1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.148 2005/06/06 19:23:03 tom 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.
40 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
42 TK/03 Merged latest SRS patch from Miles Wilton.
44 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
45 with the definition in sysexits.h (which is #included earlier).
46 Fortunately, Exim does not actually use EX_OK. The code used to try to
47 preserve the sysexits.h value, by assumimg that macro definitions were
48 scanned for macro replacements. I have been disabused of this notion,
49 so now the code just undefines EX_OK before #including unistd.h.
51 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
52 in the smtp transport. When a block could not be written in a single
53 write() function, the timeout was being re-applied to each part-write.
54 This seems wrong - if the receiver was accepting one byte at a time it
55 would take for ever. The timeout is now adjusted when this happens. It
56 doesn't have to be particularly precise.
58 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
59 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
61 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
62 without a domain if the check to <postmaster@domain> fails.
64 SC/01 Eximstats: added -xls and the ability to specify output files
65 (patch written by Frank Heydlauf).
67 SC/02 Eximstats: use FileHandles for outputing results.
69 SC/03 Eximstats: allow any combination of xls, txt, and html output.
71 SC/04 Eximstats: fixed display of large numbers with -nvr option
73 SC/05 Eximstats: fixed merging of reports with empty tables.
75 SC/06 Eximstats: added the -include_original_destination flag
77 SC/07 Eximstats: removed tabs and trailing whitespace.
79 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
81 TK/06 MBOX spool code: Add real "From " MBOX separator line
82 so the .eml file is really in mbox format (even though
83 most programs do not really care). Patch from Alex Miller.
85 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
86 The latter is generated from $received_to and is only set if the
87 message has one envelope recipient. SA can use these headers,
88 obviously out-of-the-box. Patch from Alex Miller.
94 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
95 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
97 TK/02 Fix ACL "control" statement not being available in MIME ACL.
99 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
101 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
102 to test Sieve filters that use "vacation".
104 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
105 that changes the way the GnuTLS parameters are stored in the cache file.
106 The new format can be generated externally. For backward compatibility,
107 if the data in the cache doesn't make sense, Exim assumes it has read an
108 old-format file, and it generates new data and writes a new file. This
109 means that you can't go back to an older release without removing the
112 PH/03 A redirect router that has both "unseen" and "one_time" set does not
113 work if there are any delivery delays because "one_time" forces the
114 parent to be marked "delivered", so its unseen clone is never tried
115 again. For this reason, Exim now forbids the simultaneous setting of
118 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
119 redirected to themselves ("homonym" addresses). Read the long ChangeLog
120 entry if you want to know the details. The fix, however, neglected to
121 consider the case when local delivery batching is involved. The test for
122 "previously delivered" was not happening when checking to see if an
123 address could be batched with a previous (undelivered) one; under
124 certain circumstances this could lead to multiple deliveries to the same
127 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
128 in its include files, and this causes problems building Exim.
130 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
131 header_syntax) but Exim was just ignoring anything given after a slash.
132 In particular, this caused confusion with an attempt to use "verify =
133 reverse_host_lookup/defer_ok". An error is now given when options are
134 supplied for verify items that do not have them. (Maybe reverse_host_
135 lookup should have a defer_ok option, but that's a different point.)
137 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
138 defined by RFC 821) to 2048, because there were problems with some AUTH
139 commands, and RFC 1869 says the size should be increased for extended
140 SMTP commands that take arguments.
142 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
145 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
146 "unknown" error; now it says that the functionality isn't in the binary.
148 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
149 an address' error message when a string expansion fails (syntax or
150 whatever). Otherwise the password may appear in the log. Following change
151 PH/42 below, there is no longer a chance of it appearing in a bounce
154 PH/11 Installed exipick version 20050225.0 from John Jetmore.
156 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
157 of its addresses was ever tried. (Bugzilla bug #2.)
159 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
160 the result incorrectly in the debug output. (It correctly added a newline
161 to what was transported.)
163 TF/01 Added $received_time.
165 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
166 commented out examples of how to interface to a virus scanner and to
167 SpamAssassin. Also added commented examples of av_scanner and
168 spamd_address settings.
170 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
171 and controls are allowed in which ACLs. There were a couple of minor
172 errors. Some of the entries in the conditions table (which is a table of
173 where they are NOT allowed) were getting very unwieldy; rewrote them as a
174 negation of where the condition IS allowed.
176 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
178 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
179 header file does not have a version number, so I've had to invent a new
180 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
181 API. The code is untested by me (my Linux distribution still has 0.3.2 of
182 radiusclient), but it was contributed by a Radius user.
184 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
185 files or directories, for OpenSSL.
187 PH/19 When an Exim process that is running as root has to create an Exim log
188 file, it does so in a subprocess that runs as exim:exim so as to get the
189 ownership right at creation (otherwise, other Exim processes might see
190 the file with the wrong ownership). There was no test for failure of this
191 fork() call, which would lead to the process getting stuck as it waited
192 for a non-existent subprocess. Forks do occasionally fail when resources
193 run out. I reviewed all the other calls to fork(); they all seem to check
196 PH/20 When checking for unexpected SMTP input at connect time (before writing
197 the banner), Exim was not dealing correctly with a non-positive return
198 from the read() function. If the client had disconnected by this time,
199 the result was a log entry for a synchronization error with an empty
200 string after "input=" when read() returned zero. If read() returned -1
201 (an event I could not check), uninitialized data bytes were printed.
202 There were reports of junk text (parts of files, etc) appearing after
205 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
207 PH/22 Added support for macro redefinition, and (re)definition in between
208 driver and ACL definitions.
210 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
211 forgetting to use the resulting value; it was using the unexpanded value.
213 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
214 hadn't been configured. The fix is from Juergen Kreileder, who
215 understands it better than I do:
217 "Here's what I see happening with three configured cyrus_sasl
218 authenticators configured (plain, login, cram-md5):
220 On startup auth_cyrus_sasl_init() gets called for each of these.
221 This means three calls to sasl_listmech() without a specified mech_list.
222 => SASL tests which mechs of all available mechs actually work
223 => three warnings about OTP not working
224 => the returned list contains: plain, login, cram-md5, digest-md5, ...
226 With the patch, sasl_listmech() also gets called three times. But now
227 SASL's mech_list option is set to the server_mech specified in the the
228 authenticator. Or in other words, the answer from sasl_listmech()
229 gets limited to just the mech you're testing for (which is different
231 => the return list contains just 'plain' or 'login', 'cram-md5' or
232 nothing depending on the value of ob->server_mech.
234 I've just tested the patch: Authentication still works fine,
235 unavailable mechs specified in the exim configuration are still
236 caught, and the auth.log warnings about OTP are gone."
238 PH/25 When debugging is enabled, the contents of the command line are added
239 to the debugging output, even when log_selector=+arguments is not
242 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
243 answer is "GNU", and only if the return is "GNU/something" is the answer
246 PH/27 $acl_verify_message is now set immediately after the failure of a
247 verification in an ACL, and so is available in subsequent modifiers. In
248 particular, the message can be preserved by coding like this:
250 warn !verify = sender
251 set acl_m0 = $acl_verify_message
253 Previously, $acl_verify_message was set only while expanding "message"
254 and "log_message" when a very denied access.
256 PH/28 Modified OS/os.c-Linux with
258 -#ifndef OS_LOAD_AVERAGE
259 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
261 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
262 nomenclature these days.)
264 PH/29 Installed patch from the Sieve maintainer that adds the options
265 sieve_useraddress and sieve_subaddress to the redirect router.
267 PH/30 In these circumstances:
268 . Two addresses routed to the same list of hosts;
269 . First host does not offer TLS;
270 . First host accepts first address;
271 . First host gives temporary error to second address;
272 . Second host offers TLS and a TLS session is established;
273 . Second host accepts second address.
274 Exim incorrectly logged both deliveries with the TLS parameters (cipher
275 and peerdn, if requested) that were in fact used only for the second
278 PH/31 When doing a callout as part of verifying an address, Exim was not paying
279 attention to any local part prefix or suffix that was matched by the
280 router that accepted the address. It now behaves in the same way as it
281 does for delivery: the affixes are removed from the local part unless
282 rcpt_include_affixes is set on the transport.
284 PH/32 Add the sender address, as F=<...>, to the log line when logging a
285 timeout during the DATA phase of an incoming message.
287 PH/33 Sieve envelope tests were broken for match types other than :is. I have
288 applied a patch sanctioned by the Sieve maintainer.
290 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
291 the uid or gid is negative. A case of a negative gid caused this to be
292 noticed. The fix allows for either to be negative.
294 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
295 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
298 PH/36 The change PH/12 above was broken. Fixed it.
300 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
301 the grounds that routing the same address twice would always produce the
302 same answer. This might have been true once, but it is certainly no
303 longer true now. Routing a child address may depend on the previous
304 routing that produced that child. Some complicated redirection strategies
305 went wrong when messages had multiple recipients, and made Exim's
306 behaviour dependent on the order in which the addresses were given.
308 I have moved the duplicate checking until after the routing is complete.
309 Exim scans the addresses that are assigned to local and remote
310 transports, and removes any duplicates. This means that more work will be
311 done, as duplicates will always all be routed, but duplicates are
312 presumably rare, so I don't expect this is of any significance.
314 For deliveries to pipes, files, and autoreplies, the duplicate checking
315 still happens during the routing process, since they are not going to be
318 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
319 It corrects a timeout issue with spamd. This is Ian's comment: "The
320 background is that sometimes spamd either never reads data from a
321 connection it has accepted, or it never writes response data. The exiscan
322 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
323 blindly assumes that writes won't block so it may never time out."
325 PH/39 Allow G after quota size as well as K and M.
327 PH/40 The value set for $authenticated_id in an authenticator may not contain
328 binary zeroes or newlines because the value is written to log lines and
329 to spool files. There was no check on this. Now the value is run through
330 the string_printing() function so that such characters are converted to
331 printable escape sequences.
333 PH/41 $message_linecount is a new variable that contains the total number of
334 lines in the message. Compare $body_linecount, which is the count for the
337 PH/42 Exim no longer gives details of delivery errors for specific addresses in
338 bounce and delay warning messages, except in certain special cases, which
341 (a) An SMTP error message from a remote host;
342 (b) A message specified in a :fail: redirection;
343 (c) A message specified in a "fail" command in a system filter;
344 (d) A message specified in a FAIL return from the queryprogram router;
345 (e) A message specified by the cannot_route_message router option.
347 In these cases only, Exim does include the error details in bounce and
348 warning messages. There are also a few cases where bland messages such
349 as "unrouteable address" or "local delivery error" are given.
351 PH/43 $value is now also set for the "else" part of a ${run expansion.
353 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
354 being worked on, but at least Exim now implements the latest version to
357 PH/45 In a pipe transport, although a timeout while waiting for the pipe
358 process to complete was treated as a delivery failure, a timeout while
359 writing the message to the pipe was logged, but erroneously treated as a
360 successful delivery. Such timeouts include transport filter timeouts. For
361 consistency with the overall process timeout, these timeouts are now
362 treated as errors, giving rise to delivery failures by default. However,
363 there is now a new Boolean option for the pipe transport called
364 timeout_defer, which, if set TRUE, converts the failures into defers for
365 both kinds of timeout. A transport filter timeout is now identified in
368 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
369 systems where "make" and "gmake" are different, calling "gmake" at top
370 level broke things. I've arranged for the value of $(MAKE) to be passed
371 from the Makefile to this script so that it can call the same version of
375 A note about Exim versions 4.44 and 4.50
376 ----------------------------------------
378 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
379 changes of various kinds. As a consequence, a big documentation update was
380 needed. This delayed the release for rather longer than seemed good, especially
381 in the light of a couple of (minor) security issues. Therefore, the changes
382 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
383 release. So 4.44 and 4.50 are in effect two different branches that both start
386 I have left the 4.50 change log unchanged; it contains all the changes since
387 4.43. The change log for 4.44 is below; many of its items are identical to
388 those for 4.50. This seems to be the most sensible way to preserve the
389 historical information.
395 1. Minor wording change to the doc/README.SIEVE file.
397 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
398 computation of the current number of files was incorrect.
400 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
401 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
404 4. Give more explanation in the error message when the command for a transport
405 filter fails to execute.
407 5. There are several places where Exim runs a non-Exim command in a
408 subprocess. The SIGUSR1 signal should be disabled for these processes. This
409 was being done only for the command run by the queryprogram router. It is
410 now done for all such subprocesses. The other cases are: ${run, transport
411 filters, and the commands run by the lmtp and pipe transports.
413 6. Added CONFIGURE_GROUP build-time option.
415 7. Some older OS have a limit of 256 on the maximum number of file
416 descriptors. Exim was using setrlimit() to set 1000 as a large value
417 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
418 systems. I've change it so that if it can't get 1000, it tries for 256.
420 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
421 was an oversight, and furthermore, ever since the addition of extra
422 controls (e.g. 4.43/32), the checks on when to allow different forms of
423 "control" were broken. There should now be diagnostics for all cases when a
424 control that does not make sense is encountered.
426 9. Added the /retain_sender option to "control=submission".
428 10. $recipients is now available in the predata ACL (oversight).
430 11. Tidy the search cache before the fork to do a delivery from a message
431 received from the command line. Otherwise the child will trigger a lookup
432 failure and thereby defer the delivery if it tries to use (for example) a
433 cached ldap connection that the parent has called unbind on.
435 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
436 of $address_data from the recipient verification was clobbered by the
439 13. The value of address_data from a sender verification is now available in
440 $sender_address_data in subsequent conditions in the ACL statement.
442 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
444 15. Added a new option "connect=<time>" to callout options, to set a different
447 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
448 was its contents. (It was OK if the option was not defined at all.)
450 17. A "Completed" log line is now written for messages that are removed from
451 the spool by the -Mrm option.
453 18. New variables $sender_verify_failure and $recipient_verify_failure contain
454 information about exactly what failed.
456 19. Added -dd to debug only the daemon process.
458 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
459 handles timeouts, both on the server side and network timeouts. Renamed the
460 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
462 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
463 It is now set to "smtps".
465 22. $host_address is now set to the target address during the checking of
468 23. When checking ignore_target_hosts for an ipliteral router, no host name was
469 being passed; this would have caused $sender_host_name to have been used if
470 matching the list had actually called for a host name (not very likely,
471 since this list is usually IP addresses). A host name is now passed as
474 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
475 code that specifies a non-restarting handler (typically sigaction() in
476 modern systems) in an attempt to fix a rare and obscure crash bug.
478 25. Narrowed the window for a race in the daemon that could cause it to ignore
479 SIGCHLD signals. This is not a major problem, because they are used only to
480 wake it up if nothing else does.
482 26. A malformed maildirsize file could cause Exim to calculate negative values
483 for the mailbox size or file count. Odd effects could occur as a result.
484 The maildirsize information is now recalculated if the size or filecount
487 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
488 support for a long time. Removed HAVE_SYS_VFS_H.
490 28. Installed the latest version of exipick from John Jetmore.
492 29. In an address list, if the pattern was not a regular expression, an empty
493 subject address (from a bounce message) matched only if the pattern was an
494 empty string. Non-empty patterns were not even tested. This was the wrong
495 because it is perfectly reasonable to use an empty address as part of a
496 database query. An empty address is now tested by patterns that are
497 lookups. However, all the other forms of pattern expect the subject to
498 contain a local part and a domain, and therefore, for them, an empty
499 address still always fails if the pattern is not itself empty.
501 30. Exim went into a mad DNS loop when attempting to do a callout where the
502 host was specified on an smtp transport, and looking it up yielded more
505 31. Re-factored the code for checking spool and log partition space into a
506 function that finds that data and another that does the check. The former
507 is then used to implement four new variables: $spool_space, $log_space,
508 $spool_inodes, and $log_inodes.
510 32. The RFC2047 encoding function was originally intended for short strings
511 such as real names; it was not keeping to the 75-character limit for
512 encoded words that the RFC imposes. It now respects the limit, and
513 generates multiple encoded words if necessary. To be on the safe side, I
514 have increased the buffer size for the ${rfc2047: expansion operator from
517 33. It is now permitted to omit both strings after an "if" condition; if the
518 condition is true, the result is "true". As before, when the second string
519 is omitted, a false condition yields an empty string. This makes it less
520 cumbersome to write custom ACL and router conditions.
522 34. Failure to deliver a bounce message always caused it to be frozen, even if
523 there was an errors_to setting on the router. The errors_to setting is now
526 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
527 canonical form (fully expanded) before being placed in
528 $sender_host_address.
530 36. The table in the code that translates DNS record types into text (T_A to
531 "A" for instance) was missing entries for NS and CNAME. It is just possible
532 that this could have caused confusion if both these types were looked up
533 for the same domain, because the text type is used as part of Exim's
534 per-process caching. But the chance of anyone hitting this buglet seems
537 37. The dnsdb lookup has been extended in a number of ways.
539 (1) There is a new type, "zns", which walks up the domain tree until it
540 finds some nameserver records. It should be used with care.
542 (2) There is a new type, "mxh", which is like "mx" except that it returns
543 just the host names, not the priorities.
545 (3) It is now possible to give a list of domains (or IP addresses) to be
546 looked up. The behaviour when one of the lookups defers can be
547 controlled by a keyword.
549 (4) It is now possible to specify the separator character for use when
550 multiple records are returned.
552 38. The dnslists ACL condition has been extended: it is now possible to supply
553 a list of IP addresses and/or domains to be looked up in a particular DNS
556 39. Added log_selector=+queue_time_overall.
558 40. When running the queue in the test harness, wait just a tad after forking a
559 delivery process, to get repeatability of debugging output.
561 41. Include certificate and key file names in error message when GnuTLS fails
562 to set them up, because the GnuTLS error message doesn't include the name
563 of the failing file when there is a problem reading it.
565 42. Allow both -bf and -bF in the same test run.
567 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
569 44. The "Exiscan patch" is now merged into the mainline Exim source.
571 45. Sometimes the final signoff response after QUIT could fail to get
572 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
573 < 0 before doing a fflush(). This bug looks as though it goes back to the
574 introduction of TLS in release 3.20, but "sometimes" must have been rare
575 because the tests only now provoked it.
577 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
578 (this can affect the format of dates).
580 47. exim_tidydb, when checking for the continued existence of a message for
581 which it has found a message-specific retry record, was not finding
582 messages that were in split spool directories. Consequently, it was
583 deleting retry records that should have stayed in existence.
585 48. Steve fixed some bugs in eximstats.
587 49. The SPA authentication driver was not abandoning authentication and moving
588 on to the next authenticator when an expansion was forced to fail,
589 contradicting the general specification for all authenticators. Instead it
590 was generating a temporary error. It now behaves as specified.
592 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
593 (the order specifies the preference for clients). The order is now AES256,
594 AES128, 3DES, ARCFOUR128.
596 51. Small patch to Sieve code - explicitly set From: when generating an
599 52. Exim crashed if a remote delivery caused a very long error message to be
600 recorded - for instance if somebody sent an entire SpamAssassin report back
601 as a large number of 550 error lines. This bug was coincidentally fixed by
602 increasing the size of one of Exim's internal buffers (big_buffer) that
603 happened as part of the Exiscan merge. However, to be on the safe side, I
604 have made the code more robust (and fixed the comments that describe what
607 53. Now that there can be additional text after "Completed" in log lines (if
608 the queue_time_overall log selector is set), a one-byte patch to exigrep
609 was needed to allow it to recognize "Completed" as not the last thing in
612 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
613 patch that reportedly fixes this has been added. I am not expert enough to
614 create a test for it. This is what the patch creator wrote:
616 "I found a little strange behaviour of ldap code when working with
617 Windows 2003 AD Domain, where users was placed in more than one
618 Organization Units. When I tried to give exim partial DN, the exit code
619 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
620 But simultaneously result of request was absolutely normal ldap result,
621 so I produce this patch..."
623 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
624 REFERENCE, so I have modified the code to exclude the patch when that macro
627 55. Some experimental protocols are using DNS PTR records for new purposes. The
628 keys for these records are domain names, not reversed IP addresses. The
629 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
630 leaves it alone. Component reversal etc. now happens only for IP addresses.
633 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
635 57. Double the size of the debug message buffer (to 2048) so that more of very
636 long debug lines gets shown.
638 58. The exicyclog utility now does better if the number of log files to keep
639 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
641 59. Two changes related to the smtp_active_hostname option:
643 (1) $smtp_active_hostname is now available as a variable.
644 (2) The default for smtp_banner uses $smtp_active_hostname instead
645 of $primary_hostname.
647 60. The host_aton() function is supposed to be passed a string that is known
648 to be a valid IP address. However, in the case of IPv6 addresses, it was
649 not checking this. This is a hostage to fortune. Exim now panics and dies
650 if the condition is not met. A case was found where this could be provoked
651 from a dnsdb PTR lookup with an IPv6 address that had more than 8
652 components; fortuitously, this particular loophole had already been fixed
653 by change 4.50/55 above.
655 If there are any other similar loopholes, the new check in host_aton()
656 itself should stop them being exploited. The report I received stated that
657 data on the command line could provoke the exploit when Exim was running as
658 exim, but did not say which command line option was involved. All I could
659 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
663 61. There was a buffer overflow vulnerability in the SPA authentication code
664 (which came originally from the Samba project). I have added a test to the
665 spa_base64_to_bits() function which I hope fixes it.
668 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
669 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
671 63. The daemon start-up calls getloadavg() while still root for those OS that
672 need the first call to be done as root, but it missed one case: when
673 deliver_queue_load_max is set with deliver_drop_privilege. This is
674 necessary for the benefit of the queue runner, because there is no re-exec
675 when deliver_drop_privilege is set.
677 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
680 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
681 was in use, was not putting the data itself into the right store pool;
682 consequently, it could be overwritten for a subsequent message in the same
683 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
686 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
688 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
689 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
690 the function were treating the return as a boolean value, which happened to
691 work because 0=false and not-0=true, but is not correct code.
693 68. The host_aton() function was not handling scoped IPv6 addresses (those
694 with, for example, "%eth0" on the end) correctly.
696 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
697 negated items (that is, ~something) in unsigned ints. Some compilers
698 apparently mutter when there is no cast.
700 70. If an address verification called from an ACL failed, and did not produce a
701 user-specific message (i.e. there was only a "system" message), nothing was
702 put in $acl_verify_message. In this situation, it now puts the system
705 71. Change 4.23/11 added synchronization checking at the start of an SMTP
706 session; change 4.31/43 added the unwanted input to the log line - except
707 that it did not do this in the start of session case. It now does.
709 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
710 This is wrong and can cause the other end to generate a synchronization
711 error if it is another Exim or anything else that does the synchronization
712 check. A QUIT command is no longer sent after a timeout.
714 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
717 74. The defer_ok option of callout verification was not working if it was used
718 when verifying addresses in header lines, that is, for this case:
720 verify = header_sender/callout=defer_ok
722 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
723 those file descriptors could be used for SMTP connections. If anything
724 wrote to stderr (the example that came up was "warn" in embedded Perl), it
725 could be sent to the SMTP client, causing chaos. The daemon now opens
726 stdin, stdout, and stderr to /dev/null when it puts itself into the
729 76. Arrange for output from Perl's "warn" command to be written to Exim's main
730 log by default. The user can override this with suitable Perl magic.
732 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
733 the log message when discard triggers, was not working for the DATA ACL or
734 for the non-SMTP ACL.
736 78. Error message wording change in sieve.c.
738 79. If smtp_accept_max_per_host was set, the number of connections could be
739 restricted to fewer than expected, because the daemon was trying to set up
740 a new connection before checking whether the processes handling previous
741 connections had finished. The check for completed processes is now done
742 earlier. On busy systems, this bug wouldn't be noticed because something
743 else would have woken the daemon, and it would have reaped the completed
746 80. If a message was submitted locally by a user whose login name contained one
747 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
748 It caused a spool format error. I have fixed the spool reading code. A
749 related problem was that the "from" clause in the Received: line became
750 illegal because of the space(s). It is now covered by ${quote_local_part.
752 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
755 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
757 83. Patch from Sieve maintainer to fix unterminated string problem in
760 84. Some minor changes to the Linux configuration files to help with other
761 OS variants using glibc.
763 85. One more patch for Sieve to update vacation handling to latest spec.
766 ----------------------------------------------------
767 See the note above about the 4.44 and 4.50 releases.
768 ----------------------------------------------------
774 1. Change 4.43/35 introduced a bug that caused file counts to be
775 incorrectly computed when quota_filecount was set in an appendfile
778 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
779 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
782 3. Give more explanation in the error message when the command for a transport
783 filter fails to execute.
785 4. There are several places where Exim runs a non-Exim command in a
786 subprocess. The SIGUSR1 signal should be disabled for these processes. This
787 was being done only for the command run by the queryprogram router. It is
788 now done for all such subprocesses. The other cases are: ${run, transport
789 filters, and the commands run by the lmtp and pipe transports.
791 5. Some older OS have a limit of 256 on the maximum number of file
792 descriptors. Exim was using setrlimit() to set 1000 as a large value
793 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
794 systems. I've change it so that if it can't get 1000, it tries for 256.
796 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
797 was an oversight, and furthermore, ever since the addition of extra
798 controls (e.g. 4.43/32), the checks on when to allow different forms of
799 "control" were broken. There should now be diagnostics for all cases when a
800 control that does not make sense is encountered.
802 7. $recipients is now available in the predata ACL (oversight).
804 8. Tidy the search cache before the fork to do a delivery from a message
805 received from the command line. Otherwise the child will trigger a lookup
806 failure and thereby defer the delivery if it tries to use (for example) a
807 cached ldap connection that the parent has called unbind on.
809 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
810 of $address_data from the recipient verification was clobbered by the
813 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
814 was its contents. (It was OK if the option was not defined at all.)
816 11. A "Completed" log line is now written for messages that are removed from
817 the spool by the -Mrm option.
819 12. $host_address is now set to the target address during the checking of
822 13. When checking ignore_target_hosts for an ipliteral router, no host name was
823 being passed; this would have caused $sender_host_name to have been used if
824 matching the list had actually called for a host name (not very likely,
825 since this list is usually IP addresses). A host name is now passed as
828 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
829 code that specifies a non-restarting handler (typically sigaction() in
830 modern systems) in an attempt to fix a rare and obscure crash bug.
832 15. Narrowed the window for a race in the daemon that could cause it to ignore
833 SIGCHLD signals. This is not a major problem, because they are used only to
834 wake it up if nothing else does.
836 16. A malformed maildirsize file could cause Exim to calculate negative values
837 for the mailbox size or file count. Odd effects could occur as a result.
838 The maildirsize information is now recalculated if the size or filecount
841 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
842 support for a long time. Removed HAVE_SYS_VFS_H.
844 18. Updated exipick to current release from John Jetmore.
846 19. Allow an empty sender to be matched against a lookup in an address list.
847 Previously the only cases considered were a regular expression, or an
850 20. Exim went into a mad DNS lookup loop when doing a callout where the
851 host was specified on the transport, if the DNS lookup yielded more than
854 21. The RFC2047 encoding function was originally intended for short strings
855 such as real names; it was not keeping to the 75-character limit for
856 encoded words that the RFC imposes. It now respects the limit, and
857 generates multiple encoded words if necessary. To be on the safe side, I
858 have increased the buffer size for the ${rfc2047: expansion operator from
861 22. Failure to deliver a bounce message always caused it to be frozen, even if
862 there was an errors_to setting on the router. The errors_to setting is now
865 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
866 canonical form (fully expanded) before being placed in
867 $sender_host_address.
869 24. Updated eximstats to version 1.33
871 25. Include certificate and key file names in error message when GnuTLS fails
872 to set them up, because the GnuTLS error message doesn't include the name
873 of the failing file when there is a problem reading it.
875 26. Expand error message when OpenSSL has problems setting up cert/key files.
878 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
879 (this can affect the format of dates).
881 28. exim_tidydb, when checking for the continued existence of a message for
882 which it has found a message-specific retry record, was not finding
883 messages that were in split spool directories. Consequently, it was
884 deleting retry records that should have stayed in existence.
886 29. eximstats updated to version 1.35
887 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
888 1.35 - bugfix such that pie charts by volume are generated correctly
890 30. The SPA authentication driver was not abandoning authentication and moving
891 on to the next authenticator when an expansion was forced to fail,
892 contradicting the general specification for all authenticators. Instead it
893 was generating a temporary error. It now behaves as specified.
895 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
896 (the order specifies the preference for clients). The order is now AES256,
897 AES128, 3DES, ARCFOUR128.
899 31. Small patch to Sieve code - explicitly set From: when generating an
902 32. Exim crashed if a remote delivery caused a very long error message to be
903 recorded - for instance if somebody sent an entire SpamAssassin report back
904 as a large number of 550 error lines. This bug was coincidentally fixed by
905 increasing the size of one of Exim's internal buffers (big_buffer) that
906 happened as part of the Exiscan merge. However, to be on the safe side, I
907 have made the code more robust (and fixed the comments that describe what
910 33. Some experimental protocols are using DNS PTR records for new purposes. The
911 keys for these records are domain names, not reversed IP addresses. The
912 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
913 leaves it alone. Component reversal etc. now happens only for IP addresses.
916 34. The host_aton() function is supposed to be passed a string that is known
917 to be a valid IP address. However, in the case of IPv6 addresses, it was
918 not checking this. This is a hostage to fortune. Exim now panics and dies
919 if the condition is not met. A case was found where this could be provoked
920 from a dnsdb PTR lookup with an IPv6 address that had more than 8
921 components; fortuitously, this particular loophole had already been fixed
922 by change 4.50/55 or 4.44/33 above.
924 If there are any other similar loopholes, the new check in host_aton()
925 itself should stop them being exploited. The report I received stated that
926 data on the command line could provoke the exploit when Exim was running as
927 exim, but did not say which command line option was involved. All I could
928 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
932 35. There was a buffer overflow vulnerability in the SPA authentication code
933 (which came originally from the Samba project). I have added a test to the
934 spa_base64_to_bits() function which I hope fixes it.
937 36. The daemon start-up calls getloadavg() while still root for those OS that
938 need the first call to be done as root, but it missed one case: when
939 deliver_queue_load_max is set with deliver_drop_privilege. This is
940 necessary for the benefit of the queue runner, because there is no re-exec
941 when deliver_drop_privilege is set.
943 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
944 was in use, was not putting the data itself into the right store pool;
945 consequently, it could be overwritten for a subsequent message in the same
946 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
949 38. Sometimes the final signoff response after QUIT could fail to get
950 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
951 < 0 before doing a fflush(). This bug looks as though it goes back to the
952 introduction of TLS in release 3.20, but "sometimes" must have been rare
953 because the tests only now provoked it.
959 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
960 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
961 Now it returns an integer. A number of calls were still expecting a T/F
962 return. Fortuitously, in all cases, the tests worked in OK situations,
963 which is the norm. However, things would have gone wrong on any write
964 failures on the smtp file descriptor. This function is used when sending
965 messages over SMTP and also when doing verify callouts.
967 2. When Exim is called to do synchronous delivery of a locally submitted
968 message (the -odf or -odi options), it no longer closes stderr before doing
971 3. Implemented the mua_wrapper option.
973 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
975 5. Implemented the functions header_remove(), header_testname(),
976 header_add_at_position(), and receive_remove_recipient(), and exported them
979 6. If an ACL "warn" statement specified the addition of headers, Exim already
980 inserted X-ACL-Warn: at the start if there was no header name. However, it
981 was not making this test for the second and subsequent header lines if
982 there were newlines in the string. This meant that an invalid header could
983 be inserted if Exim was badly configured.
985 7. Allow an ACL "warn" statement to add header lines at the start or after all
986 the Received: headers, as well as at the end.
988 8. Added the rcpt_4xx retry error code.
990 9. Added postmaster_mailfrom=xxx to callout verification option.
992 10. Added mailfrom=xxxx to the callout verification option, for verify=
995 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
996 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
998 12. Inserted some casts to stop certain compilers warning when using pointer
999 differences as field lengths or precisions in printf-type calls (mostly
1000 affecting debugging statements).
1002 13. Added optional readline() support for -be (dynamically loaded).
1004 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1005 same clock tick as a message's arrival, so that its received time was the
1006 same as the "first fail" time on the retry record, and that message
1007 remained on the queue past the ultimate address timeout, every queue runner
1008 would try a delivery (because it was past the ultimate address timeout) but
1009 after another failure, the ultimate address timeout, which should have then
1010 bounced the address, did not kick in. This was a "< instead of <=" error;
1011 in most cases the first failure would have been in the next clock tick
1012 after the received time, and all would be well.
1014 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1015 being recognized when the domain list was tested by the match_domain
1016 condition in an expansion string.
1018 16. Added the ${str2b64: operator.
1020 17. Exim was always calling setrlimit() to set a large limit for the number of
1021 processes, without checking whether the existing limit was already
1022 adequate. (It did check for the limit on file descriptors.) Furthermore,
1023 errors from getrlimit() and setrlimit() were being ignored. Now they are
1024 logged to the main and panic logs, but Exim does carry on, to try to do its
1025 job under whatever limits there are.
1027 18. Imported PCRE 5.0.
1029 19. Trivial typo in log message " temporarily refused connection" (the leading
1032 20. If the log selector return_path_on_delivery was set and an address was
1033 redirected to /dev/null, the delivery process crashed because it assumed
1034 that a return path would always be set for a "successful" delivery. In this
1035 case, the whole delivery is bypassed as an optimization, and therefore no
1038 21. Internal re-arrangement: the function for sending a challenge and reading
1039 a response while authentication was assuming a zero-terminated challenge
1040 string. It's now changed to take a pointer and a length, to allow for
1041 binary data in such strings.
1043 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1045 23. Exim was not respecting finduser_retries when seeking the login of the
1046 uid under which it was called; it was always trying 10 times. (The default
1047 setting of finduser_retries is zero.) Also, it was sleeping after the final
1048 failure, which is pointless.
1050 24. Implemented tls_on_connect_ports.
1052 25. Implemented acl_smtp_predata.
1054 26. If the domain in control=submission is set empty, Exim assumes that the
1055 authenticated id is a complete email address when it generates From: or
1056 Sender: header lines.
1058 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1059 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1060 chown and chgrp in /bin and hostname in /usr/bin.
1062 28. Exim was keeping the "process log" file open after each use, just as it
1063 does for the main log. This opens the possibility of it remaining open for
1064 long periods when the USR1 signal hits a daemon. Occasional processlog
1065 errors were reported, that could have been caused by this. Anyway, it seems
1066 much more sensible not to leave this file open at all, so that is what now
1069 29. The long-running daemon process does not normally write to the log once it
1070 has entered its main loop, and it closes the log before doing so. This is
1071 so that log files can straightforwardly be renamed and moved. However,
1072 there are a couple of unusual error situations where the daemon does write
1073 log entries, and I had neglected to close the log afterwards.
1075 30. The text of an SMTP error response that was received during a remote
1076 delivery was being truncated at 512 bytes. This is too short for some of
1077 the long messages that one sometimes sees. I've increased the limit to
1080 31. It is now possible to make retry rules that apply only when a message has a
1081 specific sender, in particular, an empty sender.
1083 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1084 it possible to be selective about when SMTP synchronization is enforced.
1086 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1088 32. Implemented hosts_connection_nolog.
1090 33. Added an ACL for QUIT.
1092 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1095 35. Added mailbox_size and mailbox_filecount to appendfile.
1097 36. Added control = no_multiline_responses to ACLs.
1099 37. There was a bug in the logic of the code that waits for the clock to tick
1100 in the case where the clock went backwards by a substantial amount such
1101 that the microsecond fraction of "now" was more than the microsecond
1102 fraction of "then" (but the whole seconds number was less).
1104 38. Added support for the libradius Radius client library this is found on
1105 FreeBSD (previously only the radiusclient library was supported).
1111 1. When certain lookups returned multiple values in the form name=value, the
1112 quoting of the values was not always being done properly. Specifically:
1113 (a) If the value started with a double quote, but contained no whitespace,
1115 (b) If the value contained whitespace other than a space character (i.e.
1116 tabs or newlines or carriage returns) it was not quoted.
1117 This fix has been applied to the mysql and pgsql lookups by writing a
1118 separate quoting function and calling it from the lookup code. The fix
1119 should probably also be applied to nisplus, ibase and oracle lookups, but
1120 since I cannot test any of those, I have not disturbed their existing code.
1122 2. A hit in the callout cache for a specific address caused a log line with no
1123 reason for rejecting RCPT. Now it says "Previous (cached) callout
1124 verification failure".
1126 3. There was an off-by-one bug in the queryprogram router. An over-long
1127 return line was truncated at 256 instead of 255 characters, thereby
1128 overflowing its buffer with the terminating zero. As well as fixing this, I
1129 have increased the buffer size to 1024 (and made a note to document this).
1131 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1132 when Exim is waiting for an SMTP response from a remote server, Exim
1133 restarts its select() call on the socket, thereby resetting its timeout.
1134 This is not a problem when such interrupts are rare. Somebody set up a cron
1135 job to run exiwhat every 2 minutes, which is less than the normal select()
1136 timeout (5 or 10 minutes). This meant that the select() timeout never
1137 kicked in because it was always reset. I have fixed this by comparing the
1138 time when an interrupt arrives with the time at the start of the first call
1139 to select(). If more time than the timeout has elapsed, the interrupt is
1140 treated as a timeout.
1142 5. Some internal re-factoring in preparation for the addition of Sieve
1143 extensions (by MH). In particular, the "personal" test is moved to a
1144 separate function, and given an option for scanning Cc: and Bcc: (which is
1145 not set for Exim filters).
1147 6. When Exim created an email address using the login of the caller as the
1148 local part (e.g. when creating a From: or Sender: header line), it was not
1149 quoting the local part when it contained special characters such as @.
1151 7. Installed new OpenBSD configuration files.
1153 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1154 try to make them clearer.
1156 9. Callout options, other than the timeout value, were being ignored when
1157 verifying sender addresses in header lines. For example, when using
1159 verify = header_sender/callout=no_cache
1161 the cache was (incorrectly) being used.
1163 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1164 only the Cygwin environment.
1166 11. When return_path_on_delivery was set as a log selector, if different remote
1167 addresses in the same message used different return paths and parallel
1168 remote delivery occurred, the wrong values would sometimes be logged.
1169 (Whenever a remote delivery process finished, the return path value from
1170 the most recently started remote delivery process was logged.)
1172 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1173 lines when AUTH and/or TLS are in use. This is the "received protocol"
1174 field. Exim used to use "asmtp" for authenticated SMTP, without any
1175 indication (in the protocol name) for TLS use. Now it follows the RFC and
1176 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1177 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1178 names appear in log lines as well as in Received: header lines.
1180 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1181 extensions, and comparison tests, and to fix some bugs.
1183 14. Changes to the "personal" filter test:
1185 (1) The test was buggy in that it was just doing the equivalent of
1186 "contains" tests on header lines. For example, if a user's address was
1187 anne@some.where, the "personal" test would incorrectly be true for
1189 To: susanne@some.where
1191 This test is now done by extracting each address from the header in turn,
1192 and checking the entire address. Other tests that are part of "personal"
1193 are now done using regular expressions (for example, to check local parts
1194 of addresses in From: header lines).
1196 (2) The list of non-personal local parts in From: addresses has been
1197 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1198 taken from the Sieve specification recommendations.
1200 (3) If the message contains any header line starting with "List-" it is
1201 treated as non-personal.
1203 (4) The test for "circular" in the Subject: header line has been removed
1204 because it now seems ill-conceived.
1206 15. Minor typos in src/EDITME comments corrected.
1208 16. Installed latest exipick from John Jetmore.
1210 17. If headers_add on a router specified a text string that was too long for
1211 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1212 of string_sprintf() is now avoided.
1214 18. $message_body_size was not set (it was always zero) when running the DATA
1215 ACL and the local_scan() function.
1217 19. For the "mail" command in an Exim filter, no default was being set for
1218 the once_repeat time, causing a random time value to be used if "once" was
1219 specified. (If the value happened to be <= 0, no repeat happened.) The
1220 default is now 0s, meaning "never repeat". The "vacation" command was OK
1221 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1222 (I found it when inspecting the code).
1224 20. There is now an overall timeout for performing a callout verification. It
1225 defaults to 4 times the callout timeout, which applies to individual SMTP
1226 commands during the callout. The overall timeout applies when there is more
1227 than one host that can be tried. The timeout is checked before trying the
1228 next host. This prevents very long delays if there are a large number of
1229 hosts and all are timing out (e.g. when the network connections are timing
1230 out). The value of the overall timeout can be changed by specifying an
1231 additional sub-option for "callout", called "maxwait". For example:
1233 verify = sender/callout=5s,maxwait=20s
1235 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1236 to the end before writing, but this should make it even safer).
1238 22. Exim was forgetting that it had advertised PIPELINING for the second and
1239 subsequent messages on an SMTP connection. It was also not resetting its
1240 memory on STARTTLS and an internal HELO.
1242 23. When Exim logs an SMTP synchronization error within a session, it now
1243 records whether PIPELINING has been advertised or not.
1245 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1246 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1247 rather than long int.
1249 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1251 26. Added the never_mail option to autoreply.
1257 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1258 crash if the getsockname() call failed; this can happen if a connection is
1259 closed very soon after it is established. The problem was simply in the
1260 order in which certain operations were done, causing Exim to try to write
1261 to the SMTP stream before it had set up the file descriptor. The bug has
1262 been fixed by making things happen in the correct order.
1268 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1269 before the connection was closed, thus losing the rejection response.
1271 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1272 some early Solaris releases, but causes trouble in current releases where
1273 socklen_t is defined.
1275 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1278 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1281 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1282 that releasing the top of it at the end releases what was used for sub-
1283 expansions (unless the block got too big). However, discard this block if
1284 the first thing is a variable or header, so that we can use its block when
1285 it is dynamic (useful for very large $message_headers, for example).
1287 6. Lookups now cache *every* query, not just the most recent. A new, separate
1288 store pool is used for this. It can be recovered when all lookup caches are
1289 flushed. Lookups now release memory at the end of their result strings.
1290 This has involved some general refactoring of the lookup sources.
1292 7. Some code has been added to the store_xxx() functions to reduce the amount
1293 of flapping under certain conditions.
1295 8. log_incoming_interface used to affect only the <= reception log lines. Now
1296 it causes the local interface and port to be added to several more SMTP log
1297 lines, for example "SMTP connection from", and rejection lines.
1299 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1301 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1303 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1304 could be overwritten at the end of the current message (or the start of a
1305 new message if it was set in a HELO ACL). The value is now preserved for
1306 the duration of the SMTP connection.
1308 12. If a transport had a headers_rewrite setting, and a matching header line
1309 contained an unqualified address, that address was qualified, even if it
1310 did not match any rewriting rules. The underlying bug was that the values
1311 of the flags that permit the existence of unqualified sender and recipient
1312 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1313 non-local messages, and by -bnq for local messages) were not being
1314 preserved with the message after it was received.
1316 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1317 "next input=" as part of the text comprising the host identity instead of
1318 the correct text. The code was using the same buffer for two different
1319 strings. However, depending on which order the printing function evaluated
1320 its arguments, the bug did not always show up. Under Linux, for example, my
1321 test suite worked just fine.
1323 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1324 doesn't work with some older versions of Perl. It has been changed to "my",
1325 which in any case is probably the better facility to use.
1327 15. A really picky compiler found some instances of statements for creating
1328 error messages that either had too many or two few arguments for the format
1331 16. The size of the buffer for calls to the DNS resolver has been increased
1332 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1333 for addresses that have a lot of PTR records. This alleviates a problem; it
1334 does not fully solve it.
1336 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1337 buffer now truncates the list and logs the incident, which is the same
1338 action as happens when Exim is looking up a host name and its aliases.
1339 Previously in this situation something unpredictable would happen;
1340 sometimes it was "internal error: store_reset failed".
1342 18. If a server dropped the connection unexpectedly when an Exim client was
1343 using GnuTLS and trying to read a response, the client delivery process
1344 crashed while trying to generate an error log message.
1346 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1347 string, for example:
1349 warn message = H1: something\nH2: something
1351 the text was added as a single header line from Exim's point of view
1352 though it ended up OK in the delivered message. However, searching for the
1353 second and subsequent header lines using $h_h2: did not work. This has been
1354 fixed. Similarly, if a system filter added multiple headers in this way,
1355 the routers could not see them.
1357 20. Expanded the error message when iplsearch is called with an invalid key to
1358 suggest using net-iplsearch in a host list.
1360 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1361 ACLs are no longer actually imposed (and a message to that effect is
1364 22. If a "gecos" field in a passwd entry contained escaped characters, in
1365 particular, if it contained a \" sequence, Exim got it wrong when building
1366 a From: or a Sender: header from that name. A second bug also caused
1367 incorrect handling when an unquoted " was present following a character
1368 that needed quoting.
1370 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1371 was not being matched caselessly.
1373 24. Arranged for all hyphens in the exim.8 source to be escaped with
1376 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1377 properly. Recipient callout cache records were still being keyed to include
1378 the sender, even when use_sender was set false. This led to far more
1379 callouts that were necessary. The sender is no longer included in the key
1380 when use_sender is false.
1382 26. Added "control = submission" modifier to ACLs.
1384 27. Added the ${base62d: operator to decode base 62 numbers.
1386 28. dnsdb lookups can now access SRV records.
1388 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1389 the configuration file.
1391 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1392 (-v) mode. This makes the output for a verbose queue run more intelligible.
1394 31. Added a use_postmaster feature to recipient callouts.
1396 32. Added the $body_zerocount variable, containing the number of binary zero
1397 bytes in the message body.
1399 33. The time of last modification of the "new" subdirectory is now used as the
1400 "mailbox time last read" when there is a quota error for a maildir
1403 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1405 35. Added +ignore_unknown as a special item in host lists.
1407 36. Code for decoding IPv6 addresses in host lists is now included, even if
1408 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1409 address was recognized as an IP address, but was then not correctly decoded
1410 into binary, causing unexpected and incorrect effects when compared with
1417 1. Very minor rewording of debugging text in manualroute to say "list of
1418 hosts" instead of "hostlist".
1420 2. If verify=header_syntax was set, and a header line with an unqualified
1421 address (no domain) and a large number of spaces between the end of the
1422 name and the colon was received, the reception process suffered a buffer
1423 overflow, and (when I tested it) crashed. This was caused by some obsolete
1424 code that should have been removed. The fix is to remove it!
1426 3. When running in the test harness, delay a bit after writing a bounce
1427 message to get a bit more predictability in the log output.
1429 4. Added a call to search_tidyup() just before forking a reception process. In
1430 theory, someone could use a lookup in the expansion of smtp_accept_max_
1431 per_host which, without the tidyup, could leave open a database connection.
1433 5. Added the variables $recipient_data and $sender_data which get set from a
1434 lookup success in an ACL "recipients" or "senders" condition, or a router
1435 "senders" option, similar to $domain_data and $local_part_data.
1437 6. Moved the writing of debug_print from before to after the "senders" test
1440 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1441 problems for message scanning, either using a data ACL, or using
1442 local_scan() because the Received: header was not generated till after they
1443 were called (in order to set the time as the time of reception completion).
1444 I have revised the way this works. The header is now generated after the
1445 body is received, but before the ACL or local_scan() are called. After they
1446 are run, the timestamp in the header is updated.
1452 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1453 before starting a queue runner without re-exec. This happened only when
1454 deliver_drop_privilege was set or when the Exim user was set to root. The
1455 effect of the bug was that timeouts during subsequent deliveries caused
1456 crashes instead of being properly handled. The handler is now left at its
1457 default (and expected) setting.
1459 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1460 message, again when deliver_drop_privilege is set or Exim is run as root.
1461 The bug described in (1) was not present in this case, but the tidying up
1462 of the other signals was missing. I have made the two cases consistent.
1464 3. The ignore_target_hosts setting on a manualroute router was being ignored
1465 for hosts that were looked up using the /MX notation.
1467 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1470 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1471 operated on the sender address. After changing the $sender_address to <>
1472 for the sender address verify, Exim was re-instated it as the original
1473 (before rewriting) address, but remembering that it had rewritten it, so it
1474 wasn't rewriting it again. This bug also had the effect of breaking the
1475 sender address verification caching when the sender address was rewritten.
1477 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1478 This has been changed so that if the ip literal address matches
1479 ignore_target_hosts, the router declines.
1481 7. Added expansion conditions match_domain, match_address, and match_local_
1482 part (NOT match_host).
1484 8. The placeholder for the Received: header didn't have a length field set.
1486 9. Added code to Exim itself and to exim_lock to test for a specific race
1487 condition that could lead to file corruption when using MBX delivery. The
1488 issue is with the lockfile that is created in /tmp. If this file is removed
1489 after a process has opened it but before that process has acquired a lock,
1490 there is the potential for a second process to recreate the file and also
1491 acquire a lock. This could lead to two Exim processes writing to the file
1492 at the same time. The added code performs the same test as UW imapd; it
1493 checks after acquiring the lock that its file descriptor still refers to
1494 the same named file.
1496 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1497 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1500 11. Added the smtp_active_hostname option. If used, this will typically be made
1501 to depend on the incoming interface address. Because $interface_address is
1502 not set up until the daemon has forked a reception process, error responses
1503 that can happen earlier (such as "too many connections") no longer contain
1506 12. If an expansion in a condition on a "warn" statement fails because a lookup
1507 defers, the "warn" statement is abandoned, and the next ACL statement is
1508 processed. Previously this caused the whole ACL to be aborted.
1510 13. Added the iplsearch lookup type.
1512 14. Added ident_timeout as a log selector.
1514 15. Added tls_certificate_verified as a log selector.
1516 16. Added a global option tls_require_ciphers (compare the smtp transport
1517 option of the same name). This controls incoming TLS connections.
1519 17. I finally figured out how to make tls_require_ciphers do a similar thing
1520 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1521 before starting the TLS session.
1523 18. Tabs are now shown as \t in -bP output.
1525 19. If the log selector return_path_on_delivery was set, Exim crashed when
1526 bouncing a message because it had too many Received: header lines.
1528 20. If two routers both had headers_remove settings, and the first one included
1529 a superfluous trailing colon, the final name in the first list and the
1530 first name in the second list were incorrectly joined into one item (with a
1531 colon in the middle).
1537 1. Added -C and -D options to the exinext utility, mainly to make it easier
1538 to include in the automated testing, but these could be helpful when
1539 multiple configurations are in use.
1541 2. The exinext utility was not formatting the output nicely when there was
1542 an alternate port involved in the retry record key, nor when there was a
1543 message id as well (for retries that were specific to a specific message
1544 and a specific host). It was also confused by IPv6 addresses, because of
1545 the additional colons they contain. I have fixed the IPv4 problem, and
1546 patched it up to do a reasonable job for IPv6.
1548 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1549 delivery, the log line now contains "pipelined" if PIPELINING was used.
1551 4. An SMTP transport process used to panic and die if the bind() call to set
1552 an explicit outgoing interface failed. This has been changed; it is now
1553 treated in the same way as a connect() failure.
1555 5. A reference to $sender_host_name in the part of a conditional expansion
1556 that was being skipped was still causing a DNS lookup. This no longer
1559 6. The def: expansion condition was not recognizing references to header lines
1560 that used bh_ and bheader_.
1562 7. Added the _cache feature to named lists.
1564 8. The code for checking quota_filecount in the appendfile transport was
1565 allowing one more file than it should have been.
1567 9. For compatibility with Sendmail, the command line option
1575 and sets the incoming protocol and host name (for trusted callers). The
1576 host name and its colon can be omitted when only the protocol is to be set.
1577 Note the Exim already has two private options, -pd and -ps, that refer to
1578 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1579 "s", but I don't think that's a major issue.
1581 10. A number of refactoring changes to the code, none of which should affect
1584 (a) The number of logging options was getting close to filling up the
1585 32-bit word that was used as a bit map. I have split them into two classes:
1586 those that are passed in the argument to log_write(), and those that are
1587 only ever tested independently outside of that function. These are now in
1588 separate 32-bit words, so there is plenty of room for expansion again.
1589 There is no change in the user interface or the logging behaviour.
1591 (b) When building, for example, log lines, the code previously used a
1592 macro that called string_cat() twice, in order to add two strings. This is
1593 not really sufficiently general. Furthermore, there was one instance where
1594 it was actually wrong because one of the argument was used twice, and in
1595 one call a function was used. (As it happened, calling the function twice
1596 did not affect the overall behaviour.) The macro has been replaced by a
1597 function that can join an arbitrary number of extra strings onto a growing
1600 (c) The code for expansion conditions now uses a table and a binary chop
1601 instead of a serial search (which was left over from when there were very
1602 few conditions). Also, it now recognizes conditions like "pam" even when
1603 the relevant support is not compiled in: a suitably worded error message is
1604 given if an attempt is made to use such a condition.
1606 11. Added ${time_interval:xxxxx}.
1608 12. A bug was causing one of the ddress fields not to be passed back correctly
1609 from remote delivery subprocesses. The field in question was not being
1610 subsequently used, so this caused to problems in practice.
1612 13. Added new log selectors queue_time and deliver_time.
1614 14. Might have fixed a bug in maildirsizefile handling that threw up
1615 "unexpected character" debug warnings, and recalculated the data
1616 unnecessarily. In any case, I expanded the warning message to give more
1619 15. Added the message "Restricted characters in address" to the statements in
1620 the default ACL that block characters like @ and % in local parts.
1622 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
1623 Three changes have been made:
1625 (a) There was a serious bug; a negative response to MAIL caused the whole
1626 recipient domain to be cached as invalid, thereby blocking all messages
1627 to all local parts at the same domain, from all senders. This bug has
1628 been fixed. The domain is no longer cached after a negative response to
1629 MAIL if the sender used is not empty.
1631 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
1634 (c) A new callout option, "use_sender" has been added for people who want
1635 the modified behaviour.
1641 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
1644 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
1645 indeed breaks things for older releases.
1647 3. Added additional logging to the case where there is a problem reading data
1648 from a filter that is running in a subprocess using a pipe, in order to
1649 try to track down a specific problem.
1651 4. Testing facility fudge: when running in the test harness and attempting
1652 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
1653 getting "No route to host". Convert this to a timeout.
1655 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
1658 6. Some OS don't have socklen_t but use size_t instead. This affects the
1659 fifth argument of getsockopt() amongst other things. This is now
1660 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
1661 can be set for individual OS. I have set it for SunOS5, OSF1, and
1662 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
1663 some earlier ones do not.
1665 7. Change 4.30/15 was not doing the test caselessly.
1667 8. The standard form for an IPv6 address literal was being rejected by address
1668 parsing in, for example, MAIL and RCPT commands. An example of this kind of
1669 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1670 this, as well as the form without the "IPv6" on the front (but only when
1671 address literals are enabled, of course).
1673 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
1675 10. Exim crashed if a message with an empty sender address specified by -f
1676 encountered a router with an errors_to setting. This could be provoked only
1677 by a command such as
1681 where an empty string was supplied; "<>" did not hit this bug.
1683 11. Installed PCRE release 4.5.
1685 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
1686 remained set. It is now erased.
1688 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
1689 times from message ids (which are base 36 rather than the normal 62).
1691 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
1692 were being counted as actual protocol errors, and logged if the log
1693 selector +smtp_protocol_error was set. One cannot be perfect in this test,
1694 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
1695 and DATA following a set of rejected RCPTs do not count as protocol errors.
1696 In other words, Exim assumes they were pipelined, though this may not
1697 actually be the case. Of course, in all cases the client gets an
1698 appropriate error code.
1700 15. If a lookup fails in an ACL condition, a message about the failure may
1701 be available; it is used if testing the ACL cannot continue, because most
1702 such messages specify what the cause of the deferral is. However, some
1703 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
1704 that caused an old message to be retained and used if a later statement
1705 caused a defer, replacing the real cause of the deferral.
1707 16. If an IP address had so many PTR records that the DNS lookup buffer
1708 was not large enough to hold them, Exim could crash while trying to process
1709 the truncated data. It now detects and logs this case.
1711 17. Further to 4.21/58, another change has been made: if (and only if) the
1712 first line of a message (the first header line) ends with CRLF, a bare LF
1713 in a subsequent header line has a space inserted after it, so as not to
1714 terminate the header.
1716 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
1717 unnecessarily, used atoi() instead of strtol(), and didn't check the
1718 termination when getting file sizes from file names by regex.
1720 19. Completely re-implemented the support for maildirsize files, in the light
1721 of a number of problems with the previous contributed implementation
1722 (4.30/29). In particular:
1724 . If the quota is zero, the maildirsize file is maintained, but no quota is
1727 . If the maildir directory does not exist, it is created before any attempt
1728 to write a maildirsize file.
1730 . The quota value in the file is just a cache; if the quota is changed in
1731 the transport, the new value overrides.
1733 . A regular expression is available for excluding directories from the
1736 20. The autoreply transport checks the characters in options that define the
1737 message's headers; it allows continued headers, but it was checking with
1738 isspace() after an embedded newline instead of explicitly looking for a
1741 21. If all the "regular" hosts to which an address was routed had passed their
1742 expiry times, and had not reached their retry times, the address was
1743 bounced, even if fallback hosts were defined. Now Exim should go on to try
1746 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
1747 equivalent code in the SMTP transport. Some hosts send humungous responses
1748 to HELO/EHLO, more than 1024 it seems.
1750 23. Refactoring: code in filter.c used (void *) for "any old type" but this
1751 gives compiler warnings in some environments. I've now done it "properly",
1754 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
1755 (because of problems with the built-in one) was declared to return uschar *
1756 instead of char *, causing compiler failure.
1758 25. Fixed a file descriptor leak when processing alias/forward files.
1760 26. Fixed a minor format string issue in dbfn.c.
1762 27. Typo in exim.c: ("dmbnz" for "dbmnz").
1764 28. If a filter file refered to $h_xxx or $message_headers, and the headers
1765 contained RFC 2047 "words", Exim's memory could, under certain conditions,
1768 29. When a sender address is verified, it is cached, to save repeating the test
1769 when there is more than one recipient in a message. However, when the
1770 verification involves a callout, it is possible for different callout
1771 options to be set for different recipients. It is too complicated to keep
1772 track of this in the cache, so now Exim always runs a verification when a
1773 callout is required, relying on the callout cache for the optimization.
1774 The overhead is duplication of the address routing, but this should not be
1777 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
1778 to be verified with callout=postmaster, and the main callout worked but the
1779 postmaster check failed, the verification correctly failed. However, if a
1780 subsequent RCPT command asked for sender verification *without* the
1781 postmaster check, incorrect caching caused this verification also to fail,
1784 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
1785 it was not caching the DNS options (qualify_single, search_parents) that
1786 were used when the lookup failed. A subsequent lookup with different
1787 options therefore always gave the same answer, though there were cases
1788 where it should not have. (Example: a "domains = !$mx_any" option on a
1789 dnslookup router: the "domains" option is always processed without any
1790 widening, but the router might have qualify_single set.) Now Exim uses the
1791 cached value only when the same options are set.
1793 32. Added John Jetmore's "exipick" utility to the distribution.
1795 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
1796 than a timeout (e.g. a certificate is required, and is not provided), an
1797 Exim server now closes the connection immediately. Previously it waited for
1798 the client to close - but if the client is SSL, it seems that they each
1799 wait for each other, leading to a delay before one of them times out.
1801 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
1802 maintained 0.8.x compatibility because I don't think many are using it, and
1803 it is clearly obsolete.
1805 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
1808 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
1809 client certificate was expired. A simple patch fixes this, though I don't
1810 understand the full logic of why the verify callback is called multiple
1813 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
1814 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
1815 which causes problems with some clients (such as the Certicom SSL Plus
1816 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
1817 disables the coutermeasure allowing Eudora to connect."
1819 38. Exim was not checking that a write() to a log file succeeded. This could
1820 lead to Bad Things if a log got too big, in particular if it hit a file
1821 size limit. Exim now panics and dies if it cannot write to a log file, just
1822 as it does if it cannot open a log file.
1824 39. Modified OS/Makefile-Linux so that it now contains
1826 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
1828 The two -D definitions ensure that Exim is compiled with large file
1829 support, which makes it possible to handle log files that are bigger than
1832 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
1833 instance) a domain was checked against a named list that involved a lookup,
1834 causing $domain_data to be set, then another domain was checked against the
1835 same list, then the first domain was re-checked, the value of $domain_data
1836 after the final check could be wrong. In particular, if the second check
1837 failed, it could be set empty. This bug probably also applied to
1840 41. The strip_trailing_dot option was not being applied to the address given
1841 with the -f command-line option.
1843 42. The code for reading a message's header from the spool was incrementing
1844 $received_count, but never initializing it. This meant that the value was
1845 incorrect (doubled) while delivering a message in the same process in which
1846 it was received. In the most common configuration of Exim, this never
1847 happens - a fresh exec is done - but it can happen when
1848 deliver_drop_privilege is set.
1850 43. When Exim logs an SMTP synchronization error - client data sent too soon -
1851 it now includes up to 150 characters of the unexpected data in the log
1854 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
1855 and building data strings. The size of both of these buffers was 10 000
1856 bytes - far larger than anybody would *ever* want, thought I. Needless to
1857 say, somebody hit the limit. I have increased the maximum line length to
1858 20 000 and the maximum data length of concatenated lines to 100 000. I have
1859 also fixed two bugs, because there was no checking on these buffers. Tsk,
1860 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
1861 buffer is too small.
1863 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
1864 lsearch lookups. Now it does.
1866 46. When parsing a route_list item in a manualroute router, a fixed-length
1867 buffer was used for the list of hosts. I made this 1024 bytes long,
1868 thinking that nobody would ever have a list of hosts that long. Wrong.
1869 Somebody had a whole pile of complicated expansion conditions, and the
1870 string was silently truncated, leading to an expansion error. It turns out
1871 that it is easier to change to an unlimited length (owing to other changes
1872 that have happened since this code was originally written) than to build
1873 structure for giving a limitation error. The length of the item that
1874 expands into the list of hosts is now unlimited.
1876 47. The lsearch lookup could not handle data where the length of text line was
1877 more than 4095 characters. Such lines were truncated, leading to shortened
1878 data being returned. It should now handle lines of any length.
1880 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
1881 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
1884 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
1885 replacing strings such as BIN_DIRECTORY in a source file: the replacement
1886 no longer happens in comment lines. A list of replacements is now placed
1887 at the head of all of the source files, except those whose only change is
1888 to replace PERL_COMMAND in the very first #! line.
1890 50. Replaced the slow insertion sort in queue.c, for sorting the list of
1891 messages on the queue, with a bottom-up merge sort, using code contributed
1892 by Michael Haardt. This should make operations like -bp somewhat faster on
1893 large queues. It won't affect queue runners, except when queue_run_in_order
1896 51. Installed eximstats 1.31 in the distribution.
1898 52. Added support for SRV lookups to the dnslookup router.
1900 53. If an ACL referred to $message_body or $message_body_end, the value was not
1901 reset for any messages that followed in the same SMTP session.
1903 54. The store-handling optimization for building very long strings was not
1904 differentiating between the different store pools. I don't think this
1905 actually made any difference in practice, but I've tidied it.
1907 55. While running the routers to verify a sender address, $sender_address
1908 was still set to the sender address. This is wrong, because when routing to
1909 send a bounce to the sender, it would be empty. Therefore, I have changed
1910 it so that, while verifying a sender address, $sender_address is set to <>.
1911 (There is no change to what happens when verifying a recipient address.)
1913 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
1914 target A or AAAA records (if not already returned) without resetting the
1915 qualify_single or search_parents options of the DNS resolver. These are
1916 inappropriate in this case because the targets of MX and SRV records must
1917 be FQDNs. A broken DNS record could cause trouble if it happened to have a
1918 target that, when qualified, matched something in the local domain. These
1919 two options are now turned off when doing these lookups.
1921 57. It seems that at least some releases of Reiserfs (which does not have the
1922 concept of a fixed number of inodes) returns zero and not -1 for the
1923 number of available inodes. This interacted badly with check_spool_inodes,
1924 which assumed that -1 was the "no such thing" setting. What I have done is
1925 to check that the total number of inodes is greater than zero before doing
1926 the test of how many are available.
1928 58. When a "warn" ACL statement has a log_message modifier, the message is
1929 remembered, and not repeated. This is to avoid a lot of repetition when a
1930 message has many recipients that cause the same warning to be written.
1931 Howewer, Exim was preserving the list of already written lines for an
1932 entire SMTP session, which doesn't seem right. The memory is now reset if a
1933 new message is started.
1935 59. The "rewrite" debugging flag was not showing the result of rewriting in the
1936 debugging output unless log_rewrite was also set.
1938 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
1939 of (int)(handle) when we know that handle contains (void *)(-1).
1941 61. The Exim daemon panic-logs an error return when it closes the incoming
1942 connection. However "connection reset by peer" seems to be common, and
1943 isn't really an error worthy of noting specially, so that particular error
1946 62. When Exim is trying to find all the local interfaces, it used to panic and
1947 die if the ioctl to get the interface flags failed. However, it seems that
1948 on at least one OS (Solaris 9) it is possible to have an interface that is
1949 included in the list of interfaces, but for which you get a failure error
1950 for this call. This happens when the interface is not "plumbed" into a
1951 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
1952 failure of the "get flags" call assumes that the interface is down.
1954 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
1955 makes life easier for people who are doing arithmetic on fields extracted
1956 from dates, where you often get leading zeros that should not be
1957 interpreted as octal.
1959 64. Added qualify_domain to the redirect router, to override the global
1962 65. If a pathologically long header line contained very many addresses (the
1963 report of this problem mentioned 10 000) and each of them was rewritten,
1964 Exim could use up a very large amount of memory. (It kept on making new
1965 copies of the header line as it rewrote, and never released the old ones.)
1966 At the expense of a bit more processing, the header rewriting function has
1967 been changed so that it no longer eats memory in this way.
1969 66. The generation of the Received: header has been moved from the time that a
1970 message starts to be received, to the time that it finishes. The timestamp
1971 in the Received: header should now be very close to that of the <= log
1972 line. There are two side-effects of this change:
1974 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
1975 logged header lines no longer include the local Received: line, because
1976 it has not yet been created. The same applies to a copy of the message
1977 that is returned to a non-SMTP sender when a message is rejected.
1979 (b) When a filter file is tested using -bf, no additional Received: header
1980 is added to the test message. After some thought, I decided that this
1983 This change does not affect the value of $received_for. It is still set
1984 after address rewriting, but before local_scan() is called.
1986 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
1988 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
1989 gave an unhelpful panic error message, and a defer error. I have managed to
1990 change this behaviour so that it now rejects any supplied certificate,
1991 which seems right, as the list of acceptable certificates is empty.
1993 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
1994 gave an unhelpful defer error. I have not managed to make this reject any
1995 supplied certificates, but the error message it gives is "no certificate
1996 supplied", which is not helpful.
1998 70. exigrep's output now also includes lines that are not associated with any
1999 message, but which match the given pattern. Implemented by a patch from
2000 Martin Sluka, which also tidied up the Perl a bit.
2002 71. Recipient callout verification, like sender verification, was using <> in
2003 the MAIL FROM command. This isn't really the right thing, since the actual
2004 sender may affect whether the remote host accepts the recipient or not. I
2005 have changed it to use the actual sender in the callout; this means that
2006 the cache record is now keyed on a recipient/sender pair, not just the
2007 recipient address. There doesn't seem to be a real danger of callout loops,
2008 since a callout by the remote host to check the sender would use <>.
2009 [SEE ABOVE: changed after hitting problems.]
2011 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2012 temporary errors. However, in the case of such a code being given after
2013 the end of a data transmission (i.e. after ".") Exim was failing to write
2014 a retry record for the message. (Yes, there was some broken host that was
2015 actually sending 8xx at this point.)
2017 73. An unknown lookup type in a host list could cause Exim to panic-die when
2018 the list was checked. (An example that provoked this was putting <; in the
2019 middle of a list instead of at the start.) If this happened during a DATA
2020 ACL check, a -D file could be left lying around. This kind of configuration
2021 error no longer causes Exim to die; instead it causes a defer errror. The
2022 incident is still logged to the main and panic logs.
2024 74. Buglet left over from Exim 3 conversion. The message "too many messages
2025 in one connection" was written to the rejectlog but not the mainlog, except
2026 when address rewriting (yes!) was being logged.
2028 75. Added write_rejectlog option.
2030 76. When a system filter was run not as root (that is, when system_filter_user
2031 was set), the values of the $n variables were not being returned to the
2032 main process; thus, they were not subsequently available in the $sn
2035 77. Added +return_path_on_delivery log selector.
2037 78. A connection timeout was being treated differently from recipients deferred
2038 when testing hosts_max_try with a message that was older than the host's
2039 retry timeout. (The host should not be counted, thus allowing all hosts to
2040 be tried at least once before bouncing.) This may have been the cause of an
2041 occasionally reported bug whereby a message would remain on the queue
2042 longer than the retry timeout, but would be bounced if a delivery was
2043 forced. I say "may" because I never totally pinned down the problem;
2044 setting up timeout/retry tests is difficult. See also the next item.
2046 79. The ultimate address timeout was not being applied to errors that involved
2047 a combination of host plus message (for example, a timeout on a MAIL
2048 command). When an address resolved to a number of possible hosts, and they
2049 were not all tried for each delivery (e.g. because of hosts_max_try), a
2050 message could remain on the queue longer than the retry timeout.
2052 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2055 81. Fixed an obscure SMTP outgoing bug which required at least the following
2056 conditions: (a) there was another message waiting for the same server;
2057 (b) the server returned 5xx to all RCPT commands in the first message so
2058 that the message was not completed; (c) the server dropped the connection
2059 or gave a negative response to the RSET that Exim sends to abort the
2060 transaction. The observed case was a dropped connection after DATA that had
2061 been sent in pipelining mode. That is, the server had advertised PIPELINING
2062 but was not implementing it correctly. The effect of the bug was incorrect
2063 behaviour, such as trying another host, and this could lead to a crash.
2069 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2070 and daemon.c were passed as pointers to ints; they should have been
2071 pointers to socklen_t variables (which are typically unsigned ints).
2073 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2076 3. Fixed a really odd bug that affected only the testing scheme; patching a
2077 certain fixed string in the binary changed the value of another string that
2078 happened to be identical to the end of the original first string.
2080 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2081 name", it returns that address as the IP address. On some operating
2082 systems (e.g. Solaris), it also passes back the IP address string as the
2083 "host name". However, on others (e.g. Linux), it passes back an empty
2084 string. Exim wasn't checking for this, and was changing the host name to an
2085 empty string, assuming it had been canonicized.
2087 5. Although rare, it is permitted to have more than one PTR record for a given
2088 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2089 all the names associated with an address, because they do in Solaris.
2090 However, it seems that they do not in Linux for data that comes from the
2091 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2092 I found this out when I moved to a new Linux workstation and tried to run
2093 the Exim test suite.
2095 To get round this problem I have changed the code so that it now does its
2096 own call to the DNS to look up PTR records when searching for a host name.
2097 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2098 addresses that are only in /etc/hosts are still found.
2100 This behaviour is, however, controlled by an option called host_lookup_
2101 order, which defaults to "bydns:byaddr". If people want to use the other
2102 order, or indeed, just use one or the other means of lookup, they can
2103 specify it in this variable.
2105 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2106 some operating systems, this comes back from gethostbyaddr() as an empty
2107 string, and this is what Exim used to test for. However, it seems that in
2108 other systems, "." is yielded. Exim now tests for this case too.
2110 7. The values of check_spool_space and check_log_space are now held internally
2111 as a number of kilobytes instead of an absolute number of bytes. If a
2112 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2113 kilobyte. This means that much larger values can be stored.
2115 8. Exim monitor: an attempt to get the action menu when not actually pointing
2116 at a message produces an empty menu entitled "No message selected". This
2117 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2118 no entries in it ("Shell widget menu has zero width and/or height"). So I
2119 have added a single, blank menu entry in this case.
2121 9. Added ${quote_local_part.
2123 10. MIME decoding is now applied to the contents of Subject: header lines when
2126 11. Now that a reference to $sender_host_address automatically causes a reverse
2127 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2128 host lookup before query-style lookups in lists that might use this
2129 variable. This has therefore been abolished, and the "net-" prefix is no
2130 longer necessary for query-style lookups.
2132 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2133 have been a typo for LFLAGS, so it has been changed.
2135 13. The install script calls Exim with "-C /dev/null" in order to find the
2136 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2137 to be output. Howeve, since Exim outputs its version number before the
2138 error, it didn't break the script. It just looked ugly. I fixed this by
2139 always allowing "-C /dev/null" if the caller is root.
2141 14. Ignore overlarge ACL variable number when reading spool file - insurance
2142 against a later release with more variables having written the file.
2144 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2145 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2146 this, as well as the form without the "IPv6" on the front.
2148 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2149 OS/Makefile-Darwin file.
2151 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2152 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2154 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2155 variable. However, some OS use ints for inodes. I've added cast to long int
2156 to get rid of the compiler warning.
2158 19. I had forgotten to lock out "/../" in configuration file names when
2159 ALT_CONFIG_PREFIX was set.
2161 20. Routers used for verification do not need to specify transports. However,
2162 if such a router generated a host list, and callout was configured, Exim
2163 crashed, because it could not find a port number from the (non-existent)
2164 transport. It now assumes port 25 in this circumstance.
2166 21. Added the -t option to exigrep.
2168 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2169 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2170 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2171 them is set, LOOKUP_LSEARCH is forced.
2173 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2175 24. Added sender and host information to the "rejected by local_scan()" log
2176 line; previously there was no indication of these.
2178 25. Added .include_if_exists.
2180 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2181 Linux. It turns out that not all file systems support this. Apparently some
2182 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2183 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2184 not supported on the file descriptor, is now ignored when Exim is trying to
2185 sync a directory. This applies only to Linux.
2187 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2189 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2190 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2193 29. Added support for maildirsize files from supplied patch (modified a bit).
2195 30. The use of :fail: followed by an empty string could lead Exim to respond to
2196 sender verification failures with (e.g.):
2198 550 Verification failed for <xxx>
2199 550 Sender verify failed
2201 where the first response line was missing the '-' that indicates it is not
2202 the final line of the response.
2204 31. The loop for finding the name of the user that called Exim had a hardwired
2205 limit of 10; it now uses the value of finduser_retries, which is used for
2206 all other user lookups.
2208 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2211 33. Exim was neglecting to zero errno before one call of strtol() when
2212 expanding a string and expecting an integer value. On some systems this
2213 resulted in spurious "integer overflow" errors. Also, it was casting the
2214 result into an int without checking.
2216 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2217 did not work. The code looks as if it has *never* worked, though it appears
2218 to have been documented since at least releast 1.62. I have made it work.
2220 35. The "timeout_DNS" error in retry rules, also documented since at least
2221 1.62, also never worked. As it isn't clear exactly what this means, and
2222 clearly it isn't a major issue, I have abolished the feature by treating it
2223 as "timeout", and writing a warning to the main and panic logs.
2225 36. The display of retry rules for -brt wasn't always showing the error code
2228 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2229 timeout_connect_A, timeout_connect_MX.
2231 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2232 to the empty sender.
2234 39. The daemon was not analysing the content of -oX till after it had closed
2235 stderr and disconnected from the controlling terminal. This meant that any
2236 syntax errors were only noted on the panic log, and the return code from
2237 the command was 0. By re-arranging the code a little, I've made the
2238 decoding happen first, so such errors now appear on stderr, and the return
2239 code is 1. However, the actual setting up of the sockets still happens in
2240 the disconnected process, so errors there are still only recorded on the
2243 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2244 connections (as happens on some IP stacks) was logged at start up time as
2245 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2246 it from "IPv6 and IPv4", which means that two separate sockets are being
2249 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2250 says whether AF_INET or AF_INET6 was passed as an argument.
2252 42. Exiwhat output was messed up when time zones were included in log
2255 43. Exiwhat now gives more information about the daemon's listening ports,
2256 and whether -tls-on-connect was used.
2258 44. The "port" option of the smtp transport is now expanded.
2260 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2261 silently ignored. Now an error message is written to the main and panic
2264 46. There's a new ACL modifier called "logwrite" which writes to a log file
2265 as soon as it is encountered.
2267 47. Added $local_user_uid and $local_user_gid at routing time.
2269 48. Exim crashed when trying to verify a sender address that was being
2272 49. Exim was recognizing only a space character after ".include". It now also
2273 recognizes a tab character.
2275 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2276 extracting the relevant information from the specification. The man page no
2277 longer contains scrambled data for the -d option, and I've added a section
2278 at the front about calling Exim under different names.
2280 51. Added "extra_headers" argument to the "mail" command in filter files.
2282 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2285 53. Installed eximstats 1.29.
2287 54. Added transport_filter_timeout as a generic transport option.
2289 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2290 Cc: header lines. This was required by RFC 822, but it not required by RFC
2293 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2294 incoming messages that did not have them. Now it does so only if the
2295 message originates locally, that is, if there is no associated remote host
2296 address. When Resent- header lines are present, this applies to the Resent-
2297 lines rather than the non-Resent- lines.
2299 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2300 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2302 58. Messages for configuration errors now include the name of the main
2303 configuration file - useful now that there may be more than one file in a
2304 list (.included file names were always shown).
2306 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2307 for those rare installations that do not start the daemon as root or run it
2308 setuid root. I've cut out the call to initgroups() if the daemon is not
2311 60. The Exim user and group can now be bound into the binary as text strings
2312 that are looked up at the start of Exim's processing.
2314 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2316 62. Added $mailstore_basename variable.
2318 63. Installed patch to sieve.c from Michael Haardt.
2320 64. When Exim failed to open the panic log after failing to open the main log,
2321 the original message it was trying to log was written to stderr and debug
2322 output, but if they were not available (the usual case in production), it
2323 was lost. Now it is written to syslog before the two lines that record the
2324 failures to open the logs.
2326 65. Users' Exim filters run in subprocesses under the user's uid. It is
2327 possible for a "deliver" command or an alias in a "personal" command to
2328 provoke an address rewrite. If logging of address rewriting is configured,
2329 this fails because the process is not running as root or exim. There may be
2330 a better way of dealing with this, but for the moment (because 4.30 needs
2331 to be released), I have disabled address rewrite logging when running a
2332 filter in a non-root, non-exim process.
2338 1. The buildconfig auxiliary program wasn't quoting the value set for
2339 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2340 not defined. This bug was masked in 4.22 by the effect that was fixed in
2343 2. Some messages that were rejected after a message id was allocated were
2344 shown as "incomplete" by exigrep. It no longer does this for messages that
2345 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2347 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2348 have allow_domain_literals set, the ID did not get logged in the <= line.
2349 Domain literals are now always recognized in Message-ID: header lines.
2351 4. The first argument for a ${extract expansion item is the key name or field
2352 number. Leading and trailing spaces in this item were not being ignored,
2353 causing some misleading effects.
2355 5. When deliver_drop_privilege was set, single queue runner processes started
2356 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2357 same command in the process it spins off) were not dropping privilege.
2359 6. When the daemon running as "exim" started a queue runner, it always
2360 re-executed Exim in the spun-off process. This is a waste of effort when
2361 deliver_drop_privilege is set. The new process now just calls the
2362 queue-runner function directly.
2368 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2371 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2372 ignored. Though the use of -oP was forcing the writing of a pid file, it
2373 was always written to the default place.
2375 3. If the message "no IP address found for host xxxx" is generated during
2376 incoming verification, it is now followed by identification of the incoming
2377 connection (so you can more easily find what provoked it).
2379 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2381 5. Added some features to "harden" Exim a bit more against certain attacks:
2383 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2384 be put in Local/Makefile. This is like the never_users runtime option,
2385 but it cannot be overridden. The default setting is "root".
2387 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2388 prefix string with which any file named in a -C command line option
2391 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2392 is retained for -C and -D only if the caller of Exim is root. Without
2393 it, the exim user may also use -C and -D and retain privilege.
2395 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2396 command line option is disabled.
2398 6. Macro names set by the -D option must start with an upper case letter, just
2399 like macro names defined in the configuration file.
2401 7. Added "dereference=" facility to LDAP.
2403 8. Two instances of the typo "uknown" in the source files are fixed.
2405 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2406 the Configure-Makefile script screwed up while processing it.
2408 10. Incorporated PCRE 4.4.
2410 11. The SMTP synchronization check was not operating right at the start of an
2411 SMTP session. For example, it could not catch a HELO sent before the client
2412 waited for the greeting. There is now a check for outstanding input at the
2413 point when the greeting is written. Because of the duplex, asynchronous
2414 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2415 way, but not yet received, when the check is performed.
2417 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2418 on TCP/IP sockets, because this apparently causes some broken clients to
2421 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2422 unchanged) from the Cygwin maintainer.
2424 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2425 was supported instead of testing for mbx. Effectively a typo.
2427 15. The spa authenticator server code was not checking that the input it
2428 received was valid base64.
2430 16. The debug output line for the "set" modifier in ACLs was not showing the
2431 name of the variable that was being set.
2433 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2435 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2436 lookup on its own. Something else was needed to trigger the lookup. For
2437 example, a match in host_lookup or the need for a host name in a host list.
2438 Now, if $sender_host_name is referenced and the host name has not yet been
2439 looked up, a lookup is performed. If the lookup fails, the variable remains
2440 empty, and $host_lookup_failed is set to "1".
2442 19. Added "eqi" as a case-independent comparison operator.
2444 20. The saslauthd authentication condition could segfault if neither service
2445 nor realm was specified.
2447 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2448 error message that was logged was misleading, and incoming SMTP
2449 connections were dropped. The message is now more accurate, and temporary
2450 errors are given to SMTP connections.
2452 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2453 (except RSET) with a 421 error, until QUIT is received. However, it was
2454 failing to send a response to QUIT.
2456 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2457 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2458 ACL is now run after the helo_try_verify_hosts code.
2460 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2461 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2462 case-independent, so other case variants are also recognized). Apparently
2463 some systems use these upper case variants.
2465 25. If more than two messages were waiting for the same host, and a transport
2466 filter was specified for the transport, Exim sent two messages over the
2467 same TCP/IP connection, and then failed with "socket operation on non-
2468 socket" when it tried to send the third.
2470 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2472 27. The extern definition of crypt16() in expand.c was not being excluded when
2473 the OS had its own crypt16() function.
2475 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2476 as a preferred synonym for return_size_limit, both as an option and as an
2479 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2481 30. Changed the default configuration ACL to relax the local part checking rule
2482 for addresses that are not in any local domains. For these addresses,
2483 slashes and pipe symbols are allowed within local parts, but the sequence
2484 /../ is explicitly forbidden.
2486 31. SPA server authentication was not clearing the challenge buffer before
2489 32. log_message in a "warn" ACL statement was writing to the reject log as
2490 well as to the main log, which contradicts the documentation and doesn't
2491 seem right (because no rejection is happening). So I have stopped it.
2493 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2494 However, I am unable to do any testing of this.
2496 34. Fixed an infelicity in the appendfile transport. When checking directories
2497 for a mailbox, to see if any needed to be created, it was accidentally
2498 using path names with one or more superfluous leading slashes; tracing
2499 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2501 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2502 contents are added to the log line that is written for every discarded
2503 recipient. (Previously a log_message setting was ignored.)
2505 36. The ${quote: operator now quotes the string if it is empty.
2507 37. The install script runs exim in order to find its version number. If for
2508 some reason other than non-existence or emptiness, which it checks, it
2509 could not run './exim', it was installing it with an empty version number,
2510 i.e. as "exim-". This error state is now caught, and the installation is
2513 38. An argument was missing from the function that creates an error message
2514 when Exim fails to connect to the socket for saslauthd authentication.
2515 This could cause Exim to crash, or give a corrupted message.
2517 39. Added isip, isip4, and isip6 to ${if conditions.
2519 40. The ACL variables $acl_xx are now saved with the message, and can be
2520 accessed later in routers, transports, and filters.
2522 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2523 strings in the file are not string-expanded.
2525 42. If a MAIL command specified a SIZE value that was too large to fit into an
2526 int variable, the check against message_size_limit failed. Such values are
2527 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2528 day this will have to be increased, but I don't think I want to be around
2529 when emails are that large.
2536 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2537 iconv() is not standard in FreeBSD.
2539 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2540 IPv6 enabled. The observed symptom was a segmentation fault on return from
2541 the function os_common_find_running_interfaces() in src/os.c.
2543 3. In the check_special_case() function in daemon.c I had used "errno" as an
2544 argument name, which causes warnings on some systems. This was basically a
2545 typo, since it was named "eno" in the comments!
2547 4. The code that waits for the clock to tick (at a resolution of some fraction
2548 of a second) so as to ensure message-id uniqueness was always waiting for
2549 at least one whole tick, when it could have waited for less. [This is
2550 almost certainly not relevant at current processor speeds, where it is
2551 unlikely to ever wait at all. But we try to future-proof.]
2553 5. The function that sleeps for a time interval that includes fractions of a
2554 second contained a race. It did not block SIGALRM between setting the
2555 timer, and suspending (a couple of lines later). If the interval was short
2556 and the sigsuspend() was delayed until after it had expired, the suspension
2557 never ended. On busy systems this could lead to processes getting stuck for
2560 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2561 process, before it forks any delivery processes. The open lookup caching
2562 mechanism meant that the open file or database connection was passed into
2563 the delivery process. The problem was that delivery processes always tidy
2564 up cached lookup data. This could cause a problem for the next delivery
2565 process started by the queue runner, because the external queue runner
2566 process does not know about the closure. So the next delivery process
2567 still has data in the lookup cache. In the case of a file lookup, there was
2568 no problem because closing a file descriptor in a subprocess doesn't affect
2569 the parent. However, if the lookup was caching a connection to a database,
2570 the connection was closed, and the second delivery process was likely to
2571 see errors such as "PGSQL: query failed: server closed the connection
2572 unexpectedly". The problem has been fixed by closing all cached lookups
2573 in a queue runner before running a delivery process.
2575 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2576 seem to have the "const" qualifier which it has on other OS. I've
2579 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2580 *of the same type* (client or server) with the same public name that an
2581 error should be diagnosed.
2583 9. When Exim looked up a host name for an IP address, but failed to find the
2584 original IP address when looking up the host name (a safety check), it
2585 output the message "<ip address> does not match any IP for NULL", which was
2586 confusing, to say the least. The bug was that the host name should have
2587 appeared instead of "NULL".
2589 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2590 user that is built into the binary, and the -C or -D options is used, root
2591 privilege is dropped before the configuration file is read. In addition,
2592 logging is switched to stderr instead of the normal log files. If the
2593 configuration then re-defines the Exim user, the unprivileged environment
2594 is probably not what is expected, so Exim logs a panic warning message (but
2597 However, if deliver_drop_privilege is set, the unprivileged state may well
2598 be exactly what is intended, so the warning has been cut out in that case,
2599 and Exim is allowed to try to write to its normal log files.
2605 1. smtp_return_error_details was not giving details for temporary sender
2606 or receiver verification errors.
2608 2. Diagnose a configuration error if two authenticators have the same public
2611 3. Exim used not to create the message log file for a message until the first
2612 delivery attempt. This could be confusing when incoming messages were held
2613 for policy or load reasons. The message log file is now created at the time
2614 the message is received, and an initial "Received" line is written to it.
2616 4. The automatically generated man page for command line options had a minor
2617 bug that caused no ill effects; however, a more serious problem was that
2618 the procedure for building the man page automatically didn't always
2619 operate. Consequently, release 4.20 contains an out-of-date version. This
2620 shouldn't happen again.
2622 5. When building Exim with embedded Perl support, the script that builds the
2623 Makefile was calling 'perl' to find its compile-time parameters, ignoring
2624 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
2626 6. The freeze_tell option was not being used for messages that were frozen on
2627 arrival, either by an ACL or by local_scan().
2629 7. Added the smtp_incomplete_transaction log selector.
2631 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
2632 was accepting AUTH without a new EHLO.
2634 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
2635 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
2638 10. Logging of TCP/IP connections (when configured) now happens in the main
2639 daemon process instead of the child process, so that the TCP/IP connection
2640 count is more accurate (but it can never be perfect).
2642 11. The use of "drop" in a nested ACL was not being handled correctly in the
2643 outer ACL. Now, if condition failure induced by the nested "drop" causes
2644 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
2645 or "require"), the connection is dropped.
2647 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
2648 that yield "discard" can now be used with an "accept" or a "discard" verb,
2649 but an error is generated for any others (because I can't see a useful way
2650 to define what should happen).
2652 13. When an ACL is read dynamically from a file (or anywhere else), the lines
2653 are now processed in the same way as lines in the Exim configuration file.
2654 In particular, continuation lines are supported.
2656 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
2658 15. Added -ti meaning -t -i.
2660 16. Check for letters, digits, hyphens, and dots in the names of dnslist
2661 domains, and warn by logging if others are found.
2663 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
2664 returned from GIFCONF when Exim is trying to find the list of interfaces on
2665 a host. The code in os.c has been modified to copy each ifreq to an aligned
2666 structure in all cases.
2668 Also, in some cases, the returned ifreq's were being copied to a 'struct
2669 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
2670 means the last couple of bytes of an IPv6 address could be chopped if the
2671 ifreq contained only a normal sockaddr (14 bytes storage).
2673 18. Named domain lists were not supported in the hosts_treat_as_local option.
2674 An entry such as +xxxx was not recognized, and was treated as a literal
2677 19. Ensure that header lines added by a DATA ACL are included in the reject log
2678 if the ACL subsequently rejects the message.
2680 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
2681 MD5 (which is deprecated).
2683 21. When testing a filter file using -bf, Exim was writing a message when it
2684 took the sender from a "From " line in the message, but it was not doing so
2685 when it took $return_path from a Return-Path: header line. It now does.
2687 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
2688 with a valid header line field name (a series of printing characters
2689 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
2691 23. Changed "disc" in the source to "disk" to conform to the documentation and
2692 the book and for uniformity.
2694 24. Ignore Sendmail's -Ooption=value command line item.
2696 25. When execve() failed while trying to run a command in a pipe transport,
2697 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
2698 could be confused with a return value of 69 from the command itself. This
2699 has been changed to 127, the value the shell returns if it is asked to run
2700 a non-existent command. The wording for the related log line suggests a
2701 non-existent command as the problem.
2703 26. If received_header_text expands to an empty string, do not add a Received:
2704 header line to the message. (Well, it adds a token one on the spool, but
2705 marks it "old" so that it doesn't get used or transmitted.)
2707 27. Installed eximstats 1.28 (addition of -nt option).
2709 28. There was no check for failure on the call to getsockname() in the daemon
2710 code. This can fail if there is a shortage of resources on the system, with
2711 ENOMEM, for example. A temporary error is now given on failure.
2713 29. Contrary to the C standard, it seems that in some environments, the
2714 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
2715 program. Exim now does this explicitly; it affects the formatting of
2716 timestamps using strftime().
2718 30. If exiqsumm was given junk data, it threw up some uninitialized variable
2719 complaints. I've now initialized all the variables, to avoid this.
2721 32. Header lines added by a system filter were not being "seen" during
2722 transport-time rewrites.
2724 33. The info_callback() function passed to OpenSSL is set up with type void
2725 (*)(SSL *, int, int), as described somewhere. However, when calling the
2726 function (actually a macro) that sets it up, the type void(*)() is
2727 expected. I've put in a cast to prevent warnings from picky compilers.
2729 34. If a DNS black list lookup found a CNAME record, but there were no A
2730 records associated with the domain it pointed at, Exim crashed.
2732 35. If a DNS black list lookup returned more than one A record, Exim ignored
2733 all but the first. It now scans all returned addresses if a particular IP
2734 value is being sought. In this situation, the contents of the
2735 $dnslist_value variable are a list of all the addresses, separated by a
2738 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
2739 to accept a host name and all its aliases if the forward lookup for any of
2740 them yielded the IP address of the incoming connection. Now it accepts only
2741 those names whose forward lookup yields the correct IP address. Any other
2742 names are discarded. This closes a loophole whereby a rogue DNS
2743 administrator could create reverse DNS records to break through a
2744 wildcarded host restriction in an ACL.
2746 37. If a user filter or a system filter that ran in a subprocess used any of
2747 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
2748 the wrong values were passed to the pipe command ($thisaddress had the
2749 value of $0, $0 had the value of $1, etc). This bug was introduced by
2750 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
2752 38. Improved the line breaking for long SMTP error messages from ACLs.
2753 Previously, if there was no break point between 40 and 75 characters, Exim
2754 left the rest of the message alone. Two changes have been made: (a) I've
2755 reduced the minimum length to 35 characters; (b) if it can't find a break
2756 point between 35 and 75 characters, it looks ahead and uses the first one
2757 that it finds. This may give the occasional overlong line, but at least the
2758 remaining text gets split now.
2760 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
2761 file descriptors might be low, and that setting 1000 would always raise it.
2762 It turns out that in some environments, the limit is already over 1000 and
2763 that lowering it causes trouble. So now Exim takes care not to decrease it.
2765 40. When delivering a message, the value of $return_path is set to $sender_
2766 address at the start of routing (routers may change the value). By an
2767 oversight, this default was not being set up when an address was tested by
2768 -bt or -bv, which affected the outcome if any router or filter referred to
2771 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
2772 log only when "message" or "log_message" are set. However, if one of the
2773 conditions was an address verification, or a call to a nested ACL, the
2774 messages generated by the underlying test were being passed through. This
2775 no longer happens. The underlying message is available in $acl_verify_
2776 message for both "message" and "log_message" expansions, so it can be
2777 passed through if needed.
2779 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
2780 new expansion $bh_ to give the encoded byte string without charset
2781 translation. Translation happens only if iconv() is available; HAVE_ICONV
2782 indicates this at build time. HEADERS_CHARSET gives the charset to
2783 translate to; headers_charset can change it in the configuration, and
2784 "headers charset" can change it in an individual filter file.
2786 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
2787 that creates RFC 2047 encoded "words" labels them as that charset instead
2788 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
2789 expansion operator; (ii) when Exim creates a From: line for a local
2790 message; (iii) when a header line is rewritten to include a "phrase" part.
2792 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
2793 buggy, causing it to skip the first lines of messages whose message ID
2794 ended in 'D'. This would not have bitten before Exim release 4.14, because
2795 message IDs were unlikely to end in 'D' before then. The effect was to have
2796 incorrect size information for certain domains.
2798 45. #include "config.h" was missing at the start of the crypt16.c module. This
2799 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
2802 46. If there was a timeout during a "random" callout check, Exim treated it as
2803 a failure of the random address, and carried on sending RSET and the real
2804 address. If the delay was just some slowness somewhere, the response to the
2805 original RCPT would be taken as a response to RSET and so on, causing
2806 mayhem of various kinds.
2808 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
2809 when I implemented it. It didn't allow for the fact that some option values
2810 may legitimatetly be negative (e.g. size_addition), and it didn't even do
2811 the right test for positive values.
2813 48. Domain names in DNS records are case-independent. Exim always looks them up
2814 in lower case. Some resolvers return domain names in exactly the case they
2815 appear in the zone file, that is, they may contain uppercase letters. Not
2816 all resolvers do this - some return always lower case. Exim was treating a
2817 change of case by a resolver as a change of domain, similar to a widening
2818 of a domain abbreviation. This triggered its re-routing code and so it was
2819 trying to route what was effectively the same domain again. This normally
2820 caused routing to fail (because the router wouldn't handle the domain
2821 twice). Now Exim checks for this case specially, and just changes the
2822 casing of the domain that it ultimately uses when it transmits the message
2825 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
2828 50. If a filter generated a file delivery with a non-absolute name (possible if
2829 no home directory exists for the router), the forbid_file option was not
2832 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
2833 the existing equality matching.
2835 52. Exim was using ints instead of ino_t variables in some places where it was
2836 dealing with inode numbers.
2838 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
2839 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
2840 called TMPDIR, and if it finds it is different, it changes its value.
2842 54. The smtp_printf() function is now made available to local_scan() so
2843 additional output lines can be written before returning. There is also an
2844 smtp_fflush() function to enable the detection of a dropped connection.
2845 The variables smtp_input and smtp_batched_input are exported to
2848 55. Changed the default runtime configuration: the message "Unknown user"
2849 has been removed from the ACL, and instead placed on the localuser router,
2850 using the cannot_route_message feature. This means that any verification
2851 failures that generate their own messages won't get overridden. Similarly,
2852 the "Unrouteable address" message that was in the ACL for unverifiable
2853 relay addresses has also been removed.
2855 56. Added hosts_avoid_esmtp to the smtp transport.
2857 57. The exicyclog script was not checking for the esoteric option
2858 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
2859 will work only if exicyclog is run under the appropriate euid.
2861 58. Following a discussion on the list, the rules by which Exim recognises line
2862 endings on incoming messages have been changed. The -dropcr and drop_cr
2863 options are now no-ops, retained only for backwards compatibility. The
2864 following line terminators are recognized: LF CRLF CR. However, special
2865 processing applies to CR:
2867 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
2868 nor a local message in the state where . is a terminator.
2870 (ii) If a bare CR is encountered in a header line, an extra space is added
2871 after the line terminator so as not to end the header. The reasoning
2872 behind this is that bare CRs in header lines are most likely either
2873 to be mistakes, or people trying to play silly games.
2875 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
2876 was being incorrectly given as 18 bytes larger than it should have been.
2877 This is a VOB (very old bug).
2879 60. This may never have affected anything current, but just in case it has:
2880 When the local host is found other than at the start of a list of hosts,
2881 the local host, those with the same MX, and any that follow, are discarded.
2882 When the list in question was part of a longer list of hosts, the following
2883 hosts (not currently being processed) were also being discarded. This no
2884 longer happens. I'm not sure if this situation could ever has previously
2887 61. Added the "/MX" feature to lists of hosts in the manualroute and query
2890 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
2891 header. This is something that is recommended in a new Internet Draft, and
2892 is something that is documented as being done by Sendmail. There are two
2893 possible values. For messages generated by the autoreply transport, Exim
2896 Auto-Submitted: auto-replied
2898 whereas for all other generated messages (e.g. bounces) it adds
2900 Auto-Submitted: auto-generated
2902 63. The "personal" condition in filters now includes a test for the
2903 Auto-Submitted: header. If it contains the string "auto-" the message it
2904 not considered personal.
2906 64. Added rcpt_include_affixes as a generic transport option.
2908 65. Added queue_only_override (default true).
2910 66. Added the syslog_duplication option.
2912 67. If what should have been the first header line of a message consisted of
2913 a space followed by a colon, Exim was mis-interpreting it as a header line.
2914 It isn't of course - it is syntactically invalid and should therefore be
2915 treated as the start of the message body. The misbehaviour could have
2916 caused a number of strange effects, including loss of data in subsequent
2917 header lines, and spool format errors.
2919 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
2920 client host had authenticated. This control can now be exercised by an ACL
2921 for more flexibility.
2923 69. By default, callouts do not happen when testing with -bh. There is now a
2924 variant, -bhc, which does actually run the callout code, including
2925 consulting and updating the callout cache.
2927 70. Added support for saslauthd authentication, courtesy of Alexander
2930 71. If statvfs() failed on the spool or log directories while checking their
2931 size for availability, Exim confusingly gave the error "space shortage".
2932 Furthermore, in debugging mode it crashed with a floating point exception.
2933 These checks are done if check_{spool,log}_{space,inodes} are set, and when
2934 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
2935 serious problem, Exim now writes to the main and panic logs when this
2936 happens, with details of the failure. It then refuses to accept the
2937 incoming message, giving the message "spool directory problem" or "log
2938 directory problem" with a 421 code for SMTP messages.
2940 72. When Exim is about to re-exec itself, it ensures that the file descriptors
2941 0, 1, and 2 exist, because some OS complain for execs without them (see
2942 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
2943 descriptors. However, the code omitted to check that the open succeeded,
2944 causing mysterious errors if for some reason the permissions on /dev/null
2945 got screwed. Now Exim writes a message to the main and panic logs, and
2946 bombs out if it can't open /dev/null.
2948 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
2949 interact so that it is all more flexible. It is supposed to remain
2950 backwards compatible. Also added extra_local_interfaces.
2952 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
2953 to bomb out with an assertion failure - to the client this appears as a
2954 connection drop. This problem occurs in the part of the code that was taken
2955 from the Samba project. Fortunately, the assertion is in a very simple
2956 function, so I have fixed this by reproducing the function inline in the
2957 one place where it is called, and arranging for authentication to fail
2958 instead of killing the process with assert().
2960 75. The SPA client code was not working when the server requested OEM rather
2961 than Unicode encoding.
2963 76. Added code to make require_files with a specific uid setting more usable in
2964 the case where statting the file as root fails - usually a non-root-mounted
2965 NFS file system. When this happens and the failure is EACCES, Exim now
2966 forks a subprocess and does the per-uid checking as the relevant uid.
2968 77. Added process_log_path.
2970 78. If log_file_path was not explicitly set, a setting of check_log_space or
2971 check_log_inodes was ignored.
2973 79. If a space check for the spool or log partitions fails, the incident is now
2974 logged. Of course, in the latter case the data may get lost...
2976 80. Added the %p formatting code to string_format() so that it can be used to
2977 print addresses in debug_print(). Adjusted all the address printing in the
2978 debugging in store.c to use %p rather than %d.
2980 81. There was a concern that a line of code in smtp_in.c could overflow a
2981 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
2982 initially expressed, the concern was not well-founded, because trailing
2983 spaces are removed early. However, if the trailing spaces were followed by
2984 a NULL, they did not get removed, so the overflow was possible. Two fixes
2987 (a) I re-wrote the offending code in a cleaner fashion.
2988 (b) If an incoming SMTP command contains a NULL character, it is rejected
2991 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
2992 runs initgroups(), so that if the Exim user is in any additional groups,
2993 they will be used during message reception.
2999 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob