non-root-mounted
3050 NFS file system. When this happens and the failure is EACCES, Exim now
3051 forks a subprocess and does the per-uid checking as the relevant uid.
3053 77. Added process_log_path.
3055 78. If log_file_path was not explicitly set, a setting of check_log_space or
3056 check_log_inodes was ignored.
3058 79. If a space check for the spool or log partitions fails, the incident is now
3059 logged. Of course, in the latter case the data may get lost...
3061 80. Added the %p formatting code to string_format() so that it can be used to
3062 print addresses in debug_print(). Adjusted all the address printing in the
3063 debugging in store.c to use %p rather than %d.
3065 81. There was a concern that a line of code in smtp_in.c could overflow a
3066 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3067 initially expressed, the concern was not well-founded, because trailing
3068 spaces are removed early. However, if the trailing spaces were followed by
3069 a NULL, they did not get removed, so the overflow was possible. Two fixes
3072 (a) I re-wrote the offending code in a cleaner fashion.
3073 (b) If an incoming SMTP command contains a NULL character, it is rejected
3076 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3077 runs initgroups(), so that if the Exim user is in any additional groups,
3078 they will be used during message reception.
3084 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob