1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.173 2005/06/27 10:52:20 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
10 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
12 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
13 command itself ended in error, the underlying message about the transport
14 filter timeout was being overwritten with the pipe command error. Now the
15 underlying error message should be appended to the second error message.
17 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
19 PH/02 Reduce the amount of output that "make" produces by default. Full output
20 can still be requested.
22 PH/03 The warning log line about a condition test deferring for a "warn" verb
23 was being output only once per connection, rather than after each
24 occurrence (because it was using the same function as for successful
25 "warn" verbs). This seems wrong, so I have changed it.
27 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
28 it should not have, which might have caused a crash in the right
29 circumstances, but probably never did.
31 PH/04 Installed a modified version of Tony Finch's patch to make submission
32 mode fix the return path as well as the Sender: header line, and to
33 add a /name= option so that you can make the user's friendly name appear
36 TF/03 Added the control = fakedefer ACL modifier.
38 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
39 Mark Lowes for thorough testing.
41 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
43 TK/03 Merged latest SRS patch from Miles Wilton.
45 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
46 with the definition in sysexits.h (which is #included earlier).
47 Fortunately, Exim does not actually use EX_OK. The code used to try to
48 preserve the sysexits.h value, by assumimg that macro definitions were
49 scanned for macro replacements. I have been disabused of this notion,
50 so now the code just undefines EX_OK before #including unistd.h.
52 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
53 in the smtp transport. When a block could not be written in a single
54 write() function, the timeout was being re-applied to each part-write.
55 This seems wrong - if the receiver was accepting one byte at a time it
56 would take for ever. The timeout is now adjusted when this happens. It
57 doesn't have to be particularly precise.
59 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
60 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
62 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
63 without a domain if the check to <postmaster@domain> fails.
65 SC/01 Eximstats: added -xls and the ability to specify output files
66 (patch written by Frank Heydlauf).
68 SC/02 Eximstats: use FileHandles for outputing results.
70 SC/03 Eximstats: allow any combination of xls, txt, and html output.
72 SC/04 Eximstats: fixed display of large numbers with -nvr option
74 SC/05 Eximstats: fixed merging of reports with empty tables.
76 SC/06 Eximstats: added the -include_original_destination flag
78 SC/07 Eximstats: removed tabs and trailing whitespace.
80 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
82 TK/06 MBOX spool code: Add real "From " MBOX separator line
83 so the .eml file is really in mbox format (even though
84 most programs do not really care). Patch from Alex Miller.
86 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
87 The latter is generated from $received_to and is only set if the
88 message has one envelope recipient. SA can use these headers,
89 obviously out-of-the-box. Patch from Alex Miller.
91 PH/08 The ${def test on a variable was returning false if the variable's
92 value was "0", contrary to what the specification has always said!
93 The result should be true unless the variable is empty.
95 PH/09 The syntax error of a character other than { following "${if
96 def:variable_name" (after optional whitespace) was not being diagnosed.
97 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
98 accidental colon was present, for example, could give incorrect results.
100 PH/10 Tidied the code in a number of places where the st_size field of a stat()
101 result is used (not including appendfile, where other changes are about
104 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
105 This involved changing a lot of size variables from int to off_t. It
106 should work with maildirs and everything.
108 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
109 spamd dying while we are connected to it.
111 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
114 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
115 function for BDB 4.3.
117 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
119 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
120 the sources for PCRE have been re-arranged and more files are now
123 PH/15 The code I had for printing potentially long long variables in PH/11
124 above was not the best (it lost precision). The length of off_t variables
125 is now inspected at build time, and an appropriate printing format (%ld
126 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
127 to be "long long int" or "long int". This is needed for the internal
128 formatting function string_vformat().
130 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
131 the configuration file to be ":syslog", then the script "guesses" where
132 the logs files are, rather than using the compiled in default. In our
133 case the guess is not the same as the compiled default, so the script
134 suddenly stopped working when I started to use syslog. The patch checks
135 to see if log_file_path is "". If so, it attempts to read it from exim
136 with no configuration file to get the compiled in version, before it
137 falls back to the previous guessing code."
139 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
140 implementing BATV in an Exim configuration. See NewStuff for the gory
143 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
144 Makefile that are specific to HP-UX.
146 PH/18 If the "use_postmaster" option was set for a recipient callout together
147 with the "random" option, the postmaster address was used as the MAIL
148 FROM address for the random test, but not for the subsequent recipient
149 test. It is now used for both.
151 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
152 patch removes a few documentation additions to RFC 3028, because the
153 latest draft now contains them. It adds the new en;ascii-case comparator
154 and a new error check for 8bit text in MIME parts. Comparator and
155 require names are now matched exactly. I enabled the subaddress
156 extension, but it is not well tested yet (read: it works for me)."
158 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
159 rework some of the code of TK/09 above to avoid the hardwired use of
160 "%lld" and "long long". Replaced the call to snprintf() with a call to
163 PH/21 Added another message to those in 4.51/PH/42, namely "All relevant MX
164 records point to non-existent hosts".
166 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
167 experimental DomainKeys support:
169 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
170 (2) On an error such as an illegally used "control", the wrong name for
171 the control was given.
173 These problems did NOT occur unless DomainKeys support was compiled.
175 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
177 PH/24 Added ${if match_ip condition.
179 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
180 databases so that it will be absolutely obvious if a crash occurs in the
181 DB library. This is a regular occurrence (often caused by mis-matched
184 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
185 chmod(), fcntl(), and sscanf(). These were picked up on a user's system
186 that detects such things. There doesn't seem to be a gcc warning option
187 for this - only an attribute that has to be put on the function's
188 prototype. I'm sure I haven't caught all of these, but it's a start.
190 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
191 specify a transport. However, if an address that was verified by such a
192 router was the subject of a callout, Exim crashed because it tried to
193 read the rcpt_include_affixes from the non-existent transport. Now it
194 just assumes that the setting of that option is false. This bug was
195 introduced by 4.51/PH/31.
197 PH/28 Changed -d+all to exclude +memory, because that information is very
198 rarely of interest, but it makes the output a lot bigger. People tend to
199 do -d+all out of habit.
201 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
202 code in os-type was giving problems when libc.so lives in lib64, like on
205 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
206 aren't the modern standard, and it seems that some systems' include files
207 don't always have them. Exim was already checking for some of the newer
208 ones like T_AAAA, and defining it itself. I've added checks for all the
209 record types that Exim uses.
216 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
217 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
219 TK/02 Fix ACL "control" statement not being available in MIME ACL.
221 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
223 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
224 to test Sieve filters that use "vacation".
226 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
227 that changes the way the GnuTLS parameters are stored in the cache file.
228 The new format can be generated externally. For backward compatibility,
229 if the data in the cache doesn't make sense, Exim assumes it has read an
230 old-format file, and it generates new data and writes a new file. This
231 means that you can't go back to an older release without removing the
234 PH/03 A redirect router that has both "unseen" and "one_time" set does not
235 work if there are any delivery delays because "one_time" forces the
236 parent to be marked "delivered", so its unseen clone is never tried
237 again. For this reason, Exim now forbids the simultaneous setting of
240 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
241 redirected to themselves ("homonym" addresses). Read the long ChangeLog
242 entry if you want to know the details. The fix, however, neglected to
243 consider the case when local delivery batching is involved. The test for
244 "previously delivered" was not happening when checking to see if an
245 address could be batched with a previous (undelivered) one; under
246 certain circumstances this could lead to multiple deliveries to the same
249 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
250 in its include files, and this causes problems building Exim.
252 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
253 header_syntax) but Exim was just ignoring anything given after a slash.
254 In particular, this caused confusion with an attempt to use "verify =
255 reverse_host_lookup/defer_ok". An error is now given when options are
256 supplied for verify items that do not have them. (Maybe reverse_host_
257 lookup should have a defer_ok option, but that's a different point.)
259 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
260 defined by RFC 821) to 2048, because there were problems with some AUTH
261 commands, and RFC 1869 says the size should be increased for extended
262 SMTP commands that take arguments.
264 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
267 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
268 "unknown" error; now it says that the functionality isn't in the binary.
270 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
271 an address' error message when a string expansion fails (syntax or
272 whatever). Otherwise the password may appear in the log. Following change
273 PH/42 below, there is no longer a chance of it appearing in a bounce
276 PH/11 Installed exipick version 20050225.0 from John Jetmore.
278 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
279 of its addresses was ever tried. (Bugzilla bug #2.)
281 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
282 the result incorrectly in the debug output. (It correctly added a newline
283 to what was transported.)
285 TF/01 Added $received_time.
287 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
288 commented out examples of how to interface to a virus scanner and to
289 SpamAssassin. Also added commented examples of av_scanner and
290 spamd_address settings.
292 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
293 and controls are allowed in which ACLs. There were a couple of minor
294 errors. Some of the entries in the conditions table (which is a table of
295 where they are NOT allowed) were getting very unwieldy; rewrote them as a
296 negation of where the condition IS allowed.
298 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
300 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
301 header file does not have a version number, so I've had to invent a new
302 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
303 API. The code is untested by me (my Linux distribution still has 0.3.2 of
304 radiusclient), but it was contributed by a Radius user.
306 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
307 files or directories, for OpenSSL.
309 PH/19 When an Exim process that is running as root has to create an Exim log
310 file, it does so in a subprocess that runs as exim:exim so as to get the
311 ownership right at creation (otherwise, other Exim processes might see
312 the file with the wrong ownership). There was no test for failure of this
313 fork() call, which would lead to the process getting stuck as it waited
314 for a non-existent subprocess. Forks do occasionally fail when resources
315 run out. I reviewed all the other calls to fork(); they all seem to check
318 PH/20 When checking for unexpected SMTP input at connect time (before writing
319 the banner), Exim was not dealing correctly with a non-positive return
320 from the read() function. If the client had disconnected by this time,
321 the result was a log entry for a synchronization error with an empty
322 string after "input=" when read() returned zero. If read() returned -1
323 (an event I could not check), uninitialized data bytes were printed.
324 There were reports of junk text (parts of files, etc) appearing after
327 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
329 PH/22 Added support for macro redefinition, and (re)definition in between
330 driver and ACL definitions.
332 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
333 forgetting to use the resulting value; it was using the unexpanded value.
335 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
336 hadn't been configured. The fix is from Juergen Kreileder, who
337 understands it better than I do:
339 "Here's what I see happening with three configured cyrus_sasl
340 authenticators configured (plain, login, cram-md5):
342 On startup auth_cyrus_sasl_init() gets called for each of these.
343 This means three calls to sasl_listmech() without a specified mech_list.
344 => SASL tests which mechs of all available mechs actually work
345 => three warnings about OTP not working
346 => the returned list contains: plain, login, cram-md5, digest-md5, ...
348 With the patch, sasl_listmech() also gets called three times. But now
349 SASL's mech_list option is set to the server_mech specified in the the
350 authenticator. Or in other words, the answer from sasl_listmech()
351 gets limited to just the mech you're testing for (which is different
353 => the return list contains just 'plain' or 'login', 'cram-md5' or
354 nothing depending on the value of ob->server_mech.
356 I've just tested the patch: Authentication still works fine,
357 unavailable mechs specified in the exim configuration are still
358 caught, and the auth.log warnings about OTP are gone."
360 PH/25 When debugging is enabled, the contents of the command line are added
361 to the debugging output, even when log_selector=+arguments is not
364 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
365 answer is "GNU", and only if the return is "GNU/something" is the answer
368 PH/27 $acl_verify_message is now set immediately after the failure of a
369 verification in an ACL, and so is available in subsequent modifiers. In
370 particular, the message can be preserved by coding like this:
372 warn !verify = sender
373 set acl_m0 = $acl_verify_message
375 Previously, $acl_verify_message was set only while expanding "message"
376 and "log_message" when a very denied access.
378 PH/28 Modified OS/os.c-Linux with
380 -#ifndef OS_LOAD_AVERAGE
381 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
383 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
384 nomenclature these days.)
386 PH/29 Installed patch from the Sieve maintainer that adds the options
387 sieve_useraddress and sieve_subaddress to the redirect router.
389 PH/30 In these circumstances:
390 . Two addresses routed to the same list of hosts;
391 . First host does not offer TLS;
392 . First host accepts first address;
393 . First host gives temporary error to second address;
394 . Second host offers TLS and a TLS session is established;
395 . Second host accepts second address.
396 Exim incorrectly logged both deliveries with the TLS parameters (cipher
397 and peerdn, if requested) that were in fact used only for the second
400 PH/31 When doing a callout as part of verifying an address, Exim was not paying
401 attention to any local part prefix or suffix that was matched by the
402 router that accepted the address. It now behaves in the same way as it
403 does for delivery: the affixes are removed from the local part unless
404 rcpt_include_affixes is set on the transport.
406 PH/32 Add the sender address, as F=<...>, to the log line when logging a
407 timeout during the DATA phase of an incoming message.
409 PH/33 Sieve envelope tests were broken for match types other than :is. I have
410 applied a patch sanctioned by the Sieve maintainer.
412 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
413 the uid or gid is negative. A case of a negative gid caused this to be
414 noticed. The fix allows for either to be negative.
416 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
417 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
420 PH/36 The change PH/12 above was broken. Fixed it.
422 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
423 the grounds that routing the same address twice would always produce the
424 same answer. This might have been true once, but it is certainly no
425 longer true now. Routing a child address may depend on the previous
426 routing that produced that child. Some complicated redirection strategies
427 went wrong when messages had multiple recipients, and made Exim's
428 behaviour dependent on the order in which the addresses were given.
430 I have moved the duplicate checking until after the routing is complete.
431 Exim scans the addresses that are assigned to local and remote
432 transports, and removes any duplicates. This means that more work will be
433 done, as duplicates will always all be routed, but duplicates are
434 presumably rare, so I don't expect this is of any significance.
436 For deliveries to pipes, files, and autoreplies, the duplicate checking
437 still happens during the routing process, since they are not going to be
440 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
441 It corrects a timeout issue with spamd. This is Ian's comment: "The
442 background is that sometimes spamd either never reads data from a
443 connection it has accepted, or it never writes response data. The exiscan
444 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
445 blindly assumes that writes won't block so it may never time out."
447 PH/39 Allow G after quota size as well as K and M.
449 PH/40 The value set for $authenticated_id in an authenticator may not contain
450 binary zeroes or newlines because the value is written to log lines and
451 to spool files. There was no check on this. Now the value is run through
452 the string_printing() function so that such characters are converted to
453 printable escape sequences.
455 PH/41 $message_linecount is a new variable that contains the total number of
456 lines in the message. Compare $body_linecount, which is the count for the
459 PH/42 Exim no longer gives details of delivery errors for specific addresses in
460 bounce and delay warning messages, except in certain special cases, which
463 (a) An SMTP error message from a remote host;
464 (b) A message specified in a :fail: redirection;
465 (c) A message specified in a "fail" command in a system filter;
466 (d) A message specified in a FAIL return from the queryprogram router;
467 (e) A message specified by the cannot_route_message router option.
469 In these cases only, Exim does include the error details in bounce and
470 warning messages. There are also a few cases where bland messages such
471 as "unrouteable address" or "local delivery error" are given.
473 PH/43 $value is now also set for the "else" part of a ${run expansion.
475 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
476 being worked on, but at least Exim now implements the latest version to
479 PH/45 In a pipe transport, although a timeout while waiting for the pipe
480 process to complete was treated as a delivery failure, a timeout while
481 writing the message to the pipe was logged, but erroneously treated as a
482 successful delivery. Such timeouts include transport filter timeouts. For
483 consistency with the overall process timeout, these timeouts are now
484 treated as errors, giving rise to delivery failures by default. However,
485 there is now a new Boolean option for the pipe transport called
486 timeout_defer, which, if set TRUE, converts the failures into defers for
487 both kinds of timeout. A transport filter timeout is now identified in
490 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
491 systems where "make" and "gmake" are different, calling "gmake" at top
492 level broke things. I've arranged for the value of $(MAKE) to be passed
493 from the Makefile to this script so that it can call the same version of
497 A note about Exim versions 4.44 and 4.50
498 ----------------------------------------
500 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
501 changes of various kinds. As a consequence, a big documentation update was
502 needed. This delayed the release for rather longer than seemed good, especially
503 in the light of a couple of (minor) security issues. Therefore, the changes
504 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
505 release. So 4.44 and 4.50 are in effect two different branches that both start
508 I have left the 4.50 change log unchanged; it contains all the changes since
509 4.43. The change log for 4.44 is below; many of its items are identical to
510 those for 4.50. This seems to be the most sensible way to preserve the
511 historical information.
517 1. Minor wording change to the doc/README.SIEVE file.
519 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
520 computation of the current number of files was incorrect.
522 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
523 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
526 4. Give more explanation in the error message when the command for a transport
527 filter fails to execute.
529 5. There are several places where Exim runs a non-Exim command in a
530 subprocess. The SIGUSR1 signal should be disabled for these processes. This
531 was being done only for the command run by the queryprogram router. It is
532 now done for all such subprocesses. The other cases are: ${run, transport
533 filters, and the commands run by the lmtp and pipe transports.
535 6. Added CONFIGURE_GROUP build-time option.
537 7. Some older OS have a limit of 256 on the maximum number of file
538 descriptors. Exim was using setrlimit() to set 1000 as a large value
539 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
540 systems. I've change it so that if it can't get 1000, it tries for 256.
542 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
543 was an oversight, and furthermore, ever since the addition of extra
544 controls (e.g. 4.43/32), the checks on when to allow different forms of
545 "control" were broken. There should now be diagnostics for all cases when a
546 control that does not make sense is encountered.
548 9. Added the /retain_sender option to "control=submission".
550 10. $recipients is now available in the predata ACL (oversight).
552 11. Tidy the search cache before the fork to do a delivery from a message
553 received from the command line. Otherwise the child will trigger a lookup
554 failure and thereby defer the delivery if it tries to use (for example) a
555 cached ldap connection that the parent has called unbind on.
557 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
558 of $address_data from the recipient verification was clobbered by the
561 13. The value of address_data from a sender verification is now available in
562 $sender_address_data in subsequent conditions in the ACL statement.
564 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
566 15. Added a new option "connect=<time>" to callout options, to set a different
569 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
570 was its contents. (It was OK if the option was not defined at all.)
572 17. A "Completed" log line is now written for messages that are removed from
573 the spool by the -Mrm option.
575 18. New variables $sender_verify_failure and $recipient_verify_failure contain
576 information about exactly what failed.
578 19. Added -dd to debug only the daemon process.
580 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
581 handles timeouts, both on the server side and network timeouts. Renamed the
582 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
584 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
585 It is now set to "smtps".
587 22. $host_address is now set to the target address during the checking of
590 23. When checking ignore_target_hosts for an ipliteral router, no host name was
591 being passed; this would have caused $sender_host_name to have been used if
592 matching the list had actually called for a host name (not very likely,
593 since this list is usually IP addresses). A host name is now passed as
596 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
597 code that specifies a non-restarting handler (typically sigaction() in
598 modern systems) in an attempt to fix a rare and obscure crash bug.
600 25. Narrowed the window for a race in the daemon that could cause it to ignore
601 SIGCHLD signals. This is not a major problem, because they are used only to
602 wake it up if nothing else does.
604 26. A malformed maildirsize file could cause Exim to calculate negative values
605 for the mailbox size or file count. Odd effects could occur as a result.
606 The maildirsize information is now recalculated if the size or filecount
609 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
610 support for a long time. Removed HAVE_SYS_VFS_H.
612 28. Installed the latest version of exipick from John Jetmore.
614 29. In an address list, if the pattern was not a regular expression, an empty
615 subject address (from a bounce message) matched only if the pattern was an
616 empty string. Non-empty patterns were not even tested. This was the wrong
617 because it is perfectly reasonable to use an empty address as part of a
618 database query. An empty address is now tested by patterns that are
619 lookups. However, all the other forms of pattern expect the subject to
620 contain a local part and a domain, and therefore, for them, an empty
621 address still always fails if the pattern is not itself empty.
623 30. Exim went into a mad DNS loop when attempting to do a callout where the
624 host was specified on an smtp transport, and looking it up yielded more
627 31. Re-factored the code for checking spool and log partition space into a
628 function that finds that data and another that does the check. The former
629 is then used to implement four new variables: $spool_space, $log_space,
630 $spool_inodes, and $log_inodes.
632 32. The RFC2047 encoding function was originally intended for short strings
633 such as real names; it was not keeping to the 75-character limit for
634 encoded words that the RFC imposes. It now respects the limit, and
635 generates multiple encoded words if necessary. To be on the safe side, I
636 have increased the buffer size for the ${rfc2047: expansion operator from
639 33. It is now permitted to omit both strings after an "if" condition; if the
640 condition is true, the result is "true". As before, when the second string
641 is omitted, a false condition yields an empty string. This makes it less
642 cumbersome to write custom ACL and router conditions.
644 34. Failure to deliver a bounce message always caused it to be frozen, even if
645 there was an errors_to setting on the router. The errors_to setting is now
648 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
649 canonical form (fully expanded) before being placed in
650 $sender_host_address.
652 36. The table in the code that translates DNS record types into text (T_A to
653 "A" for instance) was missing entries for NS and CNAME. It is just possible
654 that this could have caused confusion if both these types were looked up
655 for the same domain, because the text type is used as part of Exim's
656 per-process caching. But the chance of anyone hitting this buglet seems
659 37. The dnsdb lookup has been extended in a number of ways.
661 (1) There is a new type, "zns", which walks up the domain tree until it
662 finds some nameserver records. It should be used with care.
664 (2) There is a new type, "mxh", which is like "mx" except that it returns
665 just the host names, not the priorities.
667 (3) It is now possible to give a list of domains (or IP addresses) to be
668 looked up. The behaviour when one of the lookups defers can be
669 controlled by a keyword.
671 (4) It is now possible to specify the separator character for use when
672 multiple records are returned.
674 38. The dnslists ACL condition has been extended: it is now possible to supply
675 a list of IP addresses and/or domains to be looked up in a particular DNS
678 39. Added log_selector=+queue_time_overall.
680 40. When running the queue in the test harness, wait just a tad after forking a
681 delivery process, to get repeatability of debugging output.
683 41. Include certificate and key file names in error message when GnuTLS fails
684 to set them up, because the GnuTLS error message doesn't include the name
685 of the failing file when there is a problem reading it.
687 42. Allow both -bf and -bF in the same test run.
689 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
691 44. The "Exiscan patch" is now merged into the mainline Exim source.
693 45. Sometimes the final signoff response after QUIT could fail to get
694 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
695 < 0 before doing a fflush(). This bug looks as though it goes back to the
696 introduction of TLS in release 3.20, but "sometimes" must have been rare
697 because the tests only now provoked it.
699 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
700 (this can affect the format of dates).
702 47. exim_tidydb, when checking for the continued existence of a message for
703 which it has found a message-specific retry record, was not finding
704 messages that were in split spool directories. Consequently, it was
705 deleting retry records that should have stayed in existence.
707 48. Steve fixed some bugs in eximstats.
709 49. The SPA authentication driver was not abandoning authentication and moving
710 on to the next authenticator when an expansion was forced to fail,
711 contradicting the general specification for all authenticators. Instead it
712 was generating a temporary error. It now behaves as specified.
714 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
715 (the order specifies the preference for clients). The order is now AES256,
716 AES128, 3DES, ARCFOUR128.
718 51. Small patch to Sieve code - explicitly set From: when generating an
721 52. Exim crashed if a remote delivery caused a very long error message to be
722 recorded - for instance if somebody sent an entire SpamAssassin report back
723 as a large number of 550 error lines. This bug was coincidentally fixed by
724 increasing the size of one of Exim's internal buffers (big_buffer) that
725 happened as part of the Exiscan merge. However, to be on the safe side, I
726 have made the code more robust (and fixed the comments that describe what
729 53. Now that there can be additional text after "Completed" in log lines (if
730 the queue_time_overall log selector is set), a one-byte patch to exigrep
731 was needed to allow it to recognize "Completed" as not the last thing in
734 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
735 patch that reportedly fixes this has been added. I am not expert enough to
736 create a test for it. This is what the patch creator wrote:
738 "I found a little strange behaviour of ldap code when working with
739 Windows 2003 AD Domain, where users was placed in more than one
740 Organization Units. When I tried to give exim partial DN, the exit code
741 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
742 But simultaneously result of request was absolutely normal ldap result,
743 so I produce this patch..."
745 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
746 REFERENCE, so I have modified the code to exclude the patch when that macro
749 55. Some experimental protocols are using DNS PTR records for new purposes. The
750 keys for these records are domain names, not reversed IP addresses. The
751 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
752 leaves it alone. Component reversal etc. now happens only for IP addresses.
755 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
757 57. Double the size of the debug message buffer (to 2048) so that more of very
758 long debug lines gets shown.
760 58. The exicyclog utility now does better if the number of log files to keep
761 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
763 59. Two changes related to the smtp_active_hostname option:
765 (1) $smtp_active_hostname is now available as a variable.
766 (2) The default for smtp_banner uses $smtp_active_hostname instead
767 of $primary_hostname.
769 60. The host_aton() function is supposed to be passed a string that is known
770 to be a valid IP address. However, in the case of IPv6 addresses, it was
771 not checking this. This is a hostage to fortune. Exim now panics and dies
772 if the condition is not met. A case was found where this could be provoked
773 from a dnsdb PTR lookup with an IPv6 address that had more than 8
774 components; fortuitously, this particular loophole had already been fixed
775 by change 4.50/55 above.
777 If there are any other similar loopholes, the new check in host_aton()
778 itself should stop them being exploited. The report I received stated that
779 data on the command line could provoke the exploit when Exim was running as
780 exim, but did not say which command line option was involved. All I could
781 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
785 61. There was a buffer overflow vulnerability in the SPA authentication code
786 (which came originally from the Samba project). I have added a test to the
787 spa_base64_to_bits() function which I hope fixes it.
790 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
791 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
793 63. The daemon start-up calls getloadavg() while still root for those OS that
794 need the first call to be done as root, but it missed one case: when
795 deliver_queue_load_max is set with deliver_drop_privilege. This is
796 necessary for the benefit of the queue runner, because there is no re-exec
797 when deliver_drop_privilege is set.
799 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
802 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
803 was in use, was not putting the data itself into the right store pool;
804 consequently, it could be overwritten for a subsequent message in the same
805 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
808 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
810 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
811 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
812 the function were treating the return as a boolean value, which happened to
813 work because 0=false and not-0=true, but is not correct code.
815 68. The host_aton() function was not handling scoped IPv6 addresses (those
816 with, for example, "%eth0" on the end) correctly.
818 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
819 negated items (that is, ~something) in unsigned ints. Some compilers
820 apparently mutter when there is no cast.
822 70. If an address verification called from an ACL failed, and did not produce a
823 user-specific message (i.e. there was only a "system" message), nothing was
824 put in $acl_verify_message. In this situation, it now puts the system
827 71. Change 4.23/11 added synchronization checking at the start of an SMTP
828 session; change 4.31/43 added the unwanted input to the log line - except
829 that it did not do this in the start of session case. It now does.
831 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
832 This is wrong and can cause the other end to generate a synchronization
833 error if it is another Exim or anything else that does the synchronization
834 check. A QUIT command is no longer sent after a timeout.
836 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
839 74. The defer_ok option of callout verification was not working if it was used
840 when verifying addresses in header lines, that is, for this case:
842 verify = header_sender/callout=defer_ok
844 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
845 those file descriptors could be used for SMTP connections. If anything
846 wrote to stderr (the example that came up was "warn" in embedded Perl), it
847 could be sent to the SMTP client, causing chaos. The daemon now opens
848 stdin, stdout, and stderr to /dev/null when it puts itself into the
851 76. Arrange for output from Perl's "warn" command to be written to Exim's main
852 log by default. The user can override this with suitable Perl magic.
854 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
855 the log message when discard triggers, was not working for the DATA ACL or
856 for the non-SMTP ACL.
858 78. Error message wording change in sieve.c.
860 79. If smtp_accept_max_per_host was set, the number of connections could be
861 restricted to fewer than expected, because the daemon was trying to set up
862 a new connection before checking whether the processes handling previous
863 connections had finished. The check for completed processes is now done
864 earlier. On busy systems, this bug wouldn't be noticed because something
865 else would have woken the daemon, and it would have reaped the completed
868 80. If a message was submitted locally by a user whose login name contained one
869 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
870 It caused a spool format error. I have fixed the spool reading code. A
871 related problem was that the "from" clause in the Received: line became
872 illegal because of the space(s). It is now covered by ${quote_local_part.
874 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
877 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
879 83. Patch from Sieve maintainer to fix unterminated string problem in
882 84. Some minor changes to the Linux configuration files to help with other
883 OS variants using glibc.
885 85. One more patch for Sieve to update vacation handling to latest spec.
888 ----------------------------------------------------
889 See the note above about the 4.44 and 4.50 releases.
890 ----------------------------------------------------
896 1. Change 4.43/35 introduced a bug that caused file counts to be
897 incorrectly computed when quota_filecount was set in an appendfile
900 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
901 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
904 3. Give more explanation in the error message when the command for a transport
905 filter fails to execute.
907 4. There are several places where Exim runs a non-Exim command in a
908 subprocess. The SIGUSR1 signal should be disabled for these processes. This
909 was being done only for the command run by the queryprogram router. It is
910 now done for all such subprocesses. The other cases are: ${run, transport
911 filters, and the commands run by the lmtp and pipe transports.
913 5. Some older OS have a limit of 256 on the maximum number of file
914 descriptors. Exim was using setrlimit() to set 1000 as a large value
915 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
916 systems. I've change it so that if it can't get 1000, it tries for 256.
918 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
919 was an oversight, and furthermore, ever since the addition of extra
920 controls (e.g. 4.43/32), the checks on when to allow different forms of
921 "control" were broken. There should now be diagnostics for all cases when a
922 control that does not make sense is encountered.
924 7. $recipients is now available in the predata ACL (oversight).
926 8. Tidy the search cache before the fork to do a delivery from a message
927 received from the command line. Otherwise the child will trigger a lookup
928 failure and thereby defer the delivery if it tries to use (for example) a
929 cached ldap connection that the parent has called unbind on.
931 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
932 of $address_data from the recipient verification was clobbered by the
935 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
936 was its contents. (It was OK if the option was not defined at all.)
938 11. A "Completed" log line is now written for messages that are removed from
939 the spool by the -Mrm option.
941 12. $host_address is now set to the target address during the checking of
944 13. When checking ignore_target_hosts for an ipliteral router, no host name was
945 being passed; this would have caused $sender_host_name to have been used if
946 matching the list had actually called for a host name (not very likely,
947 since this list is usually IP addresses). A host name is now passed as
950 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
951 code that specifies a non-restarting handler (typically sigaction() in
952 modern systems) in an attempt to fix a rare and obscure crash bug.
954 15. Narrowed the window for a race in the daemon that could cause it to ignore
955 SIGCHLD signals. This is not a major problem, because they are used only to
956 wake it up if nothing else does.
958 16. A malformed maildirsize file could cause Exim to calculate negative values
959 for the mailbox size or file count. Odd effects could occur as a result.
960 The maildirsize information is now recalculated if the size or filecount
963 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
964 support for a long time. Removed HAVE_SYS_VFS_H.
966 18. Updated exipick to current release from John Jetmore.
968 19. Allow an empty sender to be matched against a lookup in an address list.
969 Previously the only cases considered were a regular expression, or an
972 20. Exim went into a mad DNS lookup loop when doing a callout where the
973 host was specified on the transport, if the DNS lookup yielded more than
976 21. The RFC2047 encoding function was originally intended for short strings
977 such as real names; it was not keeping to the 75-character limit for
978 encoded words that the RFC imposes. It now respects the limit, and
979 generates multiple encoded words if necessary. To be on the safe side, I
980 have increased the buffer size for the ${rfc2047: expansion operator from
983 22. Failure to deliver a bounce message always caused it to be frozen, even if
984 there was an errors_to setting on the router. The errors_to setting is now
987 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
988 canonical form (fully expanded) before being placed in
989 $sender_host_address.
991 24. Updated eximstats to version 1.33
993 25. Include certificate and key file names in error message when GnuTLS fails
994 to set them up, because the GnuTLS error message doesn't include the name
995 of the failing file when there is a problem reading it.
997 26. Expand error message when OpenSSL has problems setting up cert/key files.
1000 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1001 (this can affect the format of dates).
1003 28. exim_tidydb, when checking for the continued existence of a message for
1004 which it has found a message-specific retry record, was not finding
1005 messages that were in split spool directories. Consequently, it was
1006 deleting retry records that should have stayed in existence.
1008 29. eximstats updated to version 1.35
1009 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1010 1.35 - bugfix such that pie charts by volume are generated correctly
1012 30. The SPA authentication driver was not abandoning authentication and moving
1013 on to the next authenticator when an expansion was forced to fail,
1014 contradicting the general specification for all authenticators. Instead it
1015 was generating a temporary error. It now behaves as specified.
1017 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1018 (the order specifies the preference for clients). The order is now AES256,
1019 AES128, 3DES, ARCFOUR128.
1021 31. Small patch to Sieve code - explicitly set From: when generating an
1024 32. Exim crashed if a remote delivery caused a very long error message to be
1025 recorded - for instance if somebody sent an entire SpamAssassin report back
1026 as a large number of 550 error lines. This bug was coincidentally fixed by
1027 increasing the size of one of Exim's internal buffers (big_buffer) that
1028 happened as part of the Exiscan merge. However, to be on the safe side, I
1029 have made the code more robust (and fixed the comments that describe what
1032 33. Some experimental protocols are using DNS PTR records for new purposes. The
1033 keys for these records are domain names, not reversed IP addresses. The
1034 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1035 leaves it alone. Component reversal etc. now happens only for IP addresses.
1038 34. The host_aton() function is supposed to be passed a string that is known
1039 to be a valid IP address. However, in the case of IPv6 addresses, it was
1040 not checking this. This is a hostage to fortune. Exim now panics and dies
1041 if the condition is not met. A case was found where this could be provoked
1042 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1043 components; fortuitously, this particular loophole had already been fixed
1044 by change 4.50/55 or 4.44/33 above.
1046 If there are any other similar loopholes, the new check in host_aton()
1047 itself should stop them being exploited. The report I received stated that
1048 data on the command line could provoke the exploit when Exim was running as
1049 exim, but did not say which command line option was involved. All I could
1050 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1051 running as the user.
1054 35. There was a buffer overflow vulnerability in the SPA authentication code
1055 (which came originally from the Samba project). I have added a test to the
1056 spa_base64_to_bits() function which I hope fixes it.
1059 36. The daemon start-up calls getloadavg() while still root for those OS that
1060 need the first call to be done as root, but it missed one case: when
1061 deliver_queue_load_max is set with deliver_drop_privilege. This is
1062 necessary for the benefit of the queue runner, because there is no re-exec
1063 when deliver_drop_privilege is set.
1065 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1066 was in use, was not putting the data itself into the right store pool;
1067 consequently, it could be overwritten for a subsequent message in the same
1068 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1071 38. Sometimes the final signoff response after QUIT could fail to get
1072 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1073 < 0 before doing a fflush(). This bug looks as though it goes back to the
1074 introduction of TLS in release 3.20, but "sometimes" must have been rare
1075 because the tests only now provoked it.
1081 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1082 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1083 Now it returns an integer. A number of calls were still expecting a T/F
1084 return. Fortuitously, in all cases, the tests worked in OK situations,
1085 which is the norm. However, things would have gone wrong on any write
1086 failures on the smtp file descriptor. This function is used when sending
1087 messages over SMTP and also when doing verify callouts.
1089 2. When Exim is called to do synchronous delivery of a locally submitted
1090 message (the -odf or -odi options), it no longer closes stderr before doing
1093 3. Implemented the mua_wrapper option.
1095 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1097 5. Implemented the functions header_remove(), header_testname(),
1098 header_add_at_position(), and receive_remove_recipient(), and exported them
1101 6. If an ACL "warn" statement specified the addition of headers, Exim already
1102 inserted X-ACL-Warn: at the start if there was no header name. However, it
1103 was not making this test for the second and subsequent header lines if
1104 there were newlines in the string. This meant that an invalid header could
1105 be inserted if Exim was badly configured.
1107 7. Allow an ACL "warn" statement to add header lines at the start or after all
1108 the Received: headers, as well as at the end.
1110 8. Added the rcpt_4xx retry error code.
1112 9. Added postmaster_mailfrom=xxx to callout verification option.
1114 10. Added mailfrom=xxxx to the callout verification option, for verify=
1117 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1118 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1120 12. Inserted some casts to stop certain compilers warning when using pointer
1121 differences as field lengths or precisions in printf-type calls (mostly
1122 affecting debugging statements).
1124 13. Added optional readline() support for -be (dynamically loaded).
1126 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1127 same clock tick as a message's arrival, so that its received time was the
1128 same as the "first fail" time on the retry record, and that message
1129 remained on the queue past the ultimate address timeout, every queue runner
1130 would try a delivery (because it was past the ultimate address timeout) but
1131 after another failure, the ultimate address timeout, which should have then
1132 bounced the address, did not kick in. This was a "< instead of <=" error;
1133 in most cases the first failure would have been in the next clock tick
1134 after the received time, and all would be well.
1136 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1137 being recognized when the domain list was tested by the match_domain
1138 condition in an expansion string.
1140 16. Added the ${str2b64: operator.
1142 17. Exim was always calling setrlimit() to set a large limit for the number of
1143 processes, without checking whether the existing limit was already
1144 adequate. (It did check for the limit on file descriptors.) Furthermore,
1145 errors from getrlimit() and setrlimit() were being ignored. Now they are
1146 logged to the main and panic logs, but Exim does carry on, to try to do its
1147 job under whatever limits there are.
1149 18. Imported PCRE 5.0.
1151 19. Trivial typo in log message " temporarily refused connection" (the leading
1154 20. If the log selector return_path_on_delivery was set and an address was
1155 redirected to /dev/null, the delivery process crashed because it assumed
1156 that a return path would always be set for a "successful" delivery. In this
1157 case, the whole delivery is bypassed as an optimization, and therefore no
1160 21. Internal re-arrangement: the function for sending a challenge and reading
1161 a response while authentication was assuming a zero-terminated challenge
1162 string. It's now changed to take a pointer and a length, to allow for
1163 binary data in such strings.
1165 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1167 23. Exim was not respecting finduser_retries when seeking the login of the
1168 uid under which it was called; it was always trying 10 times. (The default
1169 setting of finduser_retries is zero.) Also, it was sleeping after the final
1170 failure, which is pointless.
1172 24. Implemented tls_on_connect_ports.
1174 25. Implemented acl_smtp_predata.
1176 26. If the domain in control=submission is set empty, Exim assumes that the
1177 authenticated id is a complete email address when it generates From: or
1178 Sender: header lines.
1180 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1181 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1182 chown and chgrp in /bin and hostname in /usr/bin.
1184 28. Exim was keeping the "process log" file open after each use, just as it
1185 does for the main log. This opens the possibility of it remaining open for
1186 long periods when the USR1 signal hits a daemon. Occasional processlog
1187 errors were reported, that could have been caused by this. Anyway, it seems
1188 much more sensible not to leave this file open at all, so that is what now
1191 29. The long-running daemon process does not normally write to the log once it
1192 has entered its main loop, and it closes the log before doing so. This is
1193 so that log files can straightforwardly be renamed and moved. However,
1194 there are a couple of unusual error situations where the daemon does write
1195 log entries, and I had neglected to close the log afterwards.
1197 30. The text of an SMTP error response that was received during a remote
1198 delivery was being truncated at 512 bytes. This is too short for some of
1199 the long messages that one sometimes sees. I've increased the limit to
1202 31. It is now possible to make retry rules that apply only when a message has a
1203 specific sender, in particular, an empty sender.
1205 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1206 it possible to be selective about when SMTP synchronization is enforced.
1208 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1210 32. Implemented hosts_connection_nolog.
1212 33. Added an ACL for QUIT.
1214 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1217 35. Added mailbox_size and mailbox_filecount to appendfile.
1219 36. Added control = no_multiline_responses to ACLs.
1221 37. There was a bug in the logic of the code that waits for the clock to tick
1222 in the case where the clock went backwards by a substantial amount such
1223 that the microsecond fraction of "now" was more than the microsecond
1224 fraction of "then" (but the whole seconds number was less).
1226 38. Added support for the libradius Radius client library this is found on
1227 FreeBSD (previously only the radiusclient library was supported).
1233 1. When certain lookups returned multiple values in the form name=value, the
1234 quoting of the values was not always being done properly. Specifically:
1235 (a) If the value started with a double quote, but contained no whitespace,
1237 (b) If the value contained whitespace other than a space character (i.e.
1238 tabs or newlines or carriage returns) it was not quoted.
1239 This fix has been applied to the mysql and pgsql lookups by writing a
1240 separate quoting function and calling it from the lookup code. The fix
1241 should probably also be applied to nisplus, ibase and oracle lookups, but
1242 since I cannot test any of those, I have not disturbed their existing code.
1244 2. A hit in the callout cache for a specific address caused a log line with no
1245 reason for rejecting RCPT. Now it says "Previous (cached) callout
1246 verification failure".
1248 3. There was an off-by-one bug in the queryprogram router. An over-long
1249 return line was truncated at 256 instead of 255 characters, thereby
1250 overflowing its buffer with the terminating zero. As well as fixing this, I
1251 have increased the buffer size to 1024 (and made a note to document this).
1253 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1254 when Exim is waiting for an SMTP response from a remote server, Exim
1255 restarts its select() call on the socket, thereby resetting its timeout.
1256 This is not a problem when such interrupts are rare. Somebody set up a cron
1257 job to run exiwhat every 2 minutes, which is less than the normal select()
1258 timeout (5 or 10 minutes). This meant that the select() timeout never
1259 kicked in because it was always reset. I have fixed this by comparing the
1260 time when an interrupt arrives with the time at the start of the first call
1261 to select(). If more time than the timeout has elapsed, the interrupt is
1262 treated as a timeout.
1264 5. Some internal re-factoring in preparation for the addition of Sieve
1265 extensions (by MH). In particular, the "personal" test is moved to a
1266 separate function, and given an option for scanning Cc: and Bcc: (which is
1267 not set for Exim filters).
1269 6. When Exim created an email address using the login of the caller as the
1270 local part (e.g. when creating a From: or Sender: header line), it was not
1271 quoting the local part when it contained special characters such as @.
1273 7. Installed new OpenBSD configuration files.
1275 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1276 try to make them clearer.
1278 9. Callout options, other than the timeout value, were being ignored when
1279 verifying sender addresses in header lines. For example, when using
1281 verify = header_sender/callout=no_cache
1283 the cache was (incorrectly) being used.
1285 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1286 only the Cygwin environment.
1288 11. When return_path_on_delivery was set as a log selector, if different remote
1289 addresses in the same message used different return paths and parallel
1290 remote delivery occurred, the wrong values would sometimes be logged.
1291 (Whenever a remote delivery process finished, the return path value from
1292 the most recently started remote delivery process was logged.)
1294 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1295 lines when AUTH and/or TLS are in use. This is the "received protocol"
1296 field. Exim used to use "asmtp" for authenticated SMTP, without any
1297 indication (in the protocol name) for TLS use. Now it follows the RFC and
1298 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1299 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1300 names appear in log lines as well as in Received: header lines.
1302 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1303 extensions, and comparison tests, and to fix some bugs.
1305 14. Changes to the "personal" filter test:
1307 (1) The test was buggy in that it was just doing the equivalent of
1308 "contains" tests on header lines. For example, if a user's address was
1309 anne@some.where, the "personal" test would incorrectly be true for
1311 To: susanne@some.where
1313 This test is now done by extracting each address from the header in turn,
1314 and checking the entire address. Other tests that are part of "personal"
1315 are now done using regular expressions (for example, to check local parts
1316 of addresses in From: header lines).
1318 (2) The list of non-personal local parts in From: addresses has been
1319 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1320 taken from the Sieve specification recommendations.
1322 (3) If the message contains any header line starting with "List-" it is
1323 treated as non-personal.
1325 (4) The test for "circular" in the Subject: header line has been removed
1326 because it now seems ill-conceived.
1328 15. Minor typos in src/EDITME comments corrected.
1330 16. Installed latest exipick from John Jetmore.
1332 17. If headers_add on a router specified a text string that was too long for
1333 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1334 of string_sprintf() is now avoided.
1336 18. $message_body_size was not set (it was always zero) when running the DATA
1337 ACL and the local_scan() function.
1339 19. For the "mail" command in an Exim filter, no default was being set for
1340 the once_repeat time, causing a random time value to be used if "once" was
1341 specified. (If the value happened to be <= 0, no repeat happened.) The
1342 default is now 0s, meaning "never repeat". The "vacation" command was OK
1343 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1344 (I found it when inspecting the code).
1346 20. There is now an overall timeout for performing a callout verification. It
1347 defaults to 4 times the callout timeout, which applies to individual SMTP
1348 commands during the callout. The overall timeout applies when there is more
1349 than one host that can be tried. The timeout is checked before trying the
1350 next host. This prevents very long delays if there are a large number of
1351 hosts and all are timing out (e.g. when the network connections are timing
1352 out). The value of the overall timeout can be changed by specifying an
1353 additional sub-option for "callout", called "maxwait". For example:
1355 verify = sender/callout=5s,maxwait=20s
1357 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1358 to the end before writing, but this should make it even safer).
1360 22. Exim was forgetting that it had advertised PIPELINING for the second and
1361 subsequent messages on an SMTP connection. It was also not resetting its
1362 memory on STARTTLS and an internal HELO.
1364 23. When Exim logs an SMTP synchronization error within a session, it now
1365 records whether PIPELINING has been advertised or not.
1367 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1368 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1369 rather than long int.
1371 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1373 26. Added the never_mail option to autoreply.
1379 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1380 crash if the getsockname() call failed; this can happen if a connection is
1381 closed very soon after it is established. The problem was simply in the
1382 order in which certain operations were done, causing Exim to try to write
1383 to the SMTP stream before it had set up the file descriptor. The bug has
1384 been fixed by making things happen in the correct order.
1390 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1391 before the connection was closed, thus losing the rejection response.
1393 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1394 some early Solaris releases, but causes trouble in current releases where
1395 socklen_t is defined.
1397 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1400 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1403 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1404 that releasing the top of it at the end releases what was used for sub-
1405 expansions (unless the block got too big). However, discard this block if
1406 the first thing is a variable or header, so that we can use its block when
1407 it is dynamic (useful for very large $message_headers, for example).
1409 6. Lookups now cache *every* query, not just the most recent. A new, separate
1410 store pool is used for this. It can be recovered when all lookup caches are
1411 flushed. Lookups now release memory at the end of their result strings.
1412 This has involved some general refactoring of the lookup sources.
1414 7. Some code has been added to the store_xxx() functions to reduce the amount
1415 of flapping under certain conditions.
1417 8. log_incoming_interface used to affect only the <= reception log lines. Now
1418 it causes the local interface and port to be added to several more SMTP log
1419 lines, for example "SMTP connection from", and rejection lines.
1421 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1423 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1425 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1426 could be overwritten at the end of the current message (or the start of a
1427 new message if it was set in a HELO ACL). The value is now preserved for
1428 the duration of the SMTP connection.
1430 12. If a transport had a headers_rewrite setting, and a matching header line
1431 contained an unqualified address, that address was qualified, even if it
1432 did not match any rewriting rules. The underlying bug was that the values
1433 of the flags that permit the existence of unqualified sender and recipient
1434 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1435 non-local messages, and by -bnq for local messages) were not being
1436 preserved with the message after it was received.
1438 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1439 "next input=" as part of the text comprising the host identity instead of
1440 the correct text. The code was using the same buffer for two different
1441 strings. However, depending on which order the printing function evaluated
1442 its arguments, the bug did not always show up. Under Linux, for example, my
1443 test suite worked just fine.
1445 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1446 doesn't work with some older versions of Perl. It has been changed to "my",
1447 which in any case is probably the better facility to use.
1449 15. A really picky compiler found some instances of statements for creating
1450 error messages that either had too many or two few arguments for the format
1453 16. The size of the buffer for calls to the DNS resolver has been increased
1454 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1455 for addresses that have a lot of PTR records. This alleviates a problem; it
1456 does not fully solve it.
1458 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1459 buffer now truncates the list and logs the incident, which is the same
1460 action as happens when Exim is looking up a host name and its aliases.
1461 Previously in this situation something unpredictable would happen;
1462 sometimes it was "internal error: store_reset failed".
1464 18. If a server dropped the connection unexpectedly when an Exim client was
1465 using GnuTLS and trying to read a response, the client delivery process
1466 crashed while trying to generate an error log message.
1468 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1469 string, for example:
1471 warn message = H1: something\nH2: something
1473 the text was added as a single header line from Exim's point of view
1474 though it ended up OK in the delivered message. However, searching for the
1475 second and subsequent header lines using $h_h2: did not work. This has been
1476 fixed. Similarly, if a system filter added multiple headers in this way,
1477 the routers could not see them.
1479 20. Expanded the error message when iplsearch is called with an invalid key to
1480 suggest using net-iplsearch in a host list.
1482 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1483 ACLs are no longer actually imposed (and a message to that effect is
1486 22. If a "gecos" field in a passwd entry contained escaped characters, in
1487 particular, if it contained a \" sequence, Exim got it wrong when building
1488 a From: or a Sender: header from that name. A second bug also caused
1489 incorrect handling when an unquoted " was present following a character
1490 that needed quoting.
1492 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1493 was not being matched caselessly.
1495 24. Arranged for all hyphens in the exim.8 source to be escaped with
1498 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1499 properly. Recipient callout cache records were still being keyed to include
1500 the sender, even when use_sender was set false. This led to far more
1501 callouts that were necessary. The sender is no longer included in the key
1502 when use_sender is false.
1504 26. Added "control = submission" modifier to ACLs.
1506 27. Added the ${base62d: operator to decode base 62 numbers.
1508 28. dnsdb lookups can now access SRV records.
1510 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1511 the configuration file.
1513 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1514 (-v) mode. This makes the output for a verbose queue run more intelligible.
1516 31. Added a use_postmaster feature to recipient callouts.
1518 32. Added the $body_zerocount variable, containing the number of binary zero
1519 bytes in the message body.
1521 33. The time of last modification of the "new" subdirectory is now used as the
1522 "mailbox time last read" when there is a quota error for a maildir
1525 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1527 35. Added +ignore_unknown as a special item in host lists.
1529 36. Code for decoding IPv6 addresses in host lists is now included, even if
1530 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1531 address was recognized as an IP address, but was then not correctly decoded
1532 into binary, causing unexpected and incorrect effects when compared with
1539 1. Very minor rewording of debugging text in manualroute to say "list of
1540 hosts" instead of "hostlist".
1542 2. If verify=header_syntax was set, and a header line with an unqualified
1543 address (no domain) and a large number of spaces between the end of the
1544 name and the colon was received, the reception process suffered a buffer
1545 overflow, and (when I tested it) crashed. This was caused by some obsolete
1546 code that should have been removed. The fix is to remove it!
1548 3. When running in the test harness, delay a bit after writing a bounce
1549 message to get a bit more predictability in the log output.
1551 4. Added a call to search_tidyup() just before forking a reception process. In
1552 theory, someone could use a lookup in the expansion of smtp_accept_max_
1553 per_host which, without the tidyup, could leave open a database connection.
1555 5. Added the variables $recipient_data and $sender_data which get set from a
1556 lookup success in an ACL "recipients" or "senders" condition, or a router
1557 "senders" option, similar to $domain_data and $local_part_data.
1559 6. Moved the writing of debug_print from before to after the "senders" test
1562 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1563 problems for message scanning, either using a data ACL, or using
1564 local_scan() because the Received: header was not generated till after they
1565 were called (in order to set the time as the time of reception completion).
1566 I have revised the way this works. The header is now generated after the
1567 body is received, but before the ACL or local_scan() are called. After they
1568 are run, the timestamp in the header is updated.
1574 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1575 before starting a queue runner without re-exec. This happened only when
1576 deliver_drop_privilege was set or when the Exim user was set to root. The
1577 effect of the bug was that timeouts during subsequent deliveries caused
1578 crashes instead of being properly handled. The handler is now left at its
1579 default (and expected) setting.
1581 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1582 message, again when deliver_drop_privilege is set or Exim is run as root.
1583 The bug described in (1) was not present in this case, but the tidying up
1584 of the other signals was missing. I have made the two cases consistent.
1586 3. The ignore_target_hosts setting on a manualroute router was being ignored
1587 for hosts that were looked up using the /MX notation.
1589 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1592 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1593 operated on the sender address. After changing the $sender_address to <>
1594 for the sender address verify, Exim was re-instated it as the original
1595 (before rewriting) address, but remembering that it had rewritten it, so it
1596 wasn't rewriting it again. This bug also had the effect of breaking the
1597 sender address verification caching when the sender address was rewritten.
1599 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1600 This has been changed so that if the ip literal address matches
1601 ignore_target_hosts, the router declines.
1603 7. Added expansion conditions match_domain, match_address, and match_local_
1604 part (NOT match_host).
1606 8. The placeholder for the Received: header didn't have a length field set.
1608 9. Added code to Exim itself and to exim_lock to test for a specific race
1609 condition that could lead to file corruption when using MBX delivery. The
1610 issue is with the lockfile that is created in /tmp. If this file is removed
1611 after a process has opened it but before that process has acquired a lock,
1612 there is the potential for a second process to recreate the file and also
1613 acquire a lock. This could lead to two Exim processes writing to the file
1614 at the same time. The added code performs the same test as UW imapd; it
1615 checks after acquiring the lock that its file descriptor still refers to
1616 the same named file.
1618 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1619 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1622 11. Added the smtp_active_hostname option. If used, this will typically be made
1623 to depend on the incoming interface address. Because $interface_address is
1624 not set up until the daemon has forked a reception process, error responses
1625 that can happen earlier (such as "too many connections") no longer contain
1628 12. If an expansion in a condition on a "warn" statement fails because a lookup
1629 defers, the "warn" statement is abandoned, and the next ACL statement is
1630 processed. Previously this caused the whole ACL to be aborted.
1632 13. Added the iplsearch lookup type.
1634 14. Added ident_timeout as a log selector.
1636 15. Added tls_certificate_verified as a log selector.
1638 16. Added a global option tls_require_ciphers (compare the smtp transport
1639 option of the same name). This controls incoming TLS connections.
1641 17. I finally figured out how to make tls_require_ciphers do a similar thing
1642 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1643 before starting the TLS session.
1645 18. Tabs are now shown as \t in -bP output.
1647 19. If the log selector return_path_on_delivery was set, Exim crashed when
1648 bouncing a message because it had too many Received: header lines.
1650 20. If two routers both had headers_remove settings, and the first one included
1651 a superfluous trailing colon, the final name in the first list and the
1652 first name in the second list were incorrectly joined into one item (with a
1653 colon in the middle).
1659 1. Added -C and -D options to the exinext utility, mainly to make it easier
1660 to include in the automated testing, but these could be helpful when
1661 multiple configurations are in use.
1663 2. The exinext utility was not formatting the output nicely when there was
1664 an alternate port involved in the retry record key, nor when there was a
1665 message id as well (for retries that were specific to a specific message
1666 and a specific host). It was also confused by IPv6 addresses, because of
1667 the additional colons they contain. I have fixed the IPv4 problem, and
1668 patched it up to do a reasonable job for IPv6.
1670 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1671 delivery, the log line now contains "pipelined" if PIPELINING was used.
1673 4. An SMTP transport process used to panic and die if the bind() call to set
1674 an explicit outgoing interface failed. This has been changed; it is now
1675 treated in the same way as a connect() failure.
1677 5. A reference to $sender_host_name in the part of a conditional expansion
1678 that was being skipped was still causing a DNS lookup. This no longer
1681 6. The def: expansion condition was not recognizing references to header lines
1682 that used bh_ and bheader_.
1684 7. Added the _cache feature to named lists.
1686 8. The code for checking quota_filecount in the appendfile transport was
1687 allowing one more file than it should have been.
1689 9. For compatibility with Sendmail, the command line option
1697 and sets the incoming protocol and host name (for trusted callers). The
1698 host name and its colon can be omitted when only the protocol is to be set.
1699 Note the Exim already has two private options, -pd and -ps, that refer to
1700 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1701 "s", but I don't think that's a major issue.
1703 10. A number of refactoring changes to the code, none of which should affect
1706 (a) The number of logging options was getting close to filling up the
1707 32-bit word that was used as a bit map. I have split them into two classes:
1708 those that are passed in the argument to log_write(), and those that are
1709 only ever tested independently outside of that function. These are now in
1710 separate 32-bit words, so there is plenty of room for expansion again.
1711 There is no change in the user interface or the logging behaviour.
1713 (b) When building, for example, log lines, the code previously used a
1714 macro that called string_cat() twice, in order to add two strings. This is
1715 not really sufficiently general. Furthermore, there was one instance where
1716 it was actually wrong because one of the argument was used twice, and in
1717 one call a function was used. (As it happened, calling the function twice
1718 did not affect the overall behaviour.) The macro has been replaced by a
1719 function that can join an arbitrary number of extra strings onto a growing
1722 (c) The code for expansion conditions now uses a table and a binary chop
1723 instead of a serial search (which was left over from when there were very
1724 few conditions). Also, it now recognizes conditions like "pam" even when
1725 the relevant support is not compiled in: a suitably worded error message is
1726 given if an attempt is made to use such a condition.
1728 11. Added ${time_interval:xxxxx}.
1730 12. A bug was causing one of the ddress fields not to be passed back correctly
1731 from remote delivery subprocesses. The field in question was not being
1732 subsequently used, so this caused to problems in practice.
1734 13. Added new log selectors queue_time and deliver_time.
1736 14. Might have fixed a bug in maildirsizefile handling that threw up
1737 "unexpected character" debug warnings, and recalculated the data
1738 unnecessarily. In any case, I expanded the warning message to give more
1741 15. Added the message "Restricted characters in address" to the statements in
1742 the default ACL that block characters like @ and % in local parts.
1744 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
1745 Three changes have been made:
1747 (a) There was a serious bug; a negative response to MAIL caused the whole
1748 recipient domain to be cached as invalid, thereby blocking all messages
1749 to all local parts at the same domain, from all senders. This bug has
1750 been fixed. The domain is no longer cached after a negative response to
1751 MAIL if the sender used is not empty.
1753 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
1756 (c) A new callout option, "use_sender" has been added for people who want
1757 the modified behaviour.
1763 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
1766 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
1767 indeed breaks things for older releases.
1769 3. Added additional logging to the case where there is a problem reading data
1770 from a filter that is running in a subprocess using a pipe, in order to
1771 try to track down a specific problem.
1773 4. Testing facility fudge: when running in the test harness and attempting
1774 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
1775 getting "No route to host". Convert this to a timeout.
1777 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
1780 6. Some OS don't have socklen_t but use size_t instead. This affects the
1781 fifth argument of getsockopt() amongst other things. This is now
1782 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
1783 can be set for individual OS. I have set it for SunOS5, OSF1, and
1784 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
1785 some earlier ones do not.
1787 7. Change 4.30/15 was not doing the test caselessly.
1789 8. The standard form for an IPv6 address literal was being rejected by address
1790 parsing in, for example, MAIL and RCPT commands. An example of this kind of
1791 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1792 this, as well as the form without the "IPv6" on the front (but only when
1793 address literals are enabled, of course).
1795 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
1797 10. Exim crashed if a message with an empty sender address specified by -f
1798 encountered a router with an errors_to setting. This could be provoked only
1799 by a command such as
1803 where an empty string was supplied; "<>" did not hit this bug.
1805 11. Installed PCRE release 4.5.
1807 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
1808 remained set. It is now erased.
1810 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
1811 times from message ids (which are base 36 rather than the normal 62).
1813 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
1814 were being counted as actual protocol errors, and logged if the log
1815 selector +smtp_protocol_error was set. One cannot be perfect in this test,
1816 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
1817 and DATA following a set of rejected RCPTs do not count as protocol errors.
1818 In other words, Exim assumes they were pipelined, though this may not
1819 actually be the case. Of course, in all cases the client gets an
1820 appropriate error code.
1822 15. If a lookup fails in an ACL condition, a message about the failure may
1823 be available; it is used if testing the ACL cannot continue, because most
1824 such messages specify what the cause of the deferral is. However, some
1825 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
1826 that caused an old message to be retained and used if a later statement
1827 caused a defer, replacing the real cause of the deferral.
1829 16. If an IP address had so many PTR records that the DNS lookup buffer
1830 was not large enough to hold them, Exim could crash while trying to process
1831 the truncated data. It now detects and logs this case.
1833 17. Further to 4.21/58, another change has been made: if (and only if) the
1834 first line of a message (the first header line) ends with CRLF, a bare LF
1835 in a subsequent header line has a space inserted after it, so as not to
1836 terminate the header.
1838 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
1839 unnecessarily, used atoi() instead of strtol(), and didn't check the
1840 termination when getting file sizes from file names by regex.
1842 19. Completely re-implemented the support for maildirsize files, in the light
1843 of a number of problems with the previous contributed implementation
1844 (4.30/29). In particular:
1846 . If the quota is zero, the maildirsize file is maintained, but no quota is
1849 . If the maildir directory does not exist, it is created before any attempt
1850 to write a maildirsize file.
1852 . The quota value in the file is just a cache; if the quota is changed in
1853 the transport, the new value overrides.
1855 . A regular expression is available for excluding directories from the
1858 20. The autoreply transport checks the characters in options that define the
1859 message's headers; it allows continued headers, but it was checking with
1860 isspace() after an embedded newline instead of explicitly looking for a
1863 21. If all the "regular" hosts to which an address was routed had passed their
1864 expiry times, and had not reached their retry times, the address was
1865 bounced, even if fallback hosts were defined. Now Exim should go on to try
1868 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
1869 equivalent code in the SMTP transport. Some hosts send humungous responses
1870 to HELO/EHLO, more than 1024 it seems.
1872 23. Refactoring: code in filter.c used (void *) for "any old type" but this
1873 gives compiler warnings in some environments. I've now done it "properly",
1876 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
1877 (because of problems with the built-in one) was declared to return uschar *
1878 instead of char *, causing compiler failure.
1880 25. Fixed a file descriptor leak when processing alias/forward files.
1882 26. Fixed a minor format string issue in dbfn.c.
1884 27. Typo in exim.c: ("dmbnz" for "dbmnz").
1886 28. If a filter file refered to $h_xxx or $message_headers, and the headers
1887 contained RFC 2047 "words", Exim's memory could, under certain conditions,
1890 29. When a sender address is verified, it is cached, to save repeating the test
1891 when there is more than one recipient in a message. However, when the
1892 verification involves a callout, it is possible for different callout
1893 options to be set for different recipients. It is too complicated to keep
1894 track of this in the cache, so now Exim always runs a verification when a
1895 callout is required, relying on the callout cache for the optimization.
1896 The overhead is duplication of the address routing, but this should not be
1899 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
1900 to be verified with callout=postmaster, and the main callout worked but the
1901 postmaster check failed, the verification correctly failed. However, if a
1902 subsequent RCPT command asked for sender verification *without* the
1903 postmaster check, incorrect caching caused this verification also to fail,
1906 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
1907 it was not caching the DNS options (qualify_single, search_parents) that
1908 were used when the lookup failed. A subsequent lookup with different
1909 options therefore always gave the same answer, though there were cases
1910 where it should not have. (Example: a "domains = !$mx_any" option on a
1911 dnslookup router: the "domains" option is always processed without any
1912 widening, but the router might have qualify_single set.) Now Exim uses the
1913 cached value only when the same options are set.
1915 32. Added John Jetmore's "exipick" utility to the distribution.
1917 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
1918 than a timeout (e.g. a certificate is required, and is not provided), an
1919 Exim server now closes the connection immediately. Previously it waited for
1920 the client to close - but if the client is SSL, it seems that they each
1921 wait for each other, leading to a delay before one of them times out.
1923 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
1924 maintained 0.8.x compatibility because I don't think many are using it, and
1925 it is clearly obsolete.
1927 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
1930 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
1931 client certificate was expired. A simple patch fixes this, though I don't
1932 understand the full logic of why the verify callback is called multiple
1935 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
1936 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
1937 which causes problems with some clients (such as the Certicom SSL Plus
1938 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
1939 disables the coutermeasure allowing Eudora to connect."
1941 38. Exim was not checking that a write() to a log file succeeded. This could
1942 lead to Bad Things if a log got too big, in particular if it hit a file
1943 size limit. Exim now panics and dies if it cannot write to a log file, just
1944 as it does if it cannot open a log file.
1946 39. Modified OS/Makefile-Linux so that it now contains
1948 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
1950 The two -D definitions ensure that Exim is compiled with large file
1951 support, which makes it possible to handle log files that are bigger than
1954 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
1955 instance) a domain was checked against a named list that involved a lookup,
1956 causing $domain_data to be set, then another domain was checked against the
1957 same list, then the first domain was re-checked, the value of $domain_data
1958 after the final check could be wrong. In particular, if the second check
1959 failed, it could be set empty. This bug probably also applied to
1962 41. The strip_trailing_dot option was not being applied to the address given
1963 with the -f command-line option.
1965 42. The code for reading a message's header from the spool was incrementing
1966 $received_count, but never initializing it. This meant that the value was
1967 incorrect (doubled) while delivering a message in the same process in which
1968 it was received. In the most common configuration of Exim, this never
1969 happens - a fresh exec is done - but it can happen when
1970 deliver_drop_privilege is set.
1972 43. When Exim logs an SMTP synchronization error - client data sent too soon -
1973 it now includes up to 150 characters of the unexpected data in the log
1976 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
1977 and building data strings. The size of both of these buffers was 10 000
1978 bytes - far larger than anybody would *ever* want, thought I. Needless to
1979 say, somebody hit the limit. I have increased the maximum line length to
1980 20 000 and the maximum data length of concatenated lines to 100 000. I have
1981 also fixed two bugs, because there was no checking on these buffers. Tsk,
1982 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1983 buffer is too small.
1985 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1986 lsearch lookups. Now it does.
1988 46. When parsing a route_list item in a manualroute router, a fixed-length
1989 buffer was used for the list of hosts. I made this 1024 bytes long,
1990 thinking that nobody would ever have a list of hosts that long. Wrong.
1991 Somebody had a whole pile of complicated expansion conditions, and the
1992 string was silently truncated, leading to an expansion error. It turns out
1993 that it is easier to change to an unlimited length (owing to other changes
1994 that have happened since this code was originally written) than to build
1995 structure for giving a limitation error. The length of the item that
1996 expands into the list of hosts is now unlimited.
1998 47. The lsearch lookup could not handle data where the length of text line was
1999 more than 4095 characters. Such lines were truncated, leading to shortened
2000 data being returned. It should now handle lines of any length.
2002 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2003 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2006 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2007 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2008 no longer happens in comment lines. A list of replacements is now placed
2009 at the head of all of the source files, except those whose only change is
2010 to replace PERL_COMMAND in the very first #! line.
2012 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2013 messages on the queue, with a bottom-up merge sort, using code contributed
2014 by Michael Haardt. This should make operations like -bp somewhat faster on
2015 large queues. It won't affect queue runners, except when queue_run_in_order
2018 51. Installed eximstats 1.31 in the distribution.
2020 52. Added support for SRV lookups to the dnslookup router.
2022 53. If an ACL referred to $message_body or $message_body_end, the value was not
2023 reset for any messages that followed in the same SMTP session.
2025 54. The store-handling optimization for building very long strings was not
2026 differentiating between the different store pools. I don't think this
2027 actually made any difference in practice, but I've tidied it.
2029 55. While running the routers to verify a sender address, $sender_address
2030 was still set to the sender address. This is wrong, because when routing to
2031 send a bounce to the sender, it would be empty. Therefore, I have changed
2032 it so that, while verifying a sender address, $sender_address is set to <>.
2033 (There is no change to what happens when verifying a recipient address.)
2035 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2036 target A or AAAA records (if not already returned) without resetting the
2037 qualify_single or search_parents options of the DNS resolver. These are
2038 inappropriate in this case because the targets of MX and SRV records must
2039 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2040 target that, when qualified, matched something in the local domain. These
2041 two options are now turned off when doing these lookups.
2043 57. It seems that at least some releases of Reiserfs (which does not have the
2044 concept of a fixed number of inodes) returns zero and not -1 for the
2045 number of available inodes. This interacted badly with check_spool_inodes,
2046 which assumed that -1 was the "no such thing" setting. What I have done is
2047 to check that the total number of inodes is greater than zero before doing
2048 the test of how many are available.
2050 58. When a "warn" ACL statement has a log_message modifier, the message is
2051 remembered, and not repeated. This is to avoid a lot of repetition when a
2052 message has many recipients that cause the same warning to be written.
2053 Howewer, Exim was preserving the list of already written lines for an
2054 entire SMTP session, which doesn't seem right. The memory is now reset if a
2055 new message is started.
2057 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2058 debugging output unless log_rewrite was also set.
2060 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2061 of (int)(handle) when we know that handle contains (void *)(-1).
2063 61. The Exim daemon panic-logs an error return when it closes the incoming
2064 connection. However "connection reset by peer" seems to be common, and
2065 isn't really an error worthy of noting specially, so that particular error
2068 62. When Exim is trying to find all the local interfaces, it used to panic and
2069 die if the ioctl to get the interface flags failed. However, it seems that
2070 on at least one OS (Solaris 9) it is possible to have an interface that is
2071 included in the list of interfaces, but for which you get a failure error
2072 for this call. This happens when the interface is not "plumbed" into a
2073 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2074 failure of the "get flags" call assumes that the interface is down.
2076 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2077 makes life easier for people who are doing arithmetic on fields extracted
2078 from dates, where you often get leading zeros that should not be
2079 interpreted as octal.
2081 64. Added qualify_domain to the redirect router, to override the global
2084 65. If a pathologically long header line contained very many addresses (the
2085 report of this problem mentioned 10 000) and each of them was rewritten,
2086 Exim could use up a very large amount of memory. (It kept on making new
2087 copies of the header line as it rewrote, and never released the old ones.)
2088 At the expense of a bit more processing, the header rewriting function has
2089 been changed so that it no longer eats memory in this way.
2091 66. The generation of the Received: header has been moved from the time that a
2092 message starts to be received, to the time that it finishes. The timestamp
2093 in the Received: header should now be very close to that of the <= log
2094 line. There are two side-effects of this change:
2096 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2097 logged header lines no longer include the local Received: line, because
2098 it has not yet been created. The same applies to a copy of the message
2099 that is returned to a non-SMTP sender when a message is rejected.
2101 (b) When a filter file is tested using -bf, no additional Received: header
2102 is added to the test message. After some thought, I decided that this
2105 This change does not affect the value of $received_for. It is still set
2106 after address rewriting, but before local_scan() is called.
2108 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2110 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2111 gave an unhelpful panic error message, and a defer error. I have managed to
2112 change this behaviour so that it now rejects any supplied certificate,
2113 which seems right, as the list of acceptable certificates is empty.
2115 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2116 gave an unhelpful defer error. I have not managed to make this reject any
2117 supplied certificates, but the error message it gives is "no certificate
2118 supplied", which is not helpful.
2120 70. exigrep's output now also includes lines that are not associated with any
2121 message, but which match the given pattern. Implemented by a patch from
2122 Martin Sluka, which also tidied up the Perl a bit.
2124 71. Recipient callout verification, like sender verification, was using <> in
2125 the MAIL FROM command. This isn't really the right thing, since the actual
2126 sender may affect whether the remote host accepts the recipient or not. I
2127 have changed it to use the actual sender in the callout; this means that
2128 the cache record is now keyed on a recipient/sender pair, not just the
2129 recipient address. There doesn't seem to be a real danger of callout loops,
2130 since a callout by the remote host to check the sender would use <>.
2131 [SEE ABOVE: changed after hitting problems.]
2133 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2134 temporary errors. However, in the case of such a code being given after
2135 the end of a data transmission (i.e. after ".") Exim was failing to write
2136 a retry record for the message. (Yes, there was some broken host that was
2137 actually sending 8xx at this point.)
2139 73. An unknown lookup type in a host list could cause Exim to panic-die when
2140 the list was checked. (An example that provoked this was putting <; in the
2141 middle of a list instead of at the start.) If this happened during a DATA
2142 ACL check, a -D file could be left lying around. This kind of configuration
2143 error no longer causes Exim to die; instead it causes a defer errror. The
2144 incident is still logged to the main and panic logs.
2146 74. Buglet left over from Exim 3 conversion. The message "too many messages
2147 in one connection" was written to the rejectlog but not the mainlog, except
2148 when address rewriting (yes!) was being logged.
2150 75. Added write_rejectlog option.
2152 76. When a system filter was run not as root (that is, when system_filter_user
2153 was set), the values of the $n variables were not being returned to the
2154 main process; thus, they were not subsequently available in the $sn
2157 77. Added +return_path_on_delivery log selector.
2159 78. A connection timeout was being treated differently from recipients deferred
2160 when testing hosts_max_try with a message that was older than the host's
2161 retry timeout. (The host should not be counted, thus allowing all hosts to
2162 be tried at least once before bouncing.) This may have been the cause of an
2163 occasionally reported bug whereby a message would remain on the queue
2164 longer than the retry timeout, but would be bounced if a delivery was
2165 forced. I say "may" because I never totally pinned down the problem;
2166 setting up timeout/retry tests is difficult. See also the next item.
2168 79. The ultimate address timeout was not being applied to errors that involved
2169 a combination of host plus message (for example, a timeout on a MAIL
2170 command). When an address resolved to a number of possible hosts, and they
2171 were not all tried for each delivery (e.g. because of hosts_max_try), a
2172 message could remain on the queue longer than the retry timeout.
2174 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2177 81. Fixed an obscure SMTP outgoing bug which required at least the following
2178 conditions: (a) there was another message waiting for the same server;
2179 (b) the server returned 5xx to all RCPT commands in the first message so
2180 that the message was not completed; (c) the server dropped the connection
2181 or gave a negative response to the RSET that Exim sends to abort the
2182 transaction. The observed case was a dropped connection after DATA that had
2183 been sent in pipelining mode. That is, the server had advertised PIPELINING
2184 but was not implementing it correctly. The effect of the bug was incorrect
2185 behaviour, such as trying another host, and this could lead to a crash.
2191 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2192 and daemon.c were passed as pointers to ints; they should have been
2193 pointers to socklen_t variables (which are typically unsigned ints).
2195 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2198 3. Fixed a really odd bug that affected only the testing scheme; patching a
2199 certain fixed string in the binary changed the value of another string that
2200 happened to be identical to the end of the original first string.
2202 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2203 name", it returns that address as the IP address. On some operating
2204 systems (e.g. Solaris), it also passes back the IP address string as the
2205 "host name". However, on others (e.g. Linux), it passes back an empty
2206 string. Exim wasn't checking for this, and was changing the host name to an
2207 empty string, assuming it had been canonicized.
2209 5. Although rare, it is permitted to have more than one PTR record for a given
2210 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2211 all the names associated with an address, because they do in Solaris.
2212 However, it seems that they do not in Linux for data that comes from the
2213 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2214 I found this out when I moved to a new Linux workstation and tried to run
2215 the Exim test suite.
2217 To get round this problem I have changed the code so that it now does its
2218 own call to the DNS to look up PTR records when searching for a host name.
2219 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2220 addresses that are only in /etc/hosts are still found.
2222 This behaviour is, however, controlled by an option called host_lookup_
2223 order, which defaults to "bydns:byaddr". If people want to use the other
2224 order, or indeed, just use one or the other means of lookup, they can
2225 specify it in this variable.
2227 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2228 some operating systems, this comes back from gethostbyaddr() as an empty
2229 string, and this is what Exim used to test for. However, it seems that in
2230 other systems, "." is yielded. Exim now tests for this case too.
2232 7. The values of check_spool_space and check_log_space are now held internally
2233 as a number of kilobytes instead of an absolute number of bytes. If a
2234 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2235 kilobyte. This means that much larger values can be stored.
2237 8. Exim monitor: an attempt to get the action menu when not actually pointing
2238 at a message produces an empty menu entitled "No message selected". This
2239 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2240 no entries in it ("Shell widget menu has zero width and/or height"). So I
2241 have added a single, blank menu entry in this case.
2243 9. Added ${quote_local_part.
2245 10. MIME decoding is now applied to the contents of Subject: header lines when
2248 11. Now that a reference to $sender_host_address automatically causes a reverse
2249 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2250 host lookup before query-style lookups in lists that might use this
2251 variable. This has therefore been abolished, and the "net-" prefix is no
2252 longer necessary for query-style lookups.
2254 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2255 have been a typo for LFLAGS, so it has been changed.
2257 13. The install script calls Exim with "-C /dev/null" in order to find the
2258 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2259 to be output. Howeve, since Exim outputs its version number before the
2260 error, it didn't break the script. It just looked ugly. I fixed this by
2261 always allowing "-C /dev/null" if the caller is root.
2263 14. Ignore overlarge ACL variable number when reading spool file - insurance
2264 against a later release with more variables having written the file.
2266 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2267 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2268 this, as well as the form without the "IPv6" on the front.
2270 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2271 OS/Makefile-Darwin file.
2273 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2274 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2276 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2277 variable. However, some OS use ints for inodes. I've added cast to long int
2278 to get rid of the compiler warning.
2280 19. I had forgotten to lock out "/../" in configuration file names when
2281 ALT_CONFIG_PREFIX was set.
2283 20. Routers used for verification do not need to specify transports. However,
2284 if such a router generated a host list, and callout was configured, Exim
2285 crashed, because it could not find a port number from the (non-existent)
2286 transport. It now assumes port 25 in this circumstance.
2288 21. Added the -t option to exigrep.
2290 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2291 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2292 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2293 them is set, LOOKUP_LSEARCH is forced.
2295 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2297 24. Added sender and host information to the "rejected by local_scan()" log
2298 line; previously there was no indication of these.
2300 25. Added .include_if_exists.
2302 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2303 Linux. It turns out that not all file systems support this. Apparently some
2304 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2305 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2306 not supported on the file descriptor, is now ignored when Exim is trying to
2307 sync a directory. This applies only to Linux.
2309 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2311 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2312 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2315 29. Added support for maildirsize files from supplied patch (modified a bit).
2317 30. The use of :fail: followed by an empty string could lead Exim to respond to
2318 sender verification failures with (e.g.):
2320 550 Verification failed for <xxx>
2321 550 Sender verify failed
2323 where the first response line was missing the '-' that indicates it is not
2324 the final line of the response.
2326 31. The loop for finding the name of the user that called Exim had a hardwired
2327 limit of 10; it now uses the value of finduser_retries, which is used for
2328 all other user lookups.
2330 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2333 33. Exim was neglecting to zero errno before one call of strtol() when
2334 expanding a string and expecting an integer value. On some systems this
2335 resulted in spurious "integer overflow" errors. Also, it was casting the
2336 result into an int without checking.
2338 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2339 did not work. The code looks as if it has *never* worked, though it appears
2340 to have been documented since at least releast 1.62. I have made it work.
2342 35. The "timeout_DNS" error in retry rules, also documented since at least
2343 1.62, also never worked. As it isn't clear exactly what this means, and
2344 clearly it isn't a major issue, I have abolished the feature by treating it
2345 as "timeout", and writing a warning to the main and panic logs.
2347 36. The display of retry rules for -brt wasn't always showing the error code
2350 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2351 timeout_connect_A, timeout_connect_MX.
2353 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2354 to the empty sender.
2356 39. The daemon was not analysing the content of -oX till after it had closed
2357 stderr and disconnected from the controlling terminal. This meant that any
2358 syntax errors were only noted on the panic log, and the return code from
2359 the command was 0. By re-arranging the code a little, I've made the
2360 decoding happen first, so such errors now appear on stderr, and the return
2361 code is 1. However, the actual setting up of the sockets still happens in
2362 the disconnected process, so errors there are still only recorded on the
2365 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2366 connections (as happens on some IP stacks) was logged at start up time as
2367 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2368 it from "IPv6 and IPv4", which means that two separate sockets are being
2371 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2372 says whether AF_INET or AF_INET6 was passed as an argument.
2374 42. Exiwhat output was messed up when time zones were included in log
2377 43. Exiwhat now gives more information about the daemon's listening ports,
2378 and whether -tls-on-connect was used.
2380 44. The "port" option of the smtp transport is now expanded.
2382 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2383 silently ignored. Now an error message is written to the main and panic
2386 46. There's a new ACL modifier called "logwrite" which writes to a log file
2387 as soon as it is encountered.
2389 47. Added $local_user_uid and $local_user_gid at routing time.
2391 48. Exim crashed when trying to verify a sender address that was being
2394 49. Exim was recognizing only a space character after ".include". It now also
2395 recognizes a tab character.
2397 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2398 extracting the relevant information from the specification. The man page no
2399 longer contains scrambled data for the -d option, and I've added a section
2400 at the front about calling Exim under different names.
2402 51. Added "extra_headers" argument to the "mail" command in filter files.
2404 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2407 53. Installed eximstats 1.29.
2409 54. Added transport_filter_timeout as a generic transport option.
2411 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2412 Cc: header lines. This was required by RFC 822, but it not required by RFC
2415 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2416 incoming messages that did not have them. Now it does so only if the
2417 message originates locally, that is, if there is no associated remote host
2418 address. When Resent- header lines are present, this applies to the Resent-
2419 lines rather than the non-Resent- lines.
2421 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2422 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2424 58. Messages for configuration errors now include the name of the main
2425 configuration file - useful now that there may be more than one file in a
2426 list (.included file names were always shown).
2428 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2429 for those rare installations that do not start the daemon as root or run it
2430 setuid root. I've cut out the call to initgroups() if the daemon is not
2433 60. The Exim user and group can now be bound into the binary as text strings
2434 that are looked up at the start of Exim's processing.
2436 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2438 62. Added $mailstore_basename variable.
2440 63. Installed patch to sieve.c from Michael Haardt.
2442 64. When Exim failed to open the panic log after failing to open the main log,
2443 the original message it was trying to log was written to stderr and debug
2444 output, but if they were not available (the usual case in production), it
2445 was lost. Now it is written to syslog before the two lines that record the
2446 failures to open the logs.
2448 65. Users' Exim filters run in subprocesses under the user's uid. It is
2449 possible for a "deliver" command or an alias in a "personal" command to
2450 provoke an address rewrite. If logging of address rewriting is configured,
2451 this fails because the process is not running as root or exim. There may be
2452 a better way of dealing with this, but for the moment (because 4.30 needs
2453 to be released), I have disabled address rewrite logging when running a
2454 filter in a non-root, non-exim process.
2460 1. The buildconfig auxiliary program wasn't quoting the value set for
2461 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2462 not defined. This bug was masked in 4.22 by the effect that was fixed in
2465 2. Some messages that were rejected after a message id was allocated were
2466 shown as "incomplete" by exigrep. It no longer does this for messages that
2467 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2469 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2470 have allow_domain_literals set, the ID did not get logged in the <= line.
2471 Domain literals are now always recognized in Message-ID: header lines.
2473 4. The first argument for a ${extract expansion item is the key name or field
2474 number. Leading and trailing spaces in this item were not being ignored,
2475 causing some misleading effects.
2477 5. When deliver_drop_privilege was set, single queue runner processes started
2478 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2479 same command in the process it spins off) were not dropping privilege.
2481 6. When the daemon running as "exim" started a queue runner, it always
2482 re-executed Exim in the spun-off process. This is a waste of effort when
2483 deliver_drop_privilege is set. The new process now just calls the
2484 queue-runner function directly.
2490 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2493 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2494 ignored. Though the use of -oP was forcing the writing of a pid file, it
2495 was always written to the default place.
2497 3. If the message "no IP address found for host xxxx" is generated during
2498 incoming verification, it is now followed by identification of the incoming
2499 connection (so you can more easily find what provoked it).
2501 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2503 5. Added some features to "harden" Exim a bit more against certain attacks:
2505 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2506 be put in Local/Makefile. This is like the never_users runtime option,
2507 but it cannot be overridden. The default setting is "root".
2509 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2510 prefix string with which any file named in a -C command line option
2513 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2514 is retained for -C and -D only if the caller of Exim is root. Without
2515 it, the exim user may also use -C and -D and retain privilege.
2517 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2518 command line option is disabled.
2520 6. Macro names set by the -D option must start with an upper case letter, just
2521 like macro names defined in the configuration file.
2523 7. Added "dereference=" facility to LDAP.
2525 8. Two instances of the typo "uknown" in the source files are fixed.
2527 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2528 the Configure-Makefile script screwed up while processing it.
2530 10. Incorporated PCRE 4.4.
2532 11. The SMTP synchronization check was not operating right at the start of an
2533 SMTP session. For example, it could not catch a HELO sent before the client
2534 waited for the greeting. There is now a check for outstanding input at the
2535 point when the greeting is written. Because of the duplex, asynchronous
2536 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2537 way, but not yet received, when the check is performed.
2539 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2540 on TCP/IP sockets, because this apparently causes some broken clients to
2543 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2544 unchanged) from the Cygwin maintainer.
2546 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2547 was supported instead of testing for mbx. Effectively a typo.
2549 15. The spa authenticator server code was not checking that the input it
2550 received was valid base64.
2552 16. The debug output line for the "set" modifier in ACLs was not showing the
2553 name of the variable that was being set.
2555 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2557 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2558 lookup on its own. Something else was needed to trigger the lookup. For
2559 example, a match in host_lookup or the need for a host name in a host list.
2560 Now, if $sender_host_name is referenced and the host name has not yet been
2561 looked up, a lookup is performed. If the lookup fails, the variable remains
2562 empty, and $host_lookup_failed is set to "1".
2564 19. Added "eqi" as a case-independent comparison operator.
2566 20. The saslauthd authentication condition could segfault if neither service
2567 nor realm was specified.
2569 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2570 error message that was logged was misleading, and incoming SMTP
2571 connections were dropped. The message is now more accurate, and temporary
2572 errors are given to SMTP connections.
2574 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2575 (except RSET) with a 421 error, until QUIT is received. However, it was
2576 failing to send a response to QUIT.
2578 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2579 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2580 ACL is now run after the helo_try_verify_hosts code.
2582 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2583 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2584 case-independent, so other case variants are also recognized). Apparently
2585 some systems use these upper case variants.
2587 25. If more than two messages were waiting for the same host, and a transport
2588 filter was specified for the transport, Exim sent two messages over the
2589 same TCP/IP connection, and then failed with "socket operation on non-
2590 socket" when it tried to send the third.
2592 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2594 27. The extern definition of crypt16() in expand.c was not being excluded when
2595 the OS had its own crypt16() function.
2597 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2598 as a preferred synonym for return_size_limit, both as an option and as an
2601 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2603 30. Changed the default configuration ACL to relax the local part checking rule
2604 for addresses that are not in any local domains. For these addresses,
2605 slashes and pipe symbols are allowed within local parts, but the sequence
2606 /../ is explicitly forbidden.
2608 31. SPA server authentication was not clearing the challenge buffer before
2611 32. log_message in a "warn" ACL statement was writing to the reject log as
2612 well as to the main log, which contradicts the documentation and doesn't
2613 seem right (because no rejection is happening). So I have stopped it.
2615 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2616 However, I am unable to do any testing of this.
2618 34. Fixed an infelicity in the appendfile transport. When checking directories
2619 for a mailbox, to see if any needed to be created, it was accidentally
2620 using path names with one or more superfluous leading slashes; tracing
2621 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2623 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2624 contents are added to the log line that is written for every discarded
2625 recipient. (Previously a log_message setting was ignored.)
2627 36. The ${quote: operator now quotes the string if it is empty.
2629 37. The install script runs exim in order to find its version number. If for
2630 some reason other than non-existence or emptiness, which it checks, it
2631 could not run './exim', it was installing it with an empty version number,
2632 i.e. as "exim-". This error state is now caught, and the installation is
2635 38. An argument was missing from the function that creates an error message
2636 when Exim fails to connect to the socket for saslauthd authentication.
2637 This could cause Exim to crash, or give a corrupted message.
2639 39. Added isip, isip4, and isip6 to ${if conditions.
2641 40. The ACL variables $acl_xx are now saved with the message, and can be
2642 accessed later in routers, transports, and filters.
2644 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2645 strings in the file are not string-expanded.
2647 42. If a MAIL command specified a SIZE value that was too large to fit into an
2648 int variable, the check against message_size_limit failed. Such values are
2649 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2650 day this will have to be increased, but I don't think I want to be around
2651 when emails are that large.
2658 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2659 iconv() is not standard in FreeBSD.
2661 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2662 IPv6 enabled. The observed symptom was a segmentation fault on return from
2663 the function os_common_find_running_interfaces() in src/os.c.
2665 3. In the check_special_case() function in daemon.c I had used "errno" as an
2666 argument name, which causes warnings on some systems. This was basically a
2667 typo, since it was named "eno" in the comments!
2669 4. The code that waits for the clock to tick (at a resolution of some fraction
2670 of a second) so as to ensure message-id uniqueness was always waiting for
2671 at least one whole tick, when it could have waited for less. [This is
2672 almost certainly not relevant at current processor speeds, where it is
2673 unlikely to ever wait at all. But we try to future-proof.]
2675 5. The function that sleeps for a time interval that includes fractions of a
2676 second contained a race. It did not block SIGALRM between setting the
2677 timer, and suspending (a couple of lines later). If the interval was short
2678 and the sigsuspend() was delayed until after it had expired, the suspension
2679 never ended. On busy systems this could lead to processes getting stuck for
2682 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2683 process, before it forks any delivery processes. The open lookup caching
2684 mechanism meant that the open file or database connection was passed into
2685 the delivery process. The problem was that delivery processes always tidy
2686 up cached lookup data. This could cause a problem for the next delivery
2687 process started by the queue runner, because the external queue runner
2688 process does not know about the closure. So the next delivery process
2689 still has data in the lookup cache. In the case of a file lookup, there was
2690 no problem because closing a file descriptor in a subprocess doesn't affect
2691 the parent. However, if the lookup was caching a connection to a database,
2692 the connection was closed, and the second delivery process was likely to
2693 see errors such as "PGSQL: query failed: server closed the connection
2694 unexpectedly". The problem has been fixed by closing all cached lookups
2695 in a queue runner before running a delivery process.
2697 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2698 seem to have the "const" qualifier which it has on other OS. I've
2701 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2702 *of the same type* (client or server) with the same public name that an
2703 error should be diagnosed.
2705 9. When Exim looked up a host name for an IP address, but failed to find the
2706 original IP address when looking up the host name (a safety check), it
2707 output the message "<ip address> does not match any IP for NULL", which was
2708 confusing, to say the least. The bug was that the host name should have
2709 appeared instead of "NULL".
2711 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2712 user that is built into the binary, and the -C or -D options is used, root
2713 privilege is dropped before the configuration file is read. In addition,
2714 logging is switched to stderr instead of the normal log files. If the
2715 configuration then re-defines the Exim user, the unprivileged environment
2716 is probably not what is expected, so Exim logs a panic warning message (but
2719 However, if deliver_drop_privilege is set, the unprivileged state may well
2720 be exactly what is intended, so the warning has been cut out in that case,
2721 and Exim is allowed to try to write to its normal log files.
2727 1. smtp_return_error_details was not giving details for temporary sender
2728 or receiver verification errors.
2730 2. Diagnose a configuration error if two authenticators have the same public
2733 3. Exim used not to create the message log file for a message until the first
2734 delivery attempt. This could be confusing when incoming messages were held
2735 for policy or load reasons. The message log file is now created at the time
2736 the message is received, and an initial "Received" line is written to it.
2738 4. The automatically generated man page for command line options had a minor
2739 bug that caused no ill effects; however, a more serious problem was that
2740 the procedure for building the man page automatically didn't always
2741 operate. Consequently, release 4.20 contains an out-of-date version. This
2742 shouldn't happen again.
2744 5. When building Exim with embedded Perl support, the script that builds the
2745 Makefile was calling 'perl' to find its compile-time parameters, ignoring
2746 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
2748 6. The freeze_tell option was not being used for messages that were frozen on
2749 arrival, either by an ACL or by local_scan().
2751 7. Added the smtp_incomplete_transaction log selector.
2753 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
2754 was accepting AUTH without a new EHLO.
2756 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
2757 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
2760 10. Logging of TCP/IP connections (when configured) now happens in the main
2761 daemon process instead of the child process, so that the TCP/IP connection
2762 count is more accurate (but it can never be perfect).
2764 11. The use of "drop" in a nested ACL was not being handled correctly in the
2765 outer ACL. Now, if condition failure induced by the nested "drop" causes
2766 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
2767 or "require"), the connection is dropped.
2769 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
2770 that yield "discard" can now be used with an "accept" or a "discard" verb,
2771 but an error is generated for any others (because I can't see a useful way
2772 to define what should happen).
2774 13. When an ACL is read dynamically from a file (or anywhere else), the lines
2775 are now processed in the same way as lines in the Exim configuration file.
2776 In particular, continuation lines are supported.
2778 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
2780 15. Added -ti meaning -t -i.
2782 16. Check for letters, digits, hyphens, and dots in the names of dnslist
2783 domains, and warn by logging if others are found.
2785 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
2786 returned from GIFCONF when Exim is trying to find the list of interfaces on
2787 a host. The code in os.c has been modified to copy each ifreq to an aligned
2788 structure in all cases.
2790 Also, in some cases, the returned ifreq's were being copied to a 'struct
2791 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
2792 means the last couple of bytes of an IPv6 address could be chopped if the
2793 ifreq contained only a normal sockaddr (14 bytes storage).
2795 18. Named domain lists were not supported in the hosts_treat_as_local option.
2796 An entry such as +xxxx was not recognized, and was treated as a literal
2799 19. Ensure that header lines added by a DATA ACL are included in the reject log
2800 if the ACL subsequently rejects the message.
2802 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
2803 MD5 (which is deprecated).
2805 21. When testing a filter file using -bf, Exim was writing a message when it
2806 took the sender from a "From " line in the message, but it was not doing so
2807 when it took $return_path from a Return-Path: header line. It now does.
2809 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
2810 with a valid header line field name (a series of printing characters
2811 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
2813 23. Changed "disc" in the source to "disk" to conform to the documentation and
2814 the book and for uniformity.
2816 24. Ignore Sendmail's -Ooption=value command line item.
2818 25. When execve() failed while trying to run a command in a pipe transport,
2819 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
2820 could be confused with a return value of 69 from the command itself. This
2821 has been changed to 127, the value the shell returns if it is asked to run
2822 a non-existent command. The wording for the related log line suggests a
2823 non-existent command as the problem.
2825 26. If received_header_text expands to an empty string, do not add a Received:
2826 header line to the message. (Well, it adds a token one on the spool, but
2827 marks it "old" so that it doesn't get used or transmitted.)
2829 27. Installed eximstats 1.28 (addition of -nt option).
2831 28. There was no check for failure on the call to getsockname() in the daemon
2832 code. This can fail if there is a shortage of resources on the system, with
2833 ENOMEM, for example. A temporary error is now given on failure.
2835 29. Contrary to the C standard, it seems that in some environments, the
2836 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
2837 program. Exim now does this explicitly; it affects the formatting of
2838 timestamps using strftime().
2840 30. If exiqsumm was given junk data, it threw up some uninitialized variable
2841 complaints. I've now initialized all the variables, to avoid this.
2843 32. Header lines added by a system filter were not being "seen" during
2844 transport-time rewrites.
2846 33. The info_callback() function passed to OpenSSL is set up with type void
2847 (*)(SSL *, int, int), as described somewhere. However, when calling the
2848 function (actually a macro) that sets it up, the type void(*)() is
2849 expected. I've put in a cast to prevent warnings from picky compilers.
2851 34. If a DNS black list lookup found a CNAME record, but there were no A
2852 records associated with the domain it pointed at, Exim crashed.
2854 35. If a DNS black list lookup returned more than one A record, Exim ignored
2855 all but the first. It now scans all returned addresses if a particular IP
2856 value is being sought. In this situation, the contents of the
2857 $dnslist_value variable are a list of all the addresses, separated by a
2860 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
2861 to accept a host name and all its aliases if the forward lookup for any of
2862 them yielded the IP address of the incoming connection. Now it accepts only
2863 those names whose forward lookup yields the correct IP address. Any other
2864 names are discarded. This closes a loophole whereby a rogue DNS
2865 administrator could create reverse DNS records to break through a
2866 wildcarded host restriction in an ACL.
2868 37. If a user filter or a system filter that ran in a subprocess used any of
2869 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
2870 the wrong values were passed to the pipe command ($thisaddress had the
2871 value of $0, $0 had the value of $1, etc). This bug was introduced by
2872 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
2874 38. Improved the line breaking for long SMTP error messages from ACLs.
2875 Previously, if there was no break point between 40 and 75 characters, Exim
2876 left the rest of the message alone. Two changes have been made: (a) I've
2877 reduced the minimum length to 35 characters; (b) if it can't find a break
2878 point between 35 and 75 characters, it looks ahead and uses the first one
2879 that it finds. This may give the occasional overlong line, but at least the
2880 remaining text gets split now.
2882 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
2883 file descriptors might be low, and that setting 1000 would always raise it.
2884 It turns out that in some environments, the limit is already over 1000 and
2885 that lowering it causes trouble. So now Exim takes care not to decrease it.
2887 40. When delivering a message, the value of $return_path is set to $sender_
2888 address at the start of routing (routers may change the value). By an
2889 oversight, this default was not being set up when an address was tested by
2890 -bt or -bv, which affected the outcome if any router or filter referred to
2893 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
2894 log only when "message" or "log_message" are set. However, if one of the
2895 conditions was an address verification, or a call to a nested ACL, the
2896 messages generated by the underlying test were being passed through. This
2897 no longer happens. The underlying message is available in $acl_verify_
2898 message for both "message" and "log_message" expansions, so it can be
2899 passed through if needed.
2901 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
2902 new expansion $bh_ to give the encoded byte string without charset
2903 translation. Translation happens only if iconv() is available; HAVE_ICONV
2904 indicates this at build time. HEADERS_CHARSET gives the charset to
2905 translate to; headers_charset can change it in the configuration, and
2906 "headers charset" can change it in an individual filter file.
2908 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
2909 that creates RFC 2047 encoded "words" labels them as that charset instead
2910 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
2911 expansion operator; (ii) when Exim creates a From: line for a local
2912 message; (iii) when a header line is rewritten to include a "phrase" part.
2914 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
2915 buggy, causing it to skip the first lines of messages whose message ID
2916 ended in 'D'. This would not have bitten before Exim release 4.14, because
2917 message IDs were unlikely to end in 'D' before then. The effect was to have
2918 incorrect size information for certain domains.
2920 45. #include "config.h" was missing at the start of the crypt16.c module. This
2921 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
2924 46. If there was a timeout during a "random" callout check, Exim treated it as
2925 a failure of the random address, and carried on sending RSET and the real
2926 address. If the delay was just some slowness somewhere, the response to the
2927 original RCPT would be taken as a response to RSET and so on, causing
2928 mayhem of various kinds.
2930 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
2931 when I implemented it. It didn't allow for the fact that some option values
2932 may legitimatetly be negative (e.g. size_addition), and it didn't even do
2933 the right test for positive values.
2935 48. Domain names in DNS records are case-independent. Exim always looks them up
2936 in lower case. Some resolvers return domain names in exactly the case they
2937 appear in the zone file, that is, they may contain uppercase letters. Not
2938 all resolvers do this - some return always lower case. Exim was treating a
2939 change of case by a resolver as a change of domain, similar to a widening
2940 of a domain abbreviation. This triggered its re-routing code and so it was
2941 trying to route what was effectively the same domain again. This normally
2942 caused routing to fail (because the router wouldn't handle the domain
2943 twice). Now Exim checks for this case specially, and just changes the
2944 casing of the domain that it ultimately uses when it transmits the message
2947 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
2950 50. If a filter generated a file delivery with a non-absolute name (possible if
2951 no home directory exists for the router), the forbid_file option was not
2954 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
2955 the existing equality matching.
2957 52. Exim was using ints instead of ino_t variables in some places where it was
2958 dealing with inode numbers.
2960 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
2961 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
2962 called TMPDIR, and if it finds it is different, it changes its value.
2964 54. The smtp_printf() function is now made available to local_scan() so
2965 additional output lines can be written before returning. There is also an
2966 smtp_fflush() function to enable the detection of a dropped connection.
2967 The variables smtp_input and smtp_batched_input are exported to
2970 55. Changed the default runtime configuration: the message "Unknown user"
2971 has been removed from the ACL, and instead placed on the localuser router,
2972 using the cannot_route_message feature. This means that any verification
2973 failures that generate their own messages won't get overridden. Similarly,
2974 the "Unrouteable address" message that was in the ACL for unverifiable
2975 relay addresses has also been removed.
2977 56. Added hosts_avoid_esmtp to the smtp transport.
2979 57. The exicyclog script was not checking for the esoteric option
2980 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2981 will work only if exicyclog is run under the appropriate euid.
2983 58. Following a discussion on the list, the rules by which Exim recognises line
2984 endings on incoming messages have been changed. The -dropcr and drop_cr
2985 options are now no-ops, retained only for backwards compatibility. The
2986 following line terminators are recognized: LF CRLF CR. However, special
2987 processing applies to CR:
2989 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2990 nor a local message in the state where . is a terminator.
2992 (ii) If a bare CR is encountered in a header line, an extra space is added
2993 after the line terminator so as not to end the header. The reasoning
2994 behind this is that bare CRs in header lines are most likely either
2995 to be mistakes, or people trying to play silly games.
2997 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
2998 was being incorrectly given as 18 bytes larger than it should have been.
2999 This is a VOB (very old bug).
3001 60. This may never have affected anything current, but just in case it has:
3002 When the local host is found other than at the start of a list of hosts,
3003 the local host, those with the same MX, and any that follow, are discarded.
3004 When the list in question was part of a longer list of hosts, the following
3005 hosts (not currently being processed) were also being discarded. This no
3006 longer happens. I'm not sure if this situation could ever has previously
3009 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3012 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3013 header. This is something that is recommended in a new Internet Draft, and
3014 is something that is documented as being done by Sendmail. There are two
3015 possible values. For messages generated by the autoreply transport, Exim
3018 Auto-Submitted: auto-replied
3020 whereas for all other generated messages (e.g. bounces) it adds
3022 Auto-Submitted: auto-generated
3024 63. The "personal" condition in filters now includes a test for the
3025 Auto-Submitted: header. If it contains the string "auto-" the message it
3026 not considered personal.
3028 64. Added rcpt_include_affixes as a generic transport option.
3030 65. Added queue_only_override (default true).
3032 66. Added the syslog_duplication option.
3034 67. If what should have been the first header line of a message consisted of
3035 a space followed by a colon, Exim was mis-interpreting it as a header line.
3036 It isn't of course - it is syntactically invalid and should therefore be
3037 treated as the start of the message body. The misbehaviour could have
3038 caused a number of strange effects, including loss of data in subsequent
3039 header lines, and spool format errors.
3041 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3042 client host had authenticated. This control can now be exercised by an ACL
3043 for more flexibility.
3045 69. By default, callouts do not happen when testing with -bh. There is now a
3046 variant, -bhc, which does actually run the callout code, including
3047 consulting and updating the callout cache.
3049 70. Added support for saslauthd authentication, courtesy of Alexander
3052 71. If statvfs() failed on the spool or log directories while checking their
3053 size for availability, Exim confusingly gave the error "space shortage".
3054 Furthermore, in debugging mode it crashed with a floating point exception.
3055 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3056 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3057 serious problem, Exim now writes to the main and panic logs when this
3058 happens, with details of the failure. It then refuses to accept the
3059 incoming message, giving the message "spool directory problem" or "log
3060 directory problem" with a 421 code for SMTP messages.
3062 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3063 0, 1, and 2 exist, because some OS complain for execs without them (see
3064 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3065 descriptors. However, the code omitted to check that the open succeeded,
3066 causing mysterious errors if for some reason the permissions on /dev/null
3067 got screwed. Now Exim writes a message to the main and panic logs, and
3068 bombs out if it can't open /dev/null.
3070 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3071 interact so that it is all more flexible. It is supposed to remain
3072 backwards compatible. Also added extra_local_interfaces.
3074 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3075 to bomb out with an assertion failure - to the client this appears as a
3076 connection drop. This problem occurs in the part of the code that was taken
3077 from the Samba project. Fortunately, the assertion is in a very simple
3078 function, so I have fixed this by reproducing the function inline in the
3079 one place where it is called, and arranging for authentication to fail
3080 instead of killing the process with assert().
3082 75. The SPA client code was not working when the server requested OEM rather
3083 than Unicode encoding.
3085 76. Added code to make require_files with a specific uid setting more usable in
3086 the case where statting the file as root fails - usually a non-root-mounted
3087 NFS file system. When this happens and the failure is EACCES, Exim now
3088 forks a subprocess and does the per-uid checking as the relevant uid.
3090 77. Added process_log_path.
3092 78. If log_file_path was not explicitly set, a setting of check_log_space or
3093 check_log_inodes was ignored.
3095 79. If a space check for the spool or log partitions fails, the incident is now
3096 logged. Of course, in the latter case the data may get lost...
3098 80. Added the %p formatting code to string_format() so that it can be used to
3099 print addresses in debug_print(). Adjusted all the address printing in the
3100 debugging in store.c to use %p rather than %d.
3102 81. There was a concern that a line of code in smtp_in.c could overflow a
3103 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3104 initially expressed, the concern was not well-founded, because trailing
3105 spaces are removed early. However, if the trailing spaces were followed by
3106 a NULL, they did not get removed, so the overflow was possible. Two fixes
3109 (a) I re-wrote the offending code in a cleaner fashion.
3110 (b) If an incoming SMTP command contains a NULL character, it is rejected
3113 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3114 runs initgroups(), so that if the Exim user is in any additional groups,
3115 they will be used during message reception.
3121 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob