1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.206 2005/08/22 10:49:04 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added the "success_on_redirect" address verification option. See
10 NewStuff for rationale and an example.
12 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
14 PH/02 Patch to exigrep to allow it to work on syslog lines.
16 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
17 fread() to skip over the body file's header line, because in Cygwin the
18 header line is locked and is inaccessible.
20 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
21 co-exist for some time) to make it clear that it is the Exim ID that is
22 referenced, not the Message-ID: header line.
24 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
25 string_format() function, because snprintf() does not exist on all
28 PH/06 The use of forbid_filter_existstest now also locks out the use of the
29 ${stat: expansion item.
31 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
32 protocol synchronization error", to keep the pedants happy.
34 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
35 well as for IRIX systems, when gcc is being used. See the host.c source
38 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
40 PH/10 Named domain lists were not working if used in a queue_smtp_domains
43 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
44 transport and to the smtp transport in LMTP mode.
46 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
48 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
49 run a filter in a subprocess. This could lead to confusion in subsequent
50 lookups in the parent process. There should also be a search_tidyup() at
51 the end of the subprocess.
53 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
54 only if the host matched helo_try_verify_hosts, which caused the
55 verification to occur when the EHLO/HELO command was issued. The ACL just
56 tested the remembered result. Now, if a previous verification attempt has
57 not happened, "verify = helo" does it there and then.
59 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
61 TK/03 Fix log output including CR from clamd.
63 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
64 exist provoked a memory error which could cause a segfault.
66 PH/15 Installed PCRE 6.2
68 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
70 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
71 of the problem. Specifically, suggested +O2 rather than +O1 for the
74 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
76 PH/20 If a delivery was routed to a non-standard port by means of an SRV
77 record, the port was not correctly logged when the outgoing_port log
78 selector was set (it logged the transort's default port).
80 PH/21 Added support for host-specific ports to manualroute, queryprogram,
81 fallback_hosts, and "hosts" in the smtp transport.
83 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
84 host errors such as "Connection refused".
86 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
87 authentication with radiusclient 0.4.9:
89 - Error returned from rc_read_config was caught wrongly
90 - Username/password not passed on to radius server due to wrong length.
92 The presumption is that some radiusclient API changes for 4.51/PH/17
93 were not taken care of correctly. The code is still untested by me (my
94 Linux distribution still has 0.3.2 of radiusclient), but it was
95 contributed by a Radius user.
97 PH/24 When doing a callout, the value of $domain wasn't set correctly when
98 expanding the "port" option of the smtp transport.
100 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
101 while reading a MIME header. Thanks to Tom Hughes for a patch.
103 PH/24 Include config.h inside local_scan.h so that configuration settings are
110 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
112 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
113 command itself ended in error, the underlying message about the transport
114 filter timeout was being overwritten with the pipe command error. Now the
115 underlying error message should be appended to the second error message.
117 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
119 PH/02 Reduce the amount of output that "make" produces by default. Full output
120 can still be requested.
122 PH/03 The warning log line about a condition test deferring for a "warn" verb
123 was being output only once per connection, rather than after each
124 occurrence (because it was using the same function as for successful
125 "warn" verbs). This seems wrong, so I have changed it.
127 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
128 it should not have, which might have caused a crash in the right
129 circumstances, but probably never did.
131 PH/04 Installed a modified version of Tony Finch's patch to make submission
132 mode fix the return path as well as the Sender: header line, and to
133 add a /name= option so that you can make the user's friendly name appear
136 TF/03 Added the control = fakedefer ACL modifier.
138 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
139 Mark Lowes for thorough testing.
141 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
143 TK/03 Merged latest SRS patch from Miles Wilton.
145 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
146 with the definition in sysexits.h (which is #included earlier).
147 Fortunately, Exim does not actually use EX_OK. The code used to try to
148 preserve the sysexits.h value, by assumimg that macro definitions were
149 scanned for macro replacements. I have been disabused of this notion,
150 so now the code just undefines EX_OK before #including unistd.h.
152 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
153 in the smtp transport. When a block could not be written in a single
154 write() function, the timeout was being re-applied to each part-write.
155 This seems wrong - if the receiver was accepting one byte at a time it
156 would take for ever. The timeout is now adjusted when this happens. It
157 doesn't have to be particularly precise.
159 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
160 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
162 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
163 without a domain if the check to <postmaster@domain> fails.
165 SC/01 Eximstats: added -xls and the ability to specify output files
166 (patch written by Frank Heydlauf).
168 SC/02 Eximstats: use FileHandles for outputing results.
170 SC/03 Eximstats: allow any combination of xls, txt, and html output.
172 SC/04 Eximstats: fixed display of large numbers with -nvr option
174 SC/05 Eximstats: fixed merging of reports with empty tables.
176 SC/06 Eximstats: added the -include_original_destination flag
178 SC/07 Eximstats: removed tabs and trailing whitespace.
180 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
182 TK/06 MBOX spool code: Add real "From " MBOX separator line
183 so the .eml file is really in mbox format (even though
184 most programs do not really care). Patch from Alex Miller.
186 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
187 The latter is generated from $received_to and is only set if the
188 message has one envelope recipient. SA can use these headers,
189 obviously out-of-the-box. Patch from Alex Miller.
191 PH/08 The ${def test on a variable was returning false if the variable's
192 value was "0", contrary to what the specification has always said!
193 The result should be true unless the variable is empty.
195 PH/09 The syntax error of a character other than { following "${if
196 def:variable_name" (after optional whitespace) was not being diagnosed.
197 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
198 accidental colon was present, for example, could give incorrect results.
200 PH/10 Tidied the code in a number of places where the st_size field of a stat()
201 result is used (not including appendfile, where other changes are about
204 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
205 This involved changing a lot of size variables from int to off_t. It
206 should work with maildirs and everything.
208 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
209 spamd dying while we are connected to it.
211 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
214 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
215 function for BDB 4.3.
217 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
219 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
220 the sources for PCRE have been re-arranged and more files are now
223 PH/15 The code I had for printing potentially long long variables in PH/11
224 above was not the best (it lost precision). The length of off_t variables
225 is now inspected at build time, and an appropriate printing format (%ld
226 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
227 to be "long long int" or "long int". This is needed for the internal
228 formatting function string_vformat().
230 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
231 the configuration file to be ":syslog", then the script "guesses" where
232 the logs files are, rather than using the compiled in default. In our
233 case the guess is not the same as the compiled default, so the script
234 suddenly stopped working when I started to use syslog. The patch checks
235 to see if log_file_path is "". If so, it attempts to read it from exim
236 with no configuration file to get the compiled in version, before it
237 falls back to the previous guessing code."
239 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
240 implementing BATV in an Exim configuration. See NewStuff for the gory
243 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
244 Makefile that are specific to HP-UX.
246 PH/18 If the "use_postmaster" option was set for a recipient callout together
247 with the "random" option, the postmaster address was used as the MAIL
248 FROM address for the random test, but not for the subsequent recipient
249 test. It is now used for both.
251 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
252 patch removes a few documentation additions to RFC 3028, because the
253 latest draft now contains them. It adds the new en;ascii-case comparator
254 and a new error check for 8bit text in MIME parts. Comparator and
255 require names are now matched exactly. I enabled the subaddress
256 extension, but it is not well tested yet (read: it works for me)."
258 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
259 rework some of the code of TK/09 above to avoid the hardwired use of
260 "%lld" and "long long". Replaced the call to snprintf() with a call to
263 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
264 records point to non-existent hosts", "retry timeout exceeded", and
265 "retry time not reached for any host after a long failure period".
267 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
268 experimental DomainKeys support:
270 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
271 (2) On an error such as an illegally used "control", the wrong name for
272 the control was given.
274 These problems did NOT occur unless DomainKeys support was compiled.
276 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
278 PH/24 Added ${if match_ip condition.
280 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
281 databases so that it will be absolutely obvious if a crash occurs in the
282 DB library. This is a regular occurrence (often caused by mis-matched
285 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
286 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
287 picked up on a user's system that detects such things. There doesn't seem
288 to be a gcc warning option for this - only an attribute that has to be
289 put on the function's prototype. It seems that in Fedora Core 4 they have
290 set this on a number of new functions. No doubt there will be more in due
293 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
294 specify a transport. However, if an address that was verified by such a
295 router was the subject of a callout, Exim crashed because it tried to
296 read the rcpt_include_affixes from the non-existent transport. Now it
297 just assumes that the setting of that option is false. This bug was
298 introduced by 4.51/PH/31.
300 PH/28 Changed -d+all to exclude +memory, because that information is very
301 rarely of interest, but it makes the output a lot bigger. People tend to
302 do -d+all out of habit.
304 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
305 code in os-type was giving problems when libc.so lives in lib64, like on
308 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
309 aren't the modern standard, and it seems that some systems' include files
310 don't always have them. Exim was already checking for some of the newer
311 ones like T_AAAA, and defining it itself. I've added checks for all the
312 record types that Exim uses.
314 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
315 not automatically generating a new one, as it is supposed to. This
316 prevented TLS from working. If the file did exist, but contained invalid
317 data, a new version was generated, as expected. It was only the case of a
318 non-existent file that was broken.
320 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
321 with a change in libdomainkeys > 0.64.
323 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
324 from DNS. If the selector record carries the flag, it now has
325 precedence over the domain-wide flag.
327 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
329 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
330 the use of a port name that isn't defined in /etc/services) occurred, the
331 message was deferred as in a normal delivery, and thus remained on the
332 spool, instead of being failed because of the mua_wrapper setting. This
333 is now fixed, and I tidied up some of the mua_wrapper messages at the
336 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
337 the messages in a hash of arrays rather than using individual hashes.
338 This is a bit cleaner and results in dramatic memory savings, albeit
339 at a slight CPU cost.
341 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
342 as requested by Marc Sherman.
344 SC/10 Eximstats: added histograms for user specified patterns as requested
347 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
349 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
350 fopen() in the content-scanning modules that did not already have it.
356 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
357 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
359 TK/02 Fix ACL "control" statement not being available in MIME ACL.
361 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
363 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
364 to test Sieve filters that use "vacation".
366 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
367 that changes the way the GnuTLS parameters are stored in the cache file.
368 The new format can be generated externally. For backward compatibility,
369 if the data in the cache doesn't make sense, Exim assumes it has read an
370 old-format file, and it generates new data and writes a new file. This
371 means that you can't go back to an older release without removing the
374 PH/03 A redirect router that has both "unseen" and "one_time" set does not
375 work if there are any delivery delays because "one_time" forces the
376 parent to be marked "delivered", so its unseen clone is never tried
377 again. For this reason, Exim now forbids the simultaneous setting of
380 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
381 redirected to themselves ("homonym" addresses). Read the long ChangeLog
382 entry if you want to know the details. The fix, however, neglected to
383 consider the case when local delivery batching is involved. The test for
384 "previously delivered" was not happening when checking to see if an
385 address could be batched with a previous (undelivered) one; under
386 certain circumstances this could lead to multiple deliveries to the same
389 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
390 in its include files, and this causes problems building Exim.
392 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
393 header_syntax) but Exim was just ignoring anything given after a slash.
394 In particular, this caused confusion with an attempt to use "verify =
395 reverse_host_lookup/defer_ok". An error is now given when options are
396 supplied for verify items that do not have them. (Maybe reverse_host_
397 lookup should have a defer_ok option, but that's a different point.)
399 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
400 defined by RFC 821) to 2048, because there were problems with some AUTH
401 commands, and RFC 1869 says the size should be increased for extended
402 SMTP commands that take arguments.
404 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
407 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
408 "unknown" error; now it says that the functionality isn't in the binary.
410 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
411 an address' error message when a string expansion fails (syntax or
412 whatever). Otherwise the password may appear in the log. Following change
413 PH/42 below, there is no longer a chance of it appearing in a bounce
416 PH/11 Installed exipick version 20050225.0 from John Jetmore.
418 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
419 of its addresses was ever tried. (Bugzilla bug #2.)
421 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
422 the result incorrectly in the debug output. (It correctly added a newline
423 to what was transported.)
425 TF/01 Added $received_time.
427 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
428 commented out examples of how to interface to a virus scanner and to
429 SpamAssassin. Also added commented examples of av_scanner and
430 spamd_address settings.
432 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
433 and controls are allowed in which ACLs. There were a couple of minor
434 errors. Some of the entries in the conditions table (which is a table of
435 where they are NOT allowed) were getting very unwieldy; rewrote them as a
436 negation of where the condition IS allowed.
438 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
440 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
441 header file does not have a version number, so I've had to invent a new
442 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
443 API. The code is untested by me (my Linux distribution still has 0.3.2 of
444 radiusclient), but it was contributed by a Radius user.
446 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
447 files or directories, for OpenSSL.
449 PH/19 When an Exim process that is running as root has to create an Exim log
450 file, it does so in a subprocess that runs as exim:exim so as to get the
451 ownership right at creation (otherwise, other Exim processes might see
452 the file with the wrong ownership). There was no test for failure of this
453 fork() call, which would lead to the process getting stuck as it waited
454 for a non-existent subprocess. Forks do occasionally fail when resources
455 run out. I reviewed all the other calls to fork(); they all seem to check
458 PH/20 When checking for unexpected SMTP input at connect time (before writing
459 the banner), Exim was not dealing correctly with a non-positive return
460 from the read() function. If the client had disconnected by this time,
461 the result was a log entry for a synchronization error with an empty
462 string after "input=" when read() returned zero. If read() returned -1
463 (an event I could not check), uninitialized data bytes were printed.
464 There were reports of junk text (parts of files, etc) appearing after
467 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
469 PH/22 Added support for macro redefinition, and (re)definition in between
470 driver and ACL definitions.
472 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
473 forgetting to use the resulting value; it was using the unexpanded value.
475 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
476 hadn't been configured. The fix is from Juergen Kreileder, who
477 understands it better than I do:
479 "Here's what I see happening with three configured cyrus_sasl
480 authenticators configured (plain, login, cram-md5):
482 On startup auth_cyrus_sasl_init() gets called for each of these.
483 This means three calls to sasl_listmech() without a specified mech_list.
484 => SASL tests which mechs of all available mechs actually work
485 => three warnings about OTP not working
486 => the returned list contains: plain, login, cram-md5, digest-md5, ...
488 With the patch, sasl_listmech() also gets called three times. But now
489 SASL's mech_list option is set to the server_mech specified in the the
490 authenticator. Or in other words, the answer from sasl_listmech()
491 gets limited to just the mech you're testing for (which is different
493 => the return list contains just 'plain' or 'login', 'cram-md5' or
494 nothing depending on the value of ob->server_mech.
496 I've just tested the patch: Authentication still works fine,
497 unavailable mechs specified in the exim configuration are still
498 caught, and the auth.log warnings about OTP are gone."
500 PH/25 When debugging is enabled, the contents of the command line are added
501 to the debugging output, even when log_selector=+arguments is not
504 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
505 answer is "GNU", and only if the return is "GNU/something" is the answer
508 PH/27 $acl_verify_message is now set immediately after the failure of a
509 verification in an ACL, and so is available in subsequent modifiers. In
510 particular, the message can be preserved by coding like this:
512 warn !verify = sender
513 set acl_m0 = $acl_verify_message
515 Previously, $acl_verify_message was set only while expanding "message"
516 and "log_message" when a very denied access.
518 PH/28 Modified OS/os.c-Linux with
520 -#ifndef OS_LOAD_AVERAGE
521 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
523 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
524 nomenclature these days.)
526 PH/29 Installed patch from the Sieve maintainer that adds the options
527 sieve_useraddress and sieve_subaddress to the redirect router.
529 PH/30 In these circumstances:
530 . Two addresses routed to the same list of hosts;
531 . First host does not offer TLS;
532 . First host accepts first address;
533 . First host gives temporary error to second address;
534 . Second host offers TLS and a TLS session is established;
535 . Second host accepts second address.
536 Exim incorrectly logged both deliveries with the TLS parameters (cipher
537 and peerdn, if requested) that were in fact used only for the second
540 PH/31 When doing a callout as part of verifying an address, Exim was not paying
541 attention to any local part prefix or suffix that was matched by the
542 router that accepted the address. It now behaves in the same way as it
543 does for delivery: the affixes are removed from the local part unless
544 rcpt_include_affixes is set on the transport.
546 PH/32 Add the sender address, as F=<...>, to the log line when logging a
547 timeout during the DATA phase of an incoming message.
549 PH/33 Sieve envelope tests were broken for match types other than :is. I have
550 applied a patch sanctioned by the Sieve maintainer.
552 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
553 the uid or gid is negative. A case of a negative gid caused this to be
554 noticed. The fix allows for either to be negative.
556 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
557 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
560 PH/36 The change PH/12 above was broken. Fixed it.
562 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
563 the grounds that routing the same address twice would always produce the
564 same answer. This might have been true once, but it is certainly no
565 longer true now. Routing a child address may depend on the previous
566 routing that produced that child. Some complicated redirection strategies
567 went wrong when messages had multiple recipients, and made Exim's
568 behaviour dependent on the order in which the addresses were given.
570 I have moved the duplicate checking until after the routing is complete.
571 Exim scans the addresses that are assigned to local and remote
572 transports, and removes any duplicates. This means that more work will be
573 done, as duplicates will always all be routed, but duplicates are
574 presumably rare, so I don't expect this is of any significance.
576 For deliveries to pipes, files, and autoreplies, the duplicate checking
577 still happens during the routing process, since they are not going to be
580 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
581 It corrects a timeout issue with spamd. This is Ian's comment: "The
582 background is that sometimes spamd either never reads data from a
583 connection it has accepted, or it never writes response data. The exiscan
584 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
585 blindly assumes that writes won't block so it may never time out."
587 PH/39 Allow G after quota size as well as K and M.
589 PH/40 The value set for $authenticated_id in an authenticator may not contain
590 binary zeroes or newlines because the value is written to log lines and
591 to spool files. There was no check on this. Now the value is run through
592 the string_printing() function so that such characters are converted to
593 printable escape sequences.
595 PH/41 $message_linecount is a new variable that contains the total number of
596 lines in the message. Compare $body_linecount, which is the count for the
599 PH/42 Exim no longer gives details of delivery errors for specific addresses in
600 bounce and delay warning messages, except in certain special cases, which
603 (a) An SMTP error message from a remote host;
604 (b) A message specified in a :fail: redirection;
605 (c) A message specified in a "fail" command in a system filter;
606 (d) A message specified in a FAIL return from the queryprogram router;
607 (e) A message specified by the cannot_route_message router option.
609 In these cases only, Exim does include the error details in bounce and
610 warning messages. There are also a few cases where bland messages such
611 as "unrouteable address" or "local delivery error" are given.
613 PH/43 $value is now also set for the "else" part of a ${run expansion.
615 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
616 being worked on, but at least Exim now implements the latest version to
619 PH/45 In a pipe transport, although a timeout while waiting for the pipe
620 process to complete was treated as a delivery failure, a timeout while
621 writing the message to the pipe was logged, but erroneously treated as a
622 successful delivery. Such timeouts include transport filter timeouts. For
623 consistency with the overall process timeout, these timeouts are now
624 treated as errors, giving rise to delivery failures by default. However,
625 there is now a new Boolean option for the pipe transport called
626 timeout_defer, which, if set TRUE, converts the failures into defers for
627 both kinds of timeout. A transport filter timeout is now identified in
630 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
631 systems where "make" and "gmake" are different, calling "gmake" at top
632 level broke things. I've arranged for the value of $(MAKE) to be passed
633 from the Makefile to this script so that it can call the same version of
637 A note about Exim versions 4.44 and 4.50
638 ----------------------------------------
640 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
641 changes of various kinds. As a consequence, a big documentation update was
642 needed. This delayed the release for rather longer than seemed good, especially
643 in the light of a couple of (minor) security issues. Therefore, the changes
644 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
645 release. So 4.44 and 4.50 are in effect two different branches that both start
648 I have left the 4.50 change log unchanged; it contains all the changes since
649 4.43. The change log for 4.44 is below; many of its items are identical to
650 those for 4.50. This seems to be the most sensible way to preserve the
651 historical information.
657 1. Minor wording change to the doc/README.SIEVE file.
659 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
660 computation of the current number of files was incorrect.
662 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
663 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
666 4. Give more explanation in the error message when the command for a transport
667 filter fails to execute.
669 5. There are several places where Exim runs a non-Exim command in a
670 subprocess. The SIGUSR1 signal should be disabled for these processes. This
671 was being done only for the command run by the queryprogram router. It is
672 now done for all such subprocesses. The other cases are: ${run, transport
673 filters, and the commands run by the lmtp and pipe transports.
675 6. Added CONFIGURE_GROUP build-time option.
677 7. Some older OS have a limit of 256 on the maximum number of file
678 descriptors. Exim was using setrlimit() to set 1000 as a large value
679 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
680 systems. I've change it so that if it can't get 1000, it tries for 256.
682 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
683 was an oversight, and furthermore, ever since the addition of extra
684 controls (e.g. 4.43/32), the checks on when to allow different forms of
685 "control" were broken. There should now be diagnostics for all cases when a
686 control that does not make sense is encountered.
688 9. Added the /retain_sender option to "control=submission".
690 10. $recipients is now available in the predata ACL (oversight).
692 11. Tidy the search cache before the fork to do a delivery from a message
693 received from the command line. Otherwise the child will trigger a lookup
694 failure and thereby defer the delivery if it tries to use (for example) a
695 cached ldap connection that the parent has called unbind on.
697 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
698 of $address_data from the recipient verification was clobbered by the
701 13. The value of address_data from a sender verification is now available in
702 $sender_address_data in subsequent conditions in the ACL statement.
704 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
706 15. Added a new option "connect=<time>" to callout options, to set a different
709 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
710 was its contents. (It was OK if the option was not defined at all.)
712 17. A "Completed" log line is now written for messages that are removed from
713 the spool by the -Mrm option.
715 18. New variables $sender_verify_failure and $recipient_verify_failure contain
716 information about exactly what failed.
718 19. Added -dd to debug only the daemon process.
720 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
721 handles timeouts, both on the server side and network timeouts. Renamed the
722 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
724 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
725 It is now set to "smtps".
727 22. $host_address is now set to the target address during the checking of
730 23. When checking ignore_target_hosts for an ipliteral router, no host name was
731 being passed; this would have caused $sender_host_name to have been used if
732 matching the list had actually called for a host name (not very likely,
733 since this list is usually IP addresses). A host name is now passed as
736 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
737 code that specifies a non-restarting handler (typically sigaction() in
738 modern systems) in an attempt to fix a rare and obscure crash bug.
740 25. Narrowed the window for a race in the daemon that could cause it to ignore
741 SIGCHLD signals. This is not a major problem, because they are used only to
742 wake it up if nothing else does.
744 26. A malformed maildirsize file could cause Exim to calculate negative values
745 for the mailbox size or file count. Odd effects could occur as a result.
746 The maildirsize information is now recalculated if the size or filecount
749 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
750 support for a long time. Removed HAVE_SYS_VFS_H.
752 28. Installed the latest version of exipick from John Jetmore.
754 29. In an address list, if the pattern was not a regular expression, an empty
755 subject address (from a bounce message) matched only if the pattern was an
756 empty string. Non-empty patterns were not even tested. This was the wrong
757 because it is perfectly reasonable to use an empty address as part of a
758 database query. An empty address is now tested by patterns that are
759 lookups. However, all the other forms of pattern expect the subject to
760 contain a local part and a domain, and therefore, for them, an empty
761 address still always fails if the pattern is not itself empty.
763 30. Exim went into a mad DNS loop when attempting to do a callout where the
764 host was specified on an smtp transport, and looking it up yielded more
767 31. Re-factored the code for checking spool and log partition space into a
768 function that finds that data and another that does the check. The former
769 is then used to implement four new variables: $spool_space, $log_space,
770 $spool_inodes, and $log_inodes.
772 32. The RFC2047 encoding function was originally intended for short strings
773 such as real names; it was not keeping to the 75-character limit for
774 encoded words that the RFC imposes. It now respects the limit, and
775 generates multiple encoded words if necessary. To be on the safe side, I
776 have increased the buffer size for the ${rfc2047: expansion operator from
779 33. It is now permitted to omit both strings after an "if" condition; if the
780 condition is true, the result is "true". As before, when the second string
781 is omitted, a false condition yields an empty string. This makes it less
782 cumbersome to write custom ACL and router conditions.
784 34. Failure to deliver a bounce message always caused it to be frozen, even if
785 there was an errors_to setting on the router. The errors_to setting is now
788 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
789 canonical form (fully expanded) before being placed in
790 $sender_host_address.
792 36. The table in the code that translates DNS record types into text (T_A to
793 "A" for instance) was missing entries for NS and CNAME. It is just possible
794 that this could have caused confusion if both these types were looked up
795 for the same domain, because the text type is used as part of Exim's
796 per-process caching. But the chance of anyone hitting this buglet seems
799 37. The dnsdb lookup has been extended in a number of ways.
801 (1) There is a new type, "zns", which walks up the domain tree until it
802 finds some nameserver records. It should be used with care.
804 (2) There is a new type, "mxh", which is like "mx" except that it returns
805 just the host names, not the priorities.
807 (3) It is now possible to give a list of domains (or IP addresses) to be
808 looked up. The behaviour when one of the lookups defers can be
809 controlled by a keyword.
811 (4) It is now possible to specify the separator character for use when
812 multiple records are returned.
814 38. The dnslists ACL condition has been extended: it is now possible to supply
815 a list of IP addresses and/or domains to be looked up in a particular DNS
818 39. Added log_selector=+queue_time_overall.
820 40. When running the queue in the test harness, wait just a tad after forking a
821 delivery process, to get repeatability of debugging output.
823 41. Include certificate and key file names in error message when GnuTLS fails
824 to set them up, because the GnuTLS error message doesn't include the name
825 of the failing file when there is a problem reading it.
827 42. Allow both -bf and -bF in the same test run.
829 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
831 44. The "Exiscan patch" is now merged into the mainline Exim source.
833 45. Sometimes the final signoff response after QUIT could fail to get
834 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
835 < 0 before doing a fflush(). This bug looks as though it goes back to the
836 introduction of TLS in release 3.20, but "sometimes" must have been rare
837 because the tests only now provoked it.
839 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
840 (this can affect the format of dates).
842 47. exim_tidydb, when checking for the continued existence of a message for
843 which it has found a message-specific retry record, was not finding
844 messages that were in split spool directories. Consequently, it was
845 deleting retry records that should have stayed in existence.
847 48. Steve fixed some bugs in eximstats.
849 49. The SPA authentication driver was not abandoning authentication and moving
850 on to the next authenticator when an expansion was forced to fail,
851 contradicting the general specification for all authenticators. Instead it
852 was generating a temporary error. It now behaves as specified.
854 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
855 (the order specifies the preference for clients). The order is now AES256,
856 AES128, 3DES, ARCFOUR128.
858 51. Small patch to Sieve code - explicitly set From: when generating an
861 52. Exim crashed if a remote delivery caused a very long error message to be
862 recorded - for instance if somebody sent an entire SpamAssassin report back
863 as a large number of 550 error lines. This bug was coincidentally fixed by
864 increasing the size of one of Exim's internal buffers (big_buffer) that
865 happened as part of the Exiscan merge. However, to be on the safe side, I
866 have made the code more robust (and fixed the comments that describe what
869 53. Now that there can be additional text after "Completed" in log lines (if
870 the queue_time_overall log selector is set), a one-byte patch to exigrep
871 was needed to allow it to recognize "Completed" as not the last thing in
874 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
875 patch that reportedly fixes this has been added. I am not expert enough to
876 create a test for it. This is what the patch creator wrote:
878 "I found a little strange behaviour of ldap code when working with
879 Windows 2003 AD Domain, where users was placed in more than one
880 Organization Units. When I tried to give exim partial DN, the exit code
881 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
882 But simultaneously result of request was absolutely normal ldap result,
883 so I produce this patch..."
885 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
886 REFERENCE, so I have modified the code to exclude the patch when that macro
889 55. Some experimental protocols are using DNS PTR records for new purposes. The
890 keys for these records are domain names, not reversed IP addresses. The
891 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
892 leaves it alone. Component reversal etc. now happens only for IP addresses.
895 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
897 57. Double the size of the debug message buffer (to 2048) so that more of very
898 long debug lines gets shown.
900 58. The exicyclog utility now does better if the number of log files to keep
901 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
903 59. Two changes related to the smtp_active_hostname option:
905 (1) $smtp_active_hostname is now available as a variable.
906 (2) The default for smtp_banner uses $smtp_active_hostname instead
907 of $primary_hostname.
909 60. The host_aton() function is supposed to be passed a string that is known
910 to be a valid IP address. However, in the case of IPv6 addresses, it was
911 not checking this. This is a hostage to fortune. Exim now panics and dies
912 if the condition is not met. A case was found where this could be provoked
913 from a dnsdb PTR lookup with an IPv6 address that had more than 8
914 components; fortuitously, this particular loophole had already been fixed
915 by change 4.50/55 above.
917 If there are any other similar loopholes, the new check in host_aton()
918 itself should stop them being exploited. The report I received stated that
919 data on the command line could provoke the exploit when Exim was running as
920 exim, but did not say which command line option was involved. All I could
921 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
925 61. There was a buffer overflow vulnerability in the SPA authentication code
926 (which came originally from the Samba project). I have added a test to the
927 spa_base64_to_bits() function which I hope fixes it.
930 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
931 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
933 63. The daemon start-up calls getloadavg() while still root for those OS that
934 need the first call to be done as root, but it missed one case: when
935 deliver_queue_load_max is set with deliver_drop_privilege. This is
936 necessary for the benefit of the queue runner, because there is no re-exec
937 when deliver_drop_privilege is set.
939 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
942 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
943 was in use, was not putting the data itself into the right store pool;
944 consequently, it could be overwritten for a subsequent message in the same
945 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
948 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
950 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
951 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
952 the function were treating the return as a boolean value, which happened to
953 work because 0=false and not-0=true, but is not correct code.
955 68. The host_aton() function was not handling scoped IPv6 addresses (those
956 with, for example, "%eth0" on the end) correctly.
958 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
959 negated items (that is, ~something) in unsigned ints. Some compilers
960 apparently mutter when there is no cast.
962 70. If an address verification called from an ACL failed, and did not produce a
963 user-specific message (i.e. there was only a "system" message), nothing was
964 put in $acl_verify_message. In this situation, it now puts the system
967 71. Change 4.23/11 added synchronization checking at the start of an SMTP
968 session; change 4.31/43 added the unwanted input to the log line - except
969 that it did not do this in the start of session case. It now does.
971 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
972 This is wrong and can cause the other end to generate a synchronization
973 error if it is another Exim or anything else that does the synchronization
974 check. A QUIT command is no longer sent after a timeout.
976 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
979 74. The defer_ok option of callout verification was not working if it was used
980 when verifying addresses in header lines, that is, for this case:
982 verify = header_sender/callout=defer_ok
984 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
985 those file descriptors could be used for SMTP connections. If anything
986 wrote to stderr (the example that came up was "warn" in embedded Perl), it
987 could be sent to the SMTP client, causing chaos. The daemon now opens
988 stdin, stdout, and stderr to /dev/null when it puts itself into the
991 76. Arrange for output from Perl's "warn" command to be written to Exim's main
992 log by default. The user can override this with suitable Perl magic.
994 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
995 the log message when discard triggers, was not working for the DATA ACL or
996 for the non-SMTP ACL.
998 78. Error message wording change in sieve.c.
1000 79. If smtp_accept_max_per_host was set, the number of connections could be
1001 restricted to fewer than expected, because the daemon was trying to set up
1002 a new connection before checking whether the processes handling previous
1003 connections had finished. The check for completed processes is now done
1004 earlier. On busy systems, this bug wouldn't be noticed because something
1005 else would have woken the daemon, and it would have reaped the completed
1008 80. If a message was submitted locally by a user whose login name contained one
1009 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1010 It caused a spool format error. I have fixed the spool reading code. A
1011 related problem was that the "from" clause in the Received: line became
1012 illegal because of the space(s). It is now covered by ${quote_local_part.
1014 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1017 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1019 83. Patch from Sieve maintainer to fix unterminated string problem in
1020 "vacation" handling.
1022 84. Some minor changes to the Linux configuration files to help with other
1023 OS variants using glibc.
1025 85. One more patch for Sieve to update vacation handling to latest spec.
1028 ----------------------------------------------------
1029 See the note above about the 4.44 and 4.50 releases.
1030 ----------------------------------------------------
1036 1. Change 4.43/35 introduced a bug that caused file counts to be
1037 incorrectly computed when quota_filecount was set in an appendfile
1040 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1041 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1044 3. Give more explanation in the error message when the command for a transport
1045 filter fails to execute.
1047 4. There are several places where Exim runs a non-Exim command in a
1048 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1049 was being done only for the command run by the queryprogram router. It is
1050 now done for all such subprocesses. The other cases are: ${run, transport
1051 filters, and the commands run by the lmtp and pipe transports.
1053 5. Some older OS have a limit of 256 on the maximum number of file
1054 descriptors. Exim was using setrlimit() to set 1000 as a large value
1055 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1056 systems. I've change it so that if it can't get 1000, it tries for 256.
1058 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1059 was an oversight, and furthermore, ever since the addition of extra
1060 controls (e.g. 4.43/32), the checks on when to allow different forms of
1061 "control" were broken. There should now be diagnostics for all cases when a
1062 control that does not make sense is encountered.
1064 7. $recipients is now available in the predata ACL (oversight).
1066 8. Tidy the search cache before the fork to do a delivery from a message
1067 received from the command line. Otherwise the child will trigger a lookup
1068 failure and thereby defer the delivery if it tries to use (for example) a
1069 cached ldap connection that the parent has called unbind on.
1071 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1072 of $address_data from the recipient verification was clobbered by the
1073 sender verification.
1075 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1076 was its contents. (It was OK if the option was not defined at all.)
1078 11. A "Completed" log line is now written for messages that are removed from
1079 the spool by the -Mrm option.
1081 12. $host_address is now set to the target address during the checking of
1082 ignore_target_hosts.
1084 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1085 being passed; this would have caused $sender_host_name to have been used if
1086 matching the list had actually called for a host name (not very likely,
1087 since this list is usually IP addresses). A host name is now passed as
1090 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1091 code that specifies a non-restarting handler (typically sigaction() in
1092 modern systems) in an attempt to fix a rare and obscure crash bug.
1094 15. Narrowed the window for a race in the daemon that could cause it to ignore
1095 SIGCHLD signals. This is not a major problem, because they are used only to
1096 wake it up if nothing else does.
1098 16. A malformed maildirsize file could cause Exim to calculate negative values
1099 for the mailbox size or file count. Odd effects could occur as a result.
1100 The maildirsize information is now recalculated if the size or filecount
1103 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1104 support for a long time. Removed HAVE_SYS_VFS_H.
1106 18. Updated exipick to current release from John Jetmore.
1108 19. Allow an empty sender to be matched against a lookup in an address list.
1109 Previously the only cases considered were a regular expression, or an
1112 20. Exim went into a mad DNS lookup loop when doing a callout where the
1113 host was specified on the transport, if the DNS lookup yielded more than
1116 21. The RFC2047 encoding function was originally intended for short strings
1117 such as real names; it was not keeping to the 75-character limit for
1118 encoded words that the RFC imposes. It now respects the limit, and
1119 generates multiple encoded words if necessary. To be on the safe side, I
1120 have increased the buffer size for the ${rfc2047: expansion operator from
1123 22. Failure to deliver a bounce message always caused it to be frozen, even if
1124 there was an errors_to setting on the router. The errors_to setting is now
1127 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1128 canonical form (fully expanded) before being placed in
1129 $sender_host_address.
1131 24. Updated eximstats to version 1.33
1133 25. Include certificate and key file names in error message when GnuTLS fails
1134 to set them up, because the GnuTLS error message doesn't include the name
1135 of the failing file when there is a problem reading it.
1137 26. Expand error message when OpenSSL has problems setting up cert/key files.
1140 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1141 (this can affect the format of dates).
1143 28. exim_tidydb, when checking for the continued existence of a message for
1144 which it has found a message-specific retry record, was not finding
1145 messages that were in split spool directories. Consequently, it was
1146 deleting retry records that should have stayed in existence.
1148 29. eximstats updated to version 1.35
1149 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1150 1.35 - bugfix such that pie charts by volume are generated correctly
1152 30. The SPA authentication driver was not abandoning authentication and moving
1153 on to the next authenticator when an expansion was forced to fail,
1154 contradicting the general specification for all authenticators. Instead it
1155 was generating a temporary error. It now behaves as specified.
1157 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1158 (the order specifies the preference for clients). The order is now AES256,
1159 AES128, 3DES, ARCFOUR128.
1161 31. Small patch to Sieve code - explicitly set From: when generating an
1164 32. Exim crashed if a remote delivery caused a very long error message to be
1165 recorded - for instance if somebody sent an entire SpamAssassin report back
1166 as a large number of 550 error lines. This bug was coincidentally fixed by
1167 increasing the size of one of Exim's internal buffers (big_buffer) that
1168 happened as part of the Exiscan merge. However, to be on the safe side, I
1169 have made the code more robust (and fixed the comments that describe what
1172 33. Some experimental protocols are using DNS PTR records for new purposes. The
1173 keys for these records are domain names, not reversed IP addresses. The
1174 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1175 leaves it alone. Component reversal etc. now happens only for IP addresses.
1178 34. The host_aton() function is supposed to be passed a string that is known
1179 to be a valid IP address. However, in the case of IPv6 addresses, it was
1180 not checking this. This is a hostage to fortune. Exim now panics and dies
1181 if the condition is not met. A case was found where this could be provoked
1182 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1183 components; fortuitously, this particular loophole had already been fixed
1184 by change 4.50/55 or 4.44/33 above.
1186 If there are any other similar loopholes, the new check in host_aton()
1187 itself should stop them being exploited. The report I received stated that
1188 data on the command line could provoke the exploit when Exim was running as
1189 exim, but did not say which command line option was involved. All I could
1190 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1191 running as the user.
1194 35. There was a buffer overflow vulnerability in the SPA authentication code
1195 (which came originally from the Samba project). I have added a test to the
1196 spa_base64_to_bits() function which I hope fixes it.
1199 36. The daemon start-up calls getloadavg() while still root for those OS that
1200 need the first call to be done as root, but it missed one case: when
1201 deliver_queue_load_max is set with deliver_drop_privilege. This is
1202 necessary for the benefit of the queue runner, because there is no re-exec
1203 when deliver_drop_privilege is set.
1205 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1206 was in use, was not putting the data itself into the right store pool;
1207 consequently, it could be overwritten for a subsequent message in the same
1208 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1211 38. Sometimes the final signoff response after QUIT could fail to get
1212 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1213 < 0 before doing a fflush(). This bug looks as though it goes back to the
1214 introduction of TLS in release 3.20, but "sometimes" must have been rare
1215 because the tests only now provoked it.
1221 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1222 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1223 Now it returns an integer. A number of calls were still expecting a T/F
1224 return. Fortuitously, in all cases, the tests worked in OK situations,
1225 which is the norm. However, things would have gone wrong on any write
1226 failures on the smtp file descriptor. This function is used when sending
1227 messages over SMTP and also when doing verify callouts.
1229 2. When Exim is called to do synchronous delivery of a locally submitted
1230 message (the -odf or -odi options), it no longer closes stderr before doing
1233 3. Implemented the mua_wrapper option.
1235 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1237 5. Implemented the functions header_remove(), header_testname(),
1238 header_add_at_position(), and receive_remove_recipient(), and exported them
1241 6. If an ACL "warn" statement specified the addition of headers, Exim already
1242 inserted X-ACL-Warn: at the start if there was no header name. However, it
1243 was not making this test for the second and subsequent header lines if
1244 there were newlines in the string. This meant that an invalid header could
1245 be inserted if Exim was badly configured.
1247 7. Allow an ACL "warn" statement to add header lines at the start or after all
1248 the Received: headers, as well as at the end.
1250 8. Added the rcpt_4xx retry error code.
1252 9. Added postmaster_mailfrom=xxx to callout verification option.
1254 10. Added mailfrom=xxxx to the callout verification option, for verify=
1257 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1258 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1260 12. Inserted some casts to stop certain compilers warning when using pointer
1261 differences as field lengths or precisions in printf-type calls (mostly
1262 affecting debugging statements).
1264 13. Added optional readline() support for -be (dynamically loaded).
1266 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1267 same clock tick as a message's arrival, so that its received time was the
1268 same as the "first fail" time on the retry record, and that message
1269 remained on the queue past the ultimate address timeout, every queue runner
1270 would try a delivery (because it was past the ultimate address timeout) but
1271 after another failure, the ultimate address timeout, which should have then
1272 bounced the address, did not kick in. This was a "< instead of <=" error;
1273 in most cases the first failure would have been in the next clock tick
1274 after the received time, and all would be well.
1276 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1277 being recognized when the domain list was tested by the match_domain
1278 condition in an expansion string.
1280 16. Added the ${str2b64: operator.
1282 17. Exim was always calling setrlimit() to set a large limit for the number of
1283 processes, without checking whether the existing limit was already
1284 adequate. (It did check for the limit on file descriptors.) Furthermore,
1285 errors from getrlimit() and setrlimit() were being ignored. Now they are
1286 logged to the main and panic logs, but Exim does carry on, to try to do its
1287 job under whatever limits there are.
1289 18. Imported PCRE 5.0.
1291 19. Trivial typo in log message " temporarily refused connection" (the leading
1294 20. If the log selector return_path_on_delivery was set and an address was
1295 redirected to /dev/null, the delivery process crashed because it assumed
1296 that a return path would always be set for a "successful" delivery. In this
1297 case, the whole delivery is bypassed as an optimization, and therefore no
1300 21. Internal re-arrangement: the function for sending a challenge and reading
1301 a response while authentication was assuming a zero-terminated challenge
1302 string. It's now changed to take a pointer and a length, to allow for
1303 binary data in such strings.
1305 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1307 23. Exim was not respecting finduser_retries when seeking the login of the
1308 uid under which it was called; it was always trying 10 times. (The default
1309 setting of finduser_retries is zero.) Also, it was sleeping after the final
1310 failure, which is pointless.
1312 24. Implemented tls_on_connect_ports.
1314 25. Implemented acl_smtp_predata.
1316 26. If the domain in control=submission is set empty, Exim assumes that the
1317 authenticated id is a complete email address when it generates From: or
1318 Sender: header lines.
1320 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1321 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1322 chown and chgrp in /bin and hostname in /usr/bin.
1324 28. Exim was keeping the "process log" file open after each use, just as it
1325 does for the main log. This opens the possibility of it remaining open for
1326 long periods when the USR1 signal hits a daemon. Occasional processlog
1327 errors were reported, that could have been caused by this. Anyway, it seems
1328 much more sensible not to leave this file open at all, so that is what now
1331 29. The long-running daemon process does not normally write to the log once it
1332 has entered its main loop, and it closes the log before doing so. This is
1333 so that log files can straightforwardly be renamed and moved. However,
1334 there are a couple of unusual error situations where the daemon does write
1335 log entries, and I had neglected to close the log afterwards.
1337 30. The text of an SMTP error response that was received during a remote
1338 delivery was being truncated at 512 bytes. This is too short for some of
1339 the long messages that one sometimes sees. I've increased the limit to
1342 31. It is now possible to make retry rules that apply only when a message has a
1343 specific sender, in particular, an empty sender.
1345 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1346 it possible to be selective about when SMTP synchronization is enforced.
1348 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1350 32. Implemented hosts_connection_nolog.
1352 33. Added an ACL for QUIT.
1354 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1357 35. Added mailbox_size and mailbox_filecount to appendfile.
1359 36. Added control = no_multiline_responses to ACLs.
1361 37. There was a bug in the logic of the code that waits for the clock to tick
1362 in the case where the clock went backwards by a substantial amount such
1363 that the microsecond fraction of "now" was more than the microsecond
1364 fraction of "then" (but the whole seconds number was less).
1366 38. Added support for the libradius Radius client library this is found on
1367 FreeBSD (previously only the radiusclient library was supported).
1373 1. When certain lookups returned multiple values in the form name=value, the
1374 quoting of the values was not always being done properly. Specifically:
1375 (a) If the value started with a double quote, but contained no whitespace,
1377 (b) If the value contained whitespace other than a space character (i.e.
1378 tabs or newlines or carriage returns) it was not quoted.
1379 This fix has been applied to the mysql and pgsql lookups by writing a
1380 separate quoting function and calling it from the lookup code. The fix
1381 should probably also be applied to nisplus, ibase and oracle lookups, but
1382 since I cannot test any of those, I have not disturbed their existing code.
1384 2. A hit in the callout cache for a specific address caused a log line with no
1385 reason for rejecting RCPT. Now it says "Previous (cached) callout
1386 verification failure".
1388 3. There was an off-by-one bug in the queryprogram router. An over-long
1389 return line was truncated at 256 instead of 255 characters, thereby
1390 overflowing its buffer with the terminating zero. As well as fixing this, I
1391 have increased the buffer size to 1024 (and made a note to document this).
1393 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1394 when Exim is waiting for an SMTP response from a remote server, Exim
1395 restarts its select() call on the socket, thereby resetting its timeout.
1396 This is not a problem when such interrupts are rare. Somebody set up a cron
1397 job to run exiwhat every 2 minutes, which is less than the normal select()
1398 timeout (5 or 10 minutes). This meant that the select() timeout never
1399 kicked in because it was always reset. I have fixed this by comparing the
1400 time when an interrupt arrives with the time at the start of the first call
1401 to select(). If more time than the timeout has elapsed, the interrupt is
1402 treated as a timeout.
1404 5. Some internal re-factoring in preparation for the addition of Sieve
1405 extensions (by MH). In particular, the "personal" test is moved to a
1406 separate function, and given an option for scanning Cc: and Bcc: (which is
1407 not set for Exim filters).
1409 6. When Exim created an email address using the login of the caller as the
1410 local part (e.g. when creating a From: or Sender: header line), it was not
1411 quoting the local part when it contained special characters such as @.
1413 7. Installed new OpenBSD configuration files.
1415 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1416 try to make them clearer.
1418 9. Callout options, other than the timeout value, were being ignored when
1419 verifying sender addresses in header lines. For example, when using
1421 verify = header_sender/callout=no_cache
1423 the cache was (incorrectly) being used.
1425 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1426 only the Cygwin environment.
1428 11. When return_path_on_delivery was set as a log selector, if different remote
1429 addresses in the same message used different return paths and parallel
1430 remote delivery occurred, the wrong values would sometimes be logged.
1431 (Whenever a remote delivery process finished, the return path value from
1432 the most recently started remote delivery process was logged.)
1434 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1435 lines when AUTH and/or TLS are in use. This is the "received protocol"
1436 field. Exim used to use "asmtp" for authenticated SMTP, without any
1437 indication (in the protocol name) for TLS use. Now it follows the RFC and
1438 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1439 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1440 names appear in log lines as well as in Received: header lines.
1442 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1443 extensions, and comparison tests, and to fix some bugs.
1445 14. Changes to the "personal" filter test:
1447 (1) The test was buggy in that it was just doing the equivalent of
1448 "contains" tests on header lines. For example, if a user's address was
1449 anne@some.where, the "personal" test would incorrectly be true for
1451 To: susanne@some.where
1453 This test is now done by extracting each address from the header in turn,
1454 and checking the entire address. Other tests that are part of "personal"
1455 are now done using regular expressions (for example, to check local parts
1456 of addresses in From: header lines).
1458 (2) The list of non-personal local parts in From: addresses has been
1459 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1460 taken from the Sieve specification recommendations.
1462 (3) If the message contains any header line starting with "List-" it is
1463 treated as non-personal.
1465 (4) The test for "circular" in the Subject: header line has been removed
1466 because it now seems ill-conceived.
1468 15. Minor typos in src/EDITME comments corrected.
1470 16. Installed latest exipick from John Jetmore.
1472 17. If headers_add on a router specified a text string that was too long for
1473 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1474 of string_sprintf() is now avoided.
1476 18. $message_body_size was not set (it was always zero) when running the DATA
1477 ACL and the local_scan() function.
1479 19. For the "mail" command in an Exim filter, no default was being set for
1480 the once_repeat time, causing a random time value to be used if "once" was
1481 specified. (If the value happened to be <= 0, no repeat happened.) The
1482 default is now 0s, meaning "never repeat". The "vacation" command was OK
1483 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1484 (I found it when inspecting the code).
1486 20. There is now an overall timeout for performing a callout verification. It
1487 defaults to 4 times the callout timeout, which applies to individual SMTP
1488 commands during the callout. The overall timeout applies when there is more
1489 than one host that can be tried. The timeout is checked before trying the
1490 next host. This prevents very long delays if there are a large number of
1491 hosts and all are timing out (e.g. when the network connections are timing
1492 out). The value of the overall timeout can be changed by specifying an
1493 additional sub-option for "callout", called "maxwait". For example:
1495 verify = sender/callout=5s,maxwait=20s
1497 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1498 to the end before writing, but this should make it even safer).
1500 22. Exim was forgetting that it had advertised PIPELINING for the second and
1501 subsequent messages on an SMTP connection. It was also not resetting its
1502 memory on STARTTLS and an internal HELO.
1504 23. When Exim logs an SMTP synchronization error within a session, it now
1505 records whether PIPELINING has been advertised or not.
1507 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1508 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1509 rather than long int.
1511 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1513 26. Added the never_mail option to autoreply.
1519 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1520 crash if the getsockname() call failed; this can happen if a connection is
1521 closed very soon after it is established. The problem was simply in the
1522 order in which certain operations were done, causing Exim to try to write
1523 to the SMTP stream before it had set up the file descriptor. The bug has
1524 been fixed by making things happen in the correct order.
1530 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1531 before the connection was closed, thus losing the rejection response.
1533 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1534 some early Solaris releases, but causes trouble in current releases where
1535 socklen_t is defined.
1537 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1540 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1543 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1544 that releasing the top of it at the end releases what was used for sub-
1545 expansions (unless the block got too big). However, discard this block if
1546 the first thing is a variable or header, so that we can use its block when
1547 it is dynamic (useful for very large $message_headers, for example).
1549 6. Lookups now cache *every* query, not just the most recent. A new, separate
1550 store pool is used for this. It can be recovered when all lookup caches are
1551 flushed. Lookups now release memory at the end of their result strings.
1552 This has involved some general refactoring of the lookup sources.
1554 7. Some code has been added to the store_xxx() functions to reduce the amount
1555 of flapping under certain conditions.
1557 8. log_incoming_interface used to affect only the <= reception log lines. Now
1558 it causes the local interface and port to be added to several more SMTP log
1559 lines, for example "SMTP connection from", and rejection lines.
1561 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1563 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1565 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1566 could be overwritten at the end of the current message (or the start of a
1567 new message if it was set in a HELO ACL). The value is now preserved for
1568 the duration of the SMTP connection.
1570 12. If a transport had a headers_rewrite setting, and a matching header line
1571 contained an unqualified address, that address was qualified, even if it
1572 did not match any rewriting rules. The underlying bug was that the values
1573 of the flags that permit the existence of unqualified sender and recipient
1574 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1575 non-local messages, and by -bnq for local messages) were not being
1576 preserved with the message after it was received.
1578 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1579 "next input=" as part of the text comprising the host identity instead of
1580 the correct text. The code was using the same buffer for two different
1581 strings. However, depending on which order the printing function evaluated
1582 its arguments, the bug did not always show up. Under Linux, for example, my
1583 test suite worked just fine.
1585 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1586 doesn't work with some older versions of Perl. It has been changed to "my",
1587 which in any case is probably the better facility to use.
1589 15. A really picky compiler found some instances of statements for creating
1590 error messages that either had too many or two few arguments for the format
1593 16. The size of the buffer for calls to the DNS resolver has been increased
1594 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1595 for addresses that have a lot of PTR records. This alleviates a problem; it
1596 does not fully solve it.
1598 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1599 buffer now truncates the list and logs the incident, which is the same
1600 action as happens when Exim is looking up a host name and its aliases.
1601 Previously in this situation something unpredictable would happen;
1602 sometimes it was "internal error: store_reset failed".
1604 18. If a server dropped the connection unexpectedly when an Exim client was
1605 using GnuTLS and trying to read a response, the client delivery process
1606 crashed while trying to generate an error log message.
1608 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1609 string, for example:
1611 warn message = H1: something\nH2: something
1613 the text was added as a single header line from Exim's point of view
1614 though it ended up OK in the delivered message. However, searching for the
1615 second and subsequent header lines using $h_h2: did not work. This has been
1616 fixed. Similarly, if a system filter added multiple headers in this way,
1617 the routers could not see them.
1619 20. Expanded the error message when iplsearch is called with an invalid key to
1620 suggest using net-iplsearch in a host list.
1622 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1623 ACLs are no longer actually imposed (and a message to that effect is
1626 22. If a "gecos" field in a passwd entry contained escaped characters, in
1627 particular, if it contained a \" sequence, Exim got it wrong when building
1628 a From: or a Sender: header from that name. A second bug also caused
1629 incorrect handling when an unquoted " was present following a character
1630 that needed quoting.
1632 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1633 was not being matched caselessly.
1635 24. Arranged for all hyphens in the exim.8 source to be escaped with
1638 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
1639 properly. Recipient callout cache records were still being keyed to include
1640 the sender, even when use_sender was set false. This led to far more
1641 callouts that were necessary. The sender is no longer included in the key
1642 when use_sender is false.
1644 26. Added "control = submission" modifier to ACLs.
1646 27. Added the ${base62d: operator to decode base 62 numbers.
1648 28. dnsdb lookups can now access SRV records.
1650 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
1651 the configuration file.
1653 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
1654 (-v) mode. This makes the output for a verbose queue run more intelligible.
1656 31. Added a use_postmaster feature to recipient callouts.
1658 32. Added the $body_zerocount variable, containing the number of binary zero
1659 bytes in the message body.
1661 33. The time of last modification of the "new" subdirectory is now used as the
1662 "mailbox time last read" when there is a quota error for a maildir
1665 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
1667 35. Added +ignore_unknown as a special item in host lists.
1669 36. Code for decoding IPv6 addresses in host lists is now included, even if
1670 IPv6 support is not being compiled. This fixes a bug in which an IPv6
1671 address was recognized as an IP address, but was then not correctly decoded
1672 into binary, causing unexpected and incorrect effects when compared with
1679 1. Very minor rewording of debugging text in manualroute to say "list of
1680 hosts" instead of "hostlist".
1682 2. If verify=header_syntax was set, and a header line with an unqualified
1683 address (no domain) and a large number of spaces between the end of the
1684 name and the colon was received, the reception process suffered a buffer
1685 overflow, and (when I tested it) crashed. This was caused by some obsolete
1686 code that should have been removed. The fix is to remove it!
1688 3. When running in the test harness, delay a bit after writing a bounce
1689 message to get a bit more predictability in the log output.
1691 4. Added a call to search_tidyup() just before forking a reception process. In
1692 theory, someone could use a lookup in the expansion of smtp_accept_max_
1693 per_host which, without the tidyup, could leave open a database connection.
1695 5. Added the variables $recipient_data and $sender_data which get set from a
1696 lookup success in an ACL "recipients" or "senders" condition, or a router
1697 "senders" option, similar to $domain_data and $local_part_data.
1699 6. Moved the writing of debug_print from before to after the "senders" test
1702 7. Change 4.31/66 (moving the time when the Received: is generated) caused
1703 problems for message scanning, either using a data ACL, or using
1704 local_scan() because the Received: header was not generated till after they
1705 were called (in order to set the time as the time of reception completion).
1706 I have revised the way this works. The header is now generated after the
1707 body is received, but before the ACL or local_scan() are called. After they
1708 are run, the timestamp in the header is updated.
1714 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
1715 before starting a queue runner without re-exec. This happened only when
1716 deliver_drop_privilege was set or when the Exim user was set to root. The
1717 effect of the bug was that timeouts during subsequent deliveries caused
1718 crashes instead of being properly handled. The handler is now left at its
1719 default (and expected) setting.
1721 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
1722 message, again when deliver_drop_privilege is set or Exim is run as root.
1723 The bug described in (1) was not present in this case, but the tidying up
1724 of the other signals was missing. I have made the two cases consistent.
1726 3. The ignore_target_hosts setting on a manualroute router was being ignored
1727 for hosts that were looked up using the /MX notation.
1729 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
1732 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
1733 operated on the sender address. After changing the $sender_address to <>
1734 for the sender address verify, Exim was re-instated it as the original
1735 (before rewriting) address, but remembering that it had rewritten it, so it
1736 wasn't rewriting it again. This bug also had the effect of breaking the
1737 sender address verification caching when the sender address was rewritten.
1739 6. The ignore_target_hosts option was being ignored by the ipliteral router.
1740 This has been changed so that if the ip literal address matches
1741 ignore_target_hosts, the router declines.
1743 7. Added expansion conditions match_domain, match_address, and match_local_
1744 part (NOT match_host).
1746 8. The placeholder for the Received: header didn't have a length field set.
1748 9. Added code to Exim itself and to exim_lock to test for a specific race
1749 condition that could lead to file corruption when using MBX delivery. The
1750 issue is with the lockfile that is created in /tmp. If this file is removed
1751 after a process has opened it but before that process has acquired a lock,
1752 there is the potential for a second process to recreate the file and also
1753 acquire a lock. This could lead to two Exim processes writing to the file
1754 at the same time. The added code performs the same test as UW imapd; it
1755 checks after acquiring the lock that its file descriptor still refers to
1756 the same named file.
1758 10. The buffer for building added header lines was of fixed size, 8192 bytes.
1759 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
1762 11. Added the smtp_active_hostname option. If used, this will typically be made
1763 to depend on the incoming interface address. Because $interface_address is
1764 not set up until the daemon has forked a reception process, error responses
1765 that can happen earlier (such as "too many connections") no longer contain
1768 12. If an expansion in a condition on a "warn" statement fails because a lookup
1769 defers, the "warn" statement is abandoned, and the next ACL statement is
1770 processed. Previously this caused the whole ACL to be aborted.
1772 13. Added the iplsearch lookup type.
1774 14. Added ident_timeout as a log selector.
1776 15. Added tls_certificate_verified as a log selector.
1778 16. Added a global option tls_require_ciphers (compare the smtp transport
1779 option of the same name). This controls incoming TLS connections.
1781 17. I finally figured out how to make tls_require_ciphers do a similar thing
1782 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
1783 before starting the TLS session.
1785 18. Tabs are now shown as \t in -bP output.
1787 19. If the log selector return_path_on_delivery was set, Exim crashed when
1788 bouncing a message because it had too many Received: header lines.
1790 20. If two routers both had headers_remove settings, and the first one included
1791 a superfluous trailing colon, the final name in the first list and the
1792 first name in the second list were incorrectly joined into one item (with a
1793 colon in the middle).
1799 1. Added -C and -D options to the exinext utility, mainly to make it easier
1800 to include in the automated testing, but these could be helpful when
1801 multiple configurations are in use.
1803 2. The exinext utility was not formatting the output nicely when there was
1804 an alternate port involved in the retry record key, nor when there was a
1805 message id as well (for retries that were specific to a specific message
1806 and a specific host). It was also confused by IPv6 addresses, because of
1807 the additional colons they contain. I have fixed the IPv4 problem, and
1808 patched it up to do a reasonable job for IPv6.
1810 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
1811 delivery, the log line now contains "pipelined" if PIPELINING was used.
1813 4. An SMTP transport process used to panic and die if the bind() call to set
1814 an explicit outgoing interface failed. This has been changed; it is now
1815 treated in the same way as a connect() failure.
1817 5. A reference to $sender_host_name in the part of a conditional expansion
1818 that was being skipped was still causing a DNS lookup. This no longer
1821 6. The def: expansion condition was not recognizing references to header lines
1822 that used bh_ and bheader_.
1824 7. Added the _cache feature to named lists.
1826 8. The code for checking quota_filecount in the appendfile transport was
1827 allowing one more file than it should have been.
1829 9. For compatibility with Sendmail, the command line option
1837 and sets the incoming protocol and host name (for trusted callers). The
1838 host name and its colon can be omitted when only the protocol is to be set.
1839 Note the Exim already has two private options, -pd and -ps, that refer to
1840 embedded Perl. It is therefore impossible to set a protocol value of "d" or
1841 "s", but I don't think that's a major issue.
1843 10. A number of refactoring changes to the code, none of which should affect
1846 (a) The number of logging options was getting close to filling up the
1847 32-bit word that was used as a bit map. I have split them into two classes:
1848 those that are passed in the argument to log_write(), and those that are
1849 only ever tested independently outside of that function. These are now in
1850 separate 32-bit words, so there is plenty of room for expansion again.
1851 There is no change in the user interface or the logging behaviour.
1853 (b) When building, for example, log lines, the code previously used a
1854 macro that called string_cat() twice, in order to add two strings. This is
1855 not really sufficiently general. Furthermore, there was one instance where
1856 it was actually wrong because one of the argument was used twice, and in
1857 one call a function was used. (As it happened, calling the function twice
1858 did not affect the overall behaviour.) The macro has been replaced by a
1859 function that can join an arbitrary number of extra strings onto a growing
1862 (c) The code for expansion conditions now uses a table and a binary chop
1863 instead of a serial search (which was left over from when there were very
1864 few conditions). Also, it now recognizes conditions like "pam" even when
1865 the relevant support is not compiled in: a suitably worded error message is
1866 given if an attempt is made to use such a condition.
1868 11. Added ${time_interval:xxxxx}.
1870 12. A bug was causing one of the ddress fields not to be passed back correctly
1871 from remote delivery subprocesses. The field in question was not being
1872 subsequently used, so this caused to problems in practice.
1874 13. Added new log selectors queue_time and deliver_time.
1876 14. Might have fixed a bug in maildirsizefile handling that threw up
1877 "unexpected character" debug warnings, and recalculated the data
1878 unnecessarily. In any case, I expanded the warning message to give more
1881 15. Added the message "Restricted characters in address" to the statements in
1882 the default ACL that block characters like @ and % in local parts.
1884 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
1885 Three changes have been made:
1887 (a) There was a serious bug; a negative response to MAIL caused the whole
1888 recipient domain to be cached as invalid, thereby blocking all messages
1889 to all local parts at the same domain, from all senders. This bug has
1890 been fixed. The domain is no longer cached after a negative response to
1891 MAIL if the sender used is not empty.
1893 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
1896 (c) A new callout option, "use_sender" has been added for people who want
1897 the modified behaviour.
1903 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
1906 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
1907 indeed breaks things for older releases.
1909 3. Added additional logging to the case where there is a problem reading data
1910 from a filter that is running in a subprocess using a pipe, in order to
1911 try to track down a specific problem.
1913 4. Testing facility fudge: when running in the test harness and attempting
1914 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
1915 getting "No route to host". Convert this to a timeout.
1917 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
1920 6. Some OS don't have socklen_t but use size_t instead. This affects the
1921 fifth argument of getsockopt() amongst other things. This is now
1922 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
1923 can be set for individual OS. I have set it for SunOS5, OSF1, and
1924 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
1925 some earlier ones do not.
1927 7. Change 4.30/15 was not doing the test caselessly.
1929 8. The standard form for an IPv6 address literal was being rejected by address
1930 parsing in, for example, MAIL and RCPT commands. An example of this kind of
1931 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
1932 this, as well as the form without the "IPv6" on the front (but only when
1933 address literals are enabled, of course).
1935 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
1937 10. Exim crashed if a message with an empty sender address specified by -f
1938 encountered a router with an errors_to setting. This could be provoked only
1939 by a command such as
1943 where an empty string was supplied; "<>" did not hit this bug.
1945 11. Installed PCRE release 4.5.
1947 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
1948 remained set. It is now erased.
1950 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
1951 times from message ids (which are base 36 rather than the normal 62).
1953 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
1954 were being counted as actual protocol errors, and logged if the log
1955 selector +smtp_protocol_error was set. One cannot be perfect in this test,
1956 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
1957 and DATA following a set of rejected RCPTs do not count as protocol errors.
1958 In other words, Exim assumes they were pipelined, though this may not
1959 actually be the case. Of course, in all cases the client gets an
1960 appropriate error code.
1962 15. If a lookup fails in an ACL condition, a message about the failure may
1963 be available; it is used if testing the ACL cannot continue, because most
1964 such messages specify what the cause of the deferral is. However, some
1965 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
1966 that caused an old message to be retained and used if a later statement
1967 caused a defer, replacing the real cause of the deferral.
1969 16. If an IP address had so many PTR records that the DNS lookup buffer
1970 was not large enough to hold them, Exim could crash while trying to process
1971 the truncated data. It now detects and logs this case.
1973 17. Further to 4.21/58, another change has been made: if (and only if) the
1974 first line of a message (the first header line) ends with CRLF, a bare LF
1975 in a subsequent header line has a space inserted after it, so as not to
1976 terminate the header.
1978 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
1979 unnecessarily, used atoi() instead of strtol(), and didn't check the
1980 termination when getting file sizes from file names by regex.
1982 19. Completely re-implemented the support for maildirsize files, in the light
1983 of a number of problems with the previous contributed implementation
1984 (4.30/29). In particular:
1986 . If the quota is zero, the maildirsize file is maintained, but no quota is
1989 . If the maildir directory does not exist, it is created before any attempt
1990 to write a maildirsize file.
1992 . The quota value in the file is just a cache; if the quota is changed in
1993 the transport, the new value overrides.
1995 . A regular expression is available for excluding directories from the
1998 20. The autoreply transport checks the characters in options that define the
1999 message's headers; it allows continued headers, but it was checking with
2000 isspace() after an embedded newline instead of explicitly looking for a
2003 21. If all the "regular" hosts to which an address was routed had passed their
2004 expiry times, and had not reached their retry times, the address was
2005 bounced, even if fallback hosts were defined. Now Exim should go on to try
2008 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2009 equivalent code in the SMTP transport. Some hosts send humungous responses
2010 to HELO/EHLO, more than 1024 it seems.
2012 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2013 gives compiler warnings in some environments. I've now done it "properly",
2016 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2017 (because of problems with the built-in one) was declared to return uschar *
2018 instead of char *, causing compiler failure.
2020 25. Fixed a file descriptor leak when processing alias/forward files.
2022 26. Fixed a minor format string issue in dbfn.c.
2024 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2026 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2027 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2030 29. When a sender address is verified, it is cached, to save repeating the test
2031 when there is more than one recipient in a message. However, when the
2032 verification involves a callout, it is possible for different callout
2033 options to be set for different recipients. It is too complicated to keep
2034 track of this in the cache, so now Exim always runs a verification when a
2035 callout is required, relying on the callout cache for the optimization.
2036 The overhead is duplication of the address routing, but this should not be
2039 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2040 to be verified with callout=postmaster, and the main callout worked but the
2041 postmaster check failed, the verification correctly failed. However, if a
2042 subsequent RCPT command asked for sender verification *without* the
2043 postmaster check, incorrect caching caused this verification also to fail,
2046 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2047 it was not caching the DNS options (qualify_single, search_parents) that
2048 were used when the lookup failed. A subsequent lookup with different
2049 options therefore always gave the same answer, though there were cases
2050 where it should not have. (Example: a "domains = !$mx_any" option on a
2051 dnslookup router: the "domains" option is always processed without any
2052 widening, but the router might have qualify_single set.) Now Exim uses the
2053 cached value only when the same options are set.
2055 32. Added John Jetmore's "exipick" utility to the distribution.
2057 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2058 than a timeout (e.g. a certificate is required, and is not provided), an
2059 Exim server now closes the connection immediately. Previously it waited for
2060 the client to close - but if the client is SSL, it seems that they each
2061 wait for each other, leading to a delay before one of them times out.
2063 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2064 maintained 0.8.x compatibility because I don't think many are using it, and
2065 it is clearly obsolete.
2067 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2070 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2071 client certificate was expired. A simple patch fixes this, though I don't
2072 understand the full logic of why the verify callback is called multiple
2075 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2076 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2077 which causes problems with some clients (such as the Certicom SSL Plus
2078 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2079 disables the coutermeasure allowing Eudora to connect."
2081 38. Exim was not checking that a write() to a log file succeeded. This could
2082 lead to Bad Things if a log got too big, in particular if it hit a file
2083 size limit. Exim now panics and dies if it cannot write to a log file, just
2084 as it does if it cannot open a log file.
2086 39. Modified OS/Makefile-Linux so that it now contains
2088 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2090 The two -D definitions ensure that Exim is compiled with large file
2091 support, which makes it possible to handle log files that are bigger than
2094 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2095 instance) a domain was checked against a named list that involved a lookup,
2096 causing $domain_data to be set, then another domain was checked against the
2097 same list, then the first domain was re-checked, the value of $domain_data
2098 after the final check could be wrong. In particular, if the second check
2099 failed, it could be set empty. This bug probably also applied to
2102 41. The strip_trailing_dot option was not being applied to the address given
2103 with the -f command-line option.
2105 42. The code for reading a message's header from the spool was incrementing
2106 $received_count, but never initializing it. This meant that the value was
2107 incorrect (doubled) while delivering a message in the same process in which
2108 it was received. In the most common configuration of Exim, this never
2109 happens - a fresh exec is done - but it can happen when
2110 deliver_drop_privilege is set.
2112 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2113 it now includes up to 150 characters of the unexpected data in the log
2116 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2117 and building data strings. The size of both of these buffers was 10 000
2118 bytes - far larger than anybody would *ever* want, thought I. Needless to
2119 say, somebody hit the limit. I have increased the maximum line length to
2120 20 000 and the maximum data length of concatenated lines to 100 000. I have
2121 also fixed two bugs, because there was no checking on these buffers. Tsk,
2122 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2123 buffer is too small.
2125 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2126 lsearch lookups. Now it does.
2128 46. When parsing a route_list item in a manualroute router, a fixed-length
2129 buffer was used for the list of hosts. I made this 1024 bytes long,
2130 thinking that nobody would ever have a list of hosts that long. Wrong.
2131 Somebody had a whole pile of complicated expansion conditions, and the
2132 string was silently truncated, leading to an expansion error. It turns out
2133 that it is easier to change to an unlimited length (owing to other changes
2134 that have happened since this code was originally written) than to build
2135 structure for giving a limitation error. The length of the item that
2136 expands into the list of hosts is now unlimited.
2138 47. The lsearch lookup could not handle data where the length of text line was
2139 more than 4095 characters. Such lines were truncated, leading to shortened
2140 data being returned. It should now handle lines of any length.
2142 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2143 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2146 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2147 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2148 no longer happens in comment lines. A list of replacements is now placed
2149 at the head of all of the source files, except those whose only change is
2150 to replace PERL_COMMAND in the very first #! line.
2152 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2153 messages on the queue, with a bottom-up merge sort, using code contributed
2154 by Michael Haardt. This should make operations like -bp somewhat faster on
2155 large queues. It won't affect queue runners, except when queue_run_in_order
2158 51. Installed eximstats 1.31 in the distribution.
2160 52. Added support for SRV lookups to the dnslookup router.
2162 53. If an ACL referred to $message_body or $message_body_end, the value was not
2163 reset for any messages that followed in the same SMTP session.
2165 54. The store-handling optimization for building very long strings was not
2166 differentiating between the different store pools. I don't think this
2167 actually made any difference in practice, but I've tidied it.
2169 55. While running the routers to verify a sender address, $sender_address
2170 was still set to the sender address. This is wrong, because when routing to
2171 send a bounce to the sender, it would be empty. Therefore, I have changed
2172 it so that, while verifying a sender address, $sender_address is set to <>.
2173 (There is no change to what happens when verifying a recipient address.)
2175 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2176 target A or AAAA records (if not already returned) without resetting the
2177 qualify_single or search_parents options of the DNS resolver. These are
2178 inappropriate in this case because the targets of MX and SRV records must
2179 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2180 target that, when qualified, matched something in the local domain. These
2181 two options are now turned off when doing these lookups.
2183 57. It seems that at least some releases of Reiserfs (which does not have the
2184 concept of a fixed number of inodes) returns zero and not -1 for the
2185 number of available inodes. This interacted badly with check_spool_inodes,
2186 which assumed that -1 was the "no such thing" setting. What I have done is
2187 to check that the total number of inodes is greater than zero before doing
2188 the test of how many are available.
2190 58. When a "warn" ACL statement has a log_message modifier, the message is
2191 remembered, and not repeated. This is to avoid a lot of repetition when a
2192 message has many recipients that cause the same warning to be written.
2193 Howewer, Exim was preserving the list of already written lines for an
2194 entire SMTP session, which doesn't seem right. The memory is now reset if a
2195 new message is started.
2197 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2198 debugging output unless log_rewrite was also set.
2200 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2201 of (int)(handle) when we know that handle contains (void *)(-1).
2203 61. The Exim daemon panic-logs an error return when it closes the incoming
2204 connection. However "connection reset by peer" seems to be common, and
2205 isn't really an error worthy of noting specially, so that particular error
2208 62. When Exim is trying to find all the local interfaces, it used to panic and
2209 die if the ioctl to get the interface flags failed. However, it seems that
2210 on at least one OS (Solaris 9) it is possible to have an interface that is
2211 included in the list of interfaces, but for which you get a failure error
2212 for this call. This happens when the interface is not "plumbed" into a
2213 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2214 failure of the "get flags" call assumes that the interface is down.
2216 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2217 makes life easier for people who are doing arithmetic on fields extracted
2218 from dates, where you often get leading zeros that should not be
2219 interpreted as octal.
2221 64. Added qualify_domain to the redirect router, to override the global
2224 65. If a pathologically long header line contained very many addresses (the
2225 report of this problem mentioned 10 000) and each of them was rewritten,
2226 Exim could use up a very large amount of memory. (It kept on making new
2227 copies of the header line as it rewrote, and never released the old ones.)
2228 At the expense of a bit more processing, the header rewriting function has
2229 been changed so that it no longer eats memory in this way.
2231 66. The generation of the Received: header has been moved from the time that a
2232 message starts to be received, to the time that it finishes. The timestamp
2233 in the Received: header should now be very close to that of the <= log
2234 line. There are two side-effects of this change:
2236 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2237 logged header lines no longer include the local Received: line, because
2238 it has not yet been created. The same applies to a copy of the message
2239 that is returned to a non-SMTP sender when a message is rejected.
2241 (b) When a filter file is tested using -bf, no additional Received: header
2242 is added to the test message. After some thought, I decided that this
2245 This change does not affect the value of $received_for. It is still set
2246 after address rewriting, but before local_scan() is called.
2248 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2250 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2251 gave an unhelpful panic error message, and a defer error. I have managed to
2252 change this behaviour so that it now rejects any supplied certificate,
2253 which seems right, as the list of acceptable certificates is empty.
2255 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2256 gave an unhelpful defer error. I have not managed to make this reject any
2257 supplied certificates, but the error message it gives is "no certificate
2258 supplied", which is not helpful.
2260 70. exigrep's output now also includes lines that are not associated with any
2261 message, but which match the given pattern. Implemented by a patch from
2262 Martin Sluka, which also tidied up the Perl a bit.
2264 71. Recipient callout verification, like sender verification, was using <> in
2265 the MAIL FROM command. This isn't really the right thing, since the actual
2266 sender may affect whether the remote host accepts the recipient or not. I
2267 have changed it to use the actual sender in the callout; this means that
2268 the cache record is now keyed on a recipient/sender pair, not just the
2269 recipient address. There doesn't seem to be a real danger of callout loops,
2270 since a callout by the remote host to check the sender would use <>.
2271 [SEE ABOVE: changed after hitting problems.]
2273 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2274 temporary errors. However, in the case of such a code being given after
2275 the end of a data transmission (i.e. after ".") Exim was failing to write
2276 a retry record for the message. (Yes, there was some broken host that was
2277 actually sending 8xx at this point.)
2279 73. An unknown lookup type in a host list could cause Exim to panic-die when
2280 the list was checked. (An example that provoked this was putting <; in the
2281 middle of a list instead of at the start.) If this happened during a DATA
2282 ACL check, a -D file could be left lying around. This kind of configuration
2283 error no longer causes Exim to die; instead it causes a defer errror. The
2284 incident is still logged to the main and panic logs.
2286 74. Buglet left over from Exim 3 conversion. The message "too many messages
2287 in one connection" was written to the rejectlog but not the mainlog, except
2288 when address rewriting (yes!) was being logged.
2290 75. Added write_rejectlog option.
2292 76. When a system filter was run not as root (that is, when system_filter_user
2293 was set), the values of the $n variables were not being returned to the
2294 main process; thus, they were not subsequently available in the $sn
2297 77. Added +return_path_on_delivery log selector.
2299 78. A connection timeout was being treated differently from recipients deferred
2300 when testing hosts_max_try with a message that was older than the host's
2301 retry timeout. (The host should not be counted, thus allowing all hosts to
2302 be tried at least once before bouncing.) This may have been the cause of an
2303 occasionally reported bug whereby a message would remain on the queue
2304 longer than the retry timeout, but would be bounced if a delivery was
2305 forced. I say "may" because I never totally pinned down the problem;
2306 setting up timeout/retry tests is difficult. See also the next item.
2308 79. The ultimate address timeout was not being applied to errors that involved
2309 a combination of host plus message (for example, a timeout on a MAIL
2310 command). When an address resolved to a number of possible hosts, and they
2311 were not all tried for each delivery (e.g. because of hosts_max_try), a
2312 message could remain on the queue longer than the retry timeout.
2314 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2317 81. Fixed an obscure SMTP outgoing bug which required at least the following
2318 conditions: (a) there was another message waiting for the same server;
2319 (b) the server returned 5xx to all RCPT commands in the first message so
2320 that the message was not completed; (c) the server dropped the connection
2321 or gave a negative response to the RSET that Exim sends to abort the
2322 transaction. The observed case was a dropped connection after DATA that had
2323 been sent in pipelining mode. That is, the server had advertised PIPELINING
2324 but was not implementing it correctly. The effect of the bug was incorrect
2325 behaviour, such as trying another host, and this could lead to a crash.
2331 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2332 and daemon.c were passed as pointers to ints; they should have been
2333 pointers to socklen_t variables (which are typically unsigned ints).
2335 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2338 3. Fixed a really odd bug that affected only the testing scheme; patching a
2339 certain fixed string in the binary changed the value of another string that
2340 happened to be identical to the end of the original first string.
2342 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2343 name", it returns that address as the IP address. On some operating
2344 systems (e.g. Solaris), it also passes back the IP address string as the
2345 "host name". However, on others (e.g. Linux), it passes back an empty
2346 string. Exim wasn't checking for this, and was changing the host name to an
2347 empty string, assuming it had been canonicized.
2349 5. Although rare, it is permitted to have more than one PTR record for a given
2350 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2351 all the names associated with an address, because they do in Solaris.
2352 However, it seems that they do not in Linux for data that comes from the
2353 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2354 I found this out when I moved to a new Linux workstation and tried to run
2355 the Exim test suite.
2357 To get round this problem I have changed the code so that it now does its
2358 own call to the DNS to look up PTR records when searching for a host name.
2359 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2360 addresses that are only in /etc/hosts are still found.
2362 This behaviour is, however, controlled by an option called host_lookup_
2363 order, which defaults to "bydns:byaddr". If people want to use the other
2364 order, or indeed, just use one or the other means of lookup, they can
2365 specify it in this variable.
2367 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2368 some operating systems, this comes back from gethostbyaddr() as an empty
2369 string, and this is what Exim used to test for. However, it seems that in
2370 other systems, "." is yielded. Exim now tests for this case too.
2372 7. The values of check_spool_space and check_log_space are now held internally
2373 as a number of kilobytes instead of an absolute number of bytes. If a
2374 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2375 kilobyte. This means that much larger values can be stored.
2377 8. Exim monitor: an attempt to get the action menu when not actually pointing
2378 at a message produces an empty menu entitled "No message selected". This
2379 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2380 no entries in it ("Shell widget menu has zero width and/or height"). So I
2381 have added a single, blank menu entry in this case.
2383 9. Added ${quote_local_part.
2385 10. MIME decoding is now applied to the contents of Subject: header lines when
2388 11. Now that a reference to $sender_host_address automatically causes a reverse
2389 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2390 host lookup before query-style lookups in lists that might use this
2391 variable. This has therefore been abolished, and the "net-" prefix is no
2392 longer necessary for query-style lookups.
2394 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2395 have been a typo for LFLAGS, so it has been changed.
2397 13. The install script calls Exim with "-C /dev/null" in order to find the
2398 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2399 to be output. Howeve, since Exim outputs its version number before the
2400 error, it didn't break the script. It just looked ugly. I fixed this by
2401 always allowing "-C /dev/null" if the caller is root.
2403 14. Ignore overlarge ACL variable number when reading spool file - insurance
2404 against a later release with more variables having written the file.
2406 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2407 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2408 this, as well as the form without the "IPv6" on the front.
2410 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2411 OS/Makefile-Darwin file.
2413 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2414 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2416 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2417 variable. However, some OS use ints for inodes. I've added cast to long int
2418 to get rid of the compiler warning.
2420 19. I had forgotten to lock out "/../" in configuration file names when
2421 ALT_CONFIG_PREFIX was set.
2423 20. Routers used for verification do not need to specify transports. However,
2424 if such a router generated a host list, and callout was configured, Exim
2425 crashed, because it could not find a port number from the (non-existent)
2426 transport. It now assumes port 25 in this circumstance.
2428 21. Added the -t option to exigrep.
2430 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2431 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2432 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2433 them is set, LOOKUP_LSEARCH is forced.
2435 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2437 24. Added sender and host information to the "rejected by local_scan()" log
2438 line; previously there was no indication of these.
2440 25. Added .include_if_exists.
2442 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2443 Linux. It turns out that not all file systems support this. Apparently some
2444 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2445 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2446 not supported on the file descriptor, is now ignored when Exim is trying to
2447 sync a directory. This applies only to Linux.
2449 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2451 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2452 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2455 29. Added support for maildirsize files from supplied patch (modified a bit).
2457 30. The use of :fail: followed by an empty string could lead Exim to respond to
2458 sender verification failures with (e.g.):
2460 550 Verification failed for <xxx>
2461 550 Sender verify failed
2463 where the first response line was missing the '-' that indicates it is not
2464 the final line of the response.
2466 31. The loop for finding the name of the user that called Exim had a hardwired
2467 limit of 10; it now uses the value of finduser_retries, which is used for
2468 all other user lookups.
2470 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2473 33. Exim was neglecting to zero errno before one call of strtol() when
2474 expanding a string and expecting an integer value. On some systems this
2475 resulted in spurious "integer overflow" errors. Also, it was casting the
2476 result into an int without checking.
2478 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2479 did not work. The code looks as if it has *never* worked, though it appears
2480 to have been documented since at least releast 1.62. I have made it work.
2482 35. The "timeout_DNS" error in retry rules, also documented since at least
2483 1.62, also never worked. As it isn't clear exactly what this means, and
2484 clearly it isn't a major issue, I have abolished the feature by treating it
2485 as "timeout", and writing a warning to the main and panic logs.
2487 36. The display of retry rules for -brt wasn't always showing the error code
2490 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2491 timeout_connect_A, timeout_connect_MX.
2493 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2494 to the empty sender.
2496 39. The daemon was not analysing the content of -oX till after it had closed
2497 stderr and disconnected from the controlling terminal. This meant that any
2498 syntax errors were only noted on the panic log, and the return code from
2499 the command was 0. By re-arranging the code a little, I've made the
2500 decoding happen first, so such errors now appear on stderr, and the return
2501 code is 1. However, the actual setting up of the sockets still happens in
2502 the disconnected process, so errors there are still only recorded on the
2505 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2506 connections (as happens on some IP stacks) was logged at start up time as
2507 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2508 it from "IPv6 and IPv4", which means that two separate sockets are being
2511 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2512 says whether AF_INET or AF_INET6 was passed as an argument.
2514 42. Exiwhat output was messed up when time zones were included in log
2517 43. Exiwhat now gives more information about the daemon's listening ports,
2518 and whether -tls-on-connect was used.
2520 44. The "port" option of the smtp transport is now expanded.
2522 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2523 silently ignored. Now an error message is written to the main and panic
2526 46. There's a new ACL modifier called "logwrite" which writes to a log file
2527 as soon as it is encountered.
2529 47. Added $local_user_uid and $local_user_gid at routing time.
2531 48. Exim crashed when trying to verify a sender address that was being
2534 49. Exim was recognizing only a space character after ".include". It now also
2535 recognizes a tab character.
2537 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2538 extracting the relevant information from the specification. The man page no
2539 longer contains scrambled data for the -d option, and I've added a section
2540 at the front about calling Exim under different names.
2542 51. Added "extra_headers" argument to the "mail" command in filter files.
2544 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2547 53. Installed eximstats 1.29.
2549 54. Added transport_filter_timeout as a generic transport option.
2551 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2552 Cc: header lines. This was required by RFC 822, but it not required by RFC
2555 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2556 incoming messages that did not have them. Now it does so only if the
2557 message originates locally, that is, if there is no associated remote host
2558 address. When Resent- header lines are present, this applies to the Resent-
2559 lines rather than the non-Resent- lines.
2561 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2562 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2564 58. Messages for configuration errors now include the name of the main
2565 configuration file - useful now that there may be more than one file in a
2566 list (.included file names were always shown).
2568 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2569 for those rare installations that do not start the daemon as root or run it
2570 setuid root. I've cut out the call to initgroups() if the daemon is not
2573 60. The Exim user and group can now be bound into the binary as text strings
2574 that are looked up at the start of Exim's processing.
2576 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2578 62. Added $mailstore_basename variable.
2580 63. Installed patch to sieve.c from Michael Haardt.
2582 64. When Exim failed to open the panic log after failing to open the main log,
2583 the original message it was trying to log was written to stderr and debug
2584 output, but if they were not available (the usual case in production), it
2585 was lost. Now it is written to syslog before the two lines that record the
2586 failures to open the logs.
2588 65. Users' Exim filters run in subprocesses under the user's uid. It is
2589 possible for a "deliver" command or an alias in a "personal" command to
2590 provoke an address rewrite. If logging of address rewriting is configured,
2591 this fails because the process is not running as root or exim. There may be
2592 a better way of dealing with this, but for the moment (because 4.30 needs
2593 to be released), I have disabled address rewrite logging when running a
2594 filter in a non-root, non-exim process.
2600 1. The buildconfig auxiliary program wasn't quoting the value set for
2601 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2602 not defined. This bug was masked in 4.22 by the effect that was fixed in
2605 2. Some messages that were rejected after a message id was allocated were
2606 shown as "incomplete" by exigrep. It no longer does this for messages that
2607 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2609 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2610 have allow_domain_literals set, the ID did not get logged in the <= line.
2611 Domain literals are now always recognized in Message-ID: header lines.
2613 4. The first argument for a ${extract expansion item is the key name or field
2614 number. Leading and trailing spaces in this item were not being ignored,
2615 causing some misleading effects.
2617 5. When deliver_drop_privilege was set, single queue runner processes started
2618 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2619 same command in the process it spins off) were not dropping privilege.
2621 6. When the daemon running as "exim" started a queue runner, it always
2622 re-executed Exim in the spun-off process. This is a waste of effort when
2623 deliver_drop_privilege is set. The new process now just calls the
2624 queue-runner function directly.
2630 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2633 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2634 ignored. Though the use of -oP was forcing the writing of a pid file, it
2635 was always written to the default place.
2637 3. If the message "no IP address found for host xxxx" is generated during
2638 incoming verification, it is now followed by identification of the incoming
2639 connection (so you can more easily find what provoked it).
2641 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
2643 5. Added some features to "harden" Exim a bit more against certain attacks:
2645 (a) There is now a build-time option called FIXED_NEVER_USERS that can
2646 be put in Local/Makefile. This is like the never_users runtime option,
2647 but it cannot be overridden. The default setting is "root".
2649 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
2650 prefix string with which any file named in a -C command line option
2653 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
2654 is retained for -C and -D only if the caller of Exim is root. Without
2655 it, the exim user may also use -C and -D and retain privilege.
2657 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
2658 command line option is disabled.
2660 6. Macro names set by the -D option must start with an upper case letter, just
2661 like macro names defined in the configuration file.
2663 7. Added "dereference=" facility to LDAP.
2665 8. Two instances of the typo "uknown" in the source files are fixed.
2667 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
2668 the Configure-Makefile script screwed up while processing it.
2670 10. Incorporated PCRE 4.4.
2672 11. The SMTP synchronization check was not operating right at the start of an
2673 SMTP session. For example, it could not catch a HELO sent before the client
2674 waited for the greeting. There is now a check for outstanding input at the
2675 point when the greeting is written. Because of the duplex, asynchronous
2676 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
2677 way, but not yet received, when the check is performed.
2679 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
2680 on TCP/IP sockets, because this apparently causes some broken clients to
2683 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
2684 unchanged) from the Cygwin maintainer.
2686 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
2687 was supported instead of testing for mbx. Effectively a typo.
2689 15. The spa authenticator server code was not checking that the input it
2690 received was valid base64.
2692 16. The debug output line for the "set" modifier in ACLs was not showing the
2693 name of the variable that was being set.
2695 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
2697 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
2698 lookup on its own. Something else was needed to trigger the lookup. For
2699 example, a match in host_lookup or the need for a host name in a host list.
2700 Now, if $sender_host_name is referenced and the host name has not yet been
2701 looked up, a lookup is performed. If the lookup fails, the variable remains
2702 empty, and $host_lookup_failed is set to "1".
2704 19. Added "eqi" as a case-independent comparison operator.
2706 20. The saslauthd authentication condition could segfault if neither service
2707 nor realm was specified.
2709 21. If an overflowing value such as "2048M" was set for message_size_limit, the
2710 error message that was logged was misleading, and incoming SMTP
2711 connections were dropped. The message is now more accurate, and temporary
2712 errors are given to SMTP connections.
2714 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
2715 (except RSET) with a 421 error, until QUIT is received. However, it was
2716 failing to send a response to QUIT.
2718 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
2719 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
2720 ACL is now run after the helo_try_verify_hosts code.
2722 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
2723 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
2724 case-independent, so other case variants are also recognized). Apparently
2725 some systems use these upper case variants.
2727 25. If more than two messages were waiting for the same host, and a transport
2728 filter was specified for the transport, Exim sent two messages over the
2729 same TCP/IP connection, and then failed with "socket operation on non-
2730 socket" when it tried to send the third.
2732 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
2734 27. The extern definition of crypt16() in expand.c was not being excluded when
2735 the OS had its own crypt16() function.
2737 28. Added bounce_return_body as a new option, and bounce_return_size_limit
2738 as a preferred synonym for return_size_limit, both as an option and as an
2741 29. Added LIBS=-liconv to OS/Makefile-OSF1.
2743 30. Changed the default configuration ACL to relax the local part checking rule
2744 for addresses that are not in any local domains. For these addresses,
2745 slashes and pipe symbols are allowed within local parts, but the sequence
2746 /../ is explicitly forbidden.
2748 31. SPA server authentication was not clearing the challenge buffer before
2751 32. log_message in a "warn" ACL statement was writing to the reject log as
2752 well as to the main log, which contradicts the documentation and doesn't
2753 seem right (because no rejection is happening). So I have stopped it.
2755 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
2756 However, I am unable to do any testing of this.
2758 34. Fixed an infelicity in the appendfile transport. When checking directories
2759 for a mailbox, to see if any needed to be created, it was accidentally
2760 using path names with one or more superfluous leading slashes; tracing
2761 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
2763 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
2764 contents are added to the log line that is written for every discarded
2765 recipient. (Previously a log_message setting was ignored.)
2767 36. The ${quote: operator now quotes the string if it is empty.
2769 37. The install script runs exim in order to find its version number. If for
2770 some reason other than non-existence or emptiness, which it checks, it
2771 could not run './exim', it was installing it with an empty version number,
2772 i.e. as "exim-". This error state is now caught, and the installation is
2775 38. An argument was missing from the function that creates an error message
2776 when Exim fails to connect to the socket for saslauthd authentication.
2777 This could cause Exim to crash, or give a corrupted message.
2779 39. Added isip, isip4, and isip6 to ${if conditions.
2781 40. The ACL variables $acl_xx are now saved with the message, and can be
2782 accessed later in routers, transports, and filters.
2784 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
2785 strings in the file are not string-expanded.
2787 42. If a MAIL command specified a SIZE value that was too large to fit into an
2788 int variable, the check against message_size_limit failed. Such values are
2789 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
2790 day this will have to be increased, but I don't think I want to be around
2791 when emails are that large.
2798 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
2799 iconv() is not standard in FreeBSD.
2801 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
2802 IPv6 enabled. The observed symptom was a segmentation fault on return from
2803 the function os_common_find_running_interfaces() in src/os.c.
2805 3. In the check_special_case() function in daemon.c I had used "errno" as an
2806 argument name, which causes warnings on some systems. This was basically a
2807 typo, since it was named "eno" in the comments!
2809 4. The code that waits for the clock to tick (at a resolution of some fraction
2810 of a second) so as to ensure message-id uniqueness was always waiting for
2811 at least one whole tick, when it could have waited for less. [This is
2812 almost certainly not relevant at current processor speeds, where it is
2813 unlikely to ever wait at all. But we try to future-proof.]
2815 5. The function that sleeps for a time interval that includes fractions of a
2816 second contained a race. It did not block SIGALRM between setting the
2817 timer, and suspending (a couple of lines later). If the interval was short
2818 and the sigsuspend() was delayed until after it had expired, the suspension
2819 never ended. On busy systems this could lead to processes getting stuck for
2822 6. Some uncommon configurations may cause a lookup to happen in a queue runner
2823 process, before it forks any delivery processes. The open lookup caching
2824 mechanism meant that the open file or database connection was passed into
2825 the delivery process. The problem was that delivery processes always tidy
2826 up cached lookup data. This could cause a problem for the next delivery
2827 process started by the queue runner, because the external queue runner
2828 process does not know about the closure. So the next delivery process
2829 still has data in the lookup cache. In the case of a file lookup, there was
2830 no problem because closing a file descriptor in a subprocess doesn't affect
2831 the parent. However, if the lookup was caching a connection to a database,
2832 the connection was closed, and the second delivery process was likely to
2833 see errors such as "PGSQL: query failed: server closed the connection
2834 unexpectedly". The problem has been fixed by closing all cached lookups
2835 in a queue runner before running a delivery process.
2837 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
2838 seem to have the "const" qualifier which it has on other OS. I've
2841 8. Change 4.21/2 was too strict. It is only if there are two authenticators
2842 *of the same type* (client or server) with the same public name that an
2843 error should be diagnosed.
2845 9. When Exim looked up a host name for an IP address, but failed to find the
2846 original IP address when looking up the host name (a safety check), it
2847 output the message "<ip address> does not match any IP for NULL", which was
2848 confusing, to say the least. The bug was that the host name should have
2849 appeared instead of "NULL".
2851 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
2852 user that is built into the binary, and the -C or -D options is used, root
2853 privilege is dropped before the configuration file is read. In addition,
2854 logging is switched to stderr instead of the normal log files. If the
2855 configuration then re-defines the Exim user, the unprivileged environment
2856 is probably not what is expected, so Exim logs a panic warning message (but
2859 However, if deliver_drop_privilege is set, the unprivileged state may well
2860 be exactly what is intended, so the warning has been cut out in that case,
2861 and Exim is allowed to try to write to its normal log files.
2867 1. smtp_return_error_details was not giving details for temporary sender
2868 or receiver verification errors.
2870 2. Diagnose a configuration error if two authenticators have the same public
2873 3. Exim used not to create the message log file for a message until the first
2874 delivery attempt. This could be confusing when incoming messages were held
2875 for policy or load reasons. The message log file is now created at the time
2876 the message is received, and an initial "Received" line is written to it.
2878 4. The automatically generated man page for command line options had a minor
2879 bug that caused no ill effects; however, a more serious problem was that
2880 the procedure for building the man page automatically didn't always
2881 operate. Consequently, release 4.20 contains an out-of-date version. This
2882 shouldn't happen again.
2884 5. When building Exim with embedded Perl support, the script that builds the
2885 Makefile was calling 'perl' to find its compile-time parameters, ignoring
2886 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
2888 6. The freeze_tell option was not being used for messages that were frozen on
2889 arrival, either by an ACL or by local_scan().
2891 7. Added the smtp_incomplete_transaction log selector.
2893 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
2894 was accepting AUTH without a new EHLO.
2896 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
2897 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
2900 10. Logging of TCP/IP connections (when configured) now happens in the main
2901 daemon process instead of the child process, so that the TCP/IP connection
2902 count is more accurate (but it can never be perfect).
2904 11. The use of "drop" in a nested ACL was not being handled correctly in the
2905 outer ACL. Now, if condition failure induced by the nested "drop" causes
2906 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
2907 or "require"), the connection is dropped.
2909 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
2910 that yield "discard" can now be used with an "accept" or a "discard" verb,
2911 but an error is generated for any others (because I can't see a useful way
2912 to define what should happen).
2914 13. When an ACL is read dynamically from a file (or anywhere else), the lines
2915 are now processed in the same way as lines in the Exim configuration file.
2916 In particular, continuation lines are supported.
2918 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
2920 15. Added -ti meaning -t -i.
2922 16. Check for letters, digits, hyphens, and dots in the names of dnslist
2923 domains, and warn by logging if others are found.
2925 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
2926 returned from GIFCONF when Exim is trying to find the list of interfaces on
2927 a host. The code in os.c has been modified to copy each ifreq to an aligned
2928 structure in all cases.
2930 Also, in some cases, the returned ifreq's were being copied to a 'struct
2931 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
2932 means the last couple of bytes of an IPv6 address could be chopped if the
2933 ifreq contained only a normal sockaddr (14 bytes storage).
2935 18. Named domain lists were not supported in the hosts_treat_as_local option.
2936 An entry such as +xxxx was not recognized, and was treated as a literal
2939 19. Ensure that header lines added by a DATA ACL are included in the reject log
2940 if the ACL subsequently rejects the message.
2942 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
2943 MD5 (which is deprecated).
2945 21. When testing a filter file using -bf, Exim was writing a message when it
2946 took the sender from a "From " line in the message, but it was not doing so
2947 when it took $return_path from a Return-Path: header line. It now does.
2949 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
2950 with a valid header line field name (a series of printing characters
2951 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
2953 23. Changed "disc" in the source to "disk" to conform to the documentation and
2954 the book and for uniformity.
2956 24. Ignore Sendmail's -Ooption=value command line item.
2958 25. When execve() failed while trying to run a command in a pipe transport,
2959 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
2960 could be confused with a return value of 69 from the command itself. This
2961 has been changed to 127, the value the shell returns if it is asked to run
2962 a non-existent command. The wording for the related log line suggests a
2963 non-existent command as the problem.
2965 26. If received_header_text expands to an empty string, do not add a Received:
2966 header line to the message. (Well, it adds a token one on the spool, but
2967 marks it "old" so that it doesn't get used or transmitted.)
2969 27. Installed eximstats 1.28 (addition of -nt option).
2971 28. There was no check for failure on the call to getsockname() in the daemon
2972 code. This can fail if there is a shortage of resources on the system, with
2973 ENOMEM, for example. A temporary error is now given on failure.
2975 29. Contrary to the C standard, it seems that in some environments, the
2976 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
2977 program. Exim now does this explicitly; it affects the formatting of
2978 timestamps using strftime().
2980 30. If exiqsumm was given junk data, it threw up some uninitialized variable
2981 complaints. I've now initialized all the variables, to avoid this.
2983 32. Header lines added by a system filter were not being "seen" during
2984 transport-time rewrites.
2986 33. The info_callback() function passed to OpenSSL is set up with type void
2987 (*)(SSL *, int, int), as described somewhere. However, when calling the
2988 function (actually a macro) that sets it up, the type void(*)() is
2989 expected. I've put in a cast to prevent warnings from picky compilers.
2991 34. If a DNS black list lookup found a CNAME record, but there were no A
2992 records associated with the domain it pointed at, Exim crashed.
2994 35. If a DNS black list lookup returned more than one A record, Exim ignored
2995 all but the first. It now scans all returned addresses if a particular IP
2996 value is being sought. In this situation, the contents of the
2997 $dnslist_value variable are a list of all the addresses, separated by a
3000 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3001 to accept a host name and all its aliases if the forward lookup for any of
3002 them yielded the IP address of the incoming connection. Now it accepts only
3003 those names whose forward lookup yields the correct IP address. Any other
3004 names are discarded. This closes a loophole whereby a rogue DNS
3005 administrator could create reverse DNS records to break through a
3006 wildcarded host restriction in an ACL.
3008 37. If a user filter or a system filter that ran in a subprocess used any of
3009 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3010 the wrong values were passed to the pipe command ($thisaddress had the
3011 value of $0, $0 had the value of $1, etc). This bug was introduced by
3012 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3014 38. Improved the line breaking for long SMTP error messages from ACLs.
3015 Previously, if there was no break point between 40 and 75 characters, Exim
3016 left the rest of the message alone. Two changes have been made: (a) I've
3017 reduced the minimum length to 35 characters; (b) if it can't find a break
3018 point between 35 and 75 characters, it looks ahead and uses the first one
3019 that it finds. This may give the occasional overlong line, but at least the
3020 remaining text gets split now.
3022 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3023 file descriptors might be low, and that setting 1000 would always raise it.
3024 It turns out that in some environments, the limit is already over 1000 and
3025 that lowering it causes trouble. So now Exim takes care not to decrease it.
3027 40. When delivering a message, the value of $return_path is set to $sender_
3028 address at the start of routing (routers may change the value). By an
3029 oversight, this default was not being set up when an address was tested by
3030 -bt or -bv, which affected the outcome if any router or filter referred to
3033 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3034 log only when "message" or "log_message" are set. However, if one of the
3035 conditions was an address verification, or a call to a nested ACL, the
3036 messages generated by the underlying test were being passed through. This
3037 no longer happens. The underlying message is available in $acl_verify_
3038 message for both "message" and "log_message" expansions, so it can be
3039 passed through if needed.
3041 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3042 new expansion $bh_ to give the encoded byte string without charset
3043 translation. Translation happens only if iconv() is available; HAVE_ICONV
3044 indicates this at build time. HEADERS_CHARSET gives the charset to
3045 translate to; headers_charset can change it in the configuration, and
3046 "headers charset" can change it in an individual filter file.
3048 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3049 that creates RFC 2047 encoded "words" labels them as that charset instead
3050 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3051 expansion operator; (ii) when Exim creates a From: line for a local
3052 message; (iii) when a header line is rewritten to include a "phrase" part.
3054 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3055 buggy, causing it to skip the first lines of messages whose message ID
3056 ended in 'D'. This would not have bitten before Exim release 4.14, because
3057 message IDs were unlikely to end in 'D' before then. The effect was to have
3058 incorrect size information for certain domains.
3060 45. #include "config.h" was missing at the start of the crypt16.c module. This
3061 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3064 46. If there was a timeout during a "random" callout check, Exim treated it as
3065 a failure of the random address, and carried on sending RSET and the real
3066 address. If the delay was just some slowness somewhere, the response to the
3067 original RCPT would be taken as a response to RSET and so on, causing
3068 mayhem of various kinds.
3070 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3071 when I implemented it. It didn't allow for the fact that some option values
3072 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3073 the right test for positive values.
3075 48. Domain names in DNS records are case-independent. Exim always looks them up
3076 in lower case. Some resolvers return domain names in exactly the case they
3077 appear in the zone file, that is, they may contain uppercase letters. Not
3078 all resolvers do this - some return always lower case. Exim was treating a
3079 change of case by a resolver as a change of domain, similar to a widening
3080 of a domain abbreviation. This triggered its re-routing code and so it was
3081 trying to route what was effectively the same domain again. This normally
3082 caused routing to fail (because the router wouldn't handle the domain
3083 twice). Now Exim checks for this case specially, and just changes the
3084 casing of the domain that it ultimately uses when it transmits the message
3087 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3090 50. If a filter generated a file delivery with a non-absolute name (possible if
3091 no home directory exists for the router), the forbid_file option was not
3094 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3095 the existing equality matching.
3097 52. Exim was using ints instead of ino_t variables in some places where it was
3098 dealing with inode numbers.
3100 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3101 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3102 called TMPDIR, and if it finds it is different, it changes its value.
3104 54. The smtp_printf() function is now made available to local_scan() so
3105 additional output lines can be written before returning. There is also an
3106 smtp_fflush() function to enable the detection of a dropped connection.
3107 The variables smtp_input and smtp_batched_input are exported to
3110 55. Changed the default runtime configuration: the message "Unknown user"
3111 has been removed from the ACL, and instead placed on the localuser router,
3112 using the cannot_route_message feature. This means that any verification
3113 failures that generate their own messages won't get overridden. Similarly,
3114 the "Unrouteable address" message that was in the ACL for unverifiable
3115 relay addresses has also been removed.
3117 56. Added hosts_avoid_esmtp to the smtp transport.
3119 57. The exicyclog script was not checking for the esoteric option
3120 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3121 will work only if exicyclog is run under the appropriate euid.
3123 58. Following a discussion on the list, the rules by which Exim recognises line
3124 endings on incoming messages have been changed. The -dropcr and drop_cr
3125 options are now no-ops, retained only for backwards compatibility. The
3126 following line terminators are recognized: LF CRLF CR. However, special
3127 processing applies to CR:
3129 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3130 nor a local message in the state where . is a terminator.
3132 (ii) If a bare CR is encountered in a header line, an extra space is added
3133 after the line terminator so as not to end the header. The reasoning
3134 behind this is that bare CRs in header lines are most likely either
3135 to be mistakes, or people trying to play silly games.
3137 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3138 was being incorrectly given as 18 bytes larger than it should have been.
3139 This is a VOB (very old bug).
3141 60. This may never have affected anything current, but just in case it has:
3142 When the local host is found other than at the start of a list of hosts,
3143 the local host, those with the same MX, and any that follow, are discarded.
3144 When the list in question was part of a longer list of hosts, the following
3145 hosts (not currently being processed) were also being discarded. This no
3146 longer happens. I'm not sure if this situation could ever has previously
3149 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3152 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3153 header. This is something that is recommended in a new Internet Draft, and
3154 is something that is documented as being done by Sendmail. There are two
3155 possible values. For messages generated by the autoreply transport, Exim
3158 Auto-Submitted: auto-replied
3160 whereas for all other generated messages (e.g. bounces) it adds
3162 Auto-Submitted: auto-generated
3164 63. The "personal" condition in filters now includes a test for the
3165 Auto-Submitted: header. If it contains the string "auto-" the message it
3166 not considered personal.
3168 64. Added rcpt_include_affixes as a generic transport option.
3170 65. Added queue_only_override (default true).
3172 66. Added the syslog_duplication option.
3174 67. If what should have been the first header line of a message consisted of
3175 a space followed by a colon, Exim was mis-interpreting it as a header line.
3176 It isn't of course - it is syntactically invalid and should therefore be
3177 treated as the start of the message body. The misbehaviour could have
3178 caused a number of strange effects, including loss of data in subsequent
3179 header lines, and spool format errors.
3181 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3182 client host had authenticated. This control can now be exercised by an ACL
3183 for more flexibility.
3185 69. By default, callouts do not happen when testing with -bh. There is now a
3186 variant, -bhc, which does actually run the callout code, including
3187 consulting and updating the callout cache.
3189 70. Added support for saslauthd authentication, courtesy of Alexander
3192 71. If statvfs() failed on the spool or log directories while checking their
3193 size for availability, Exim confusingly gave the error "space shortage".
3194 Furthermore, in debugging mode it crashed with a floating point exception.
3195 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3196 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3197 serious problem, Exim now writes to the main and panic logs when this
3198 happens, with details of the failure. It then refuses to accept the
3199 incoming message, giving the message "spool directory problem" or "log
3200 directory problem" with a 421 code for SMTP messages.
3202 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3203 0, 1, and 2 exist, because some OS complain for execs without them (see
3204 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3205 descriptors. However, the code omitted to check that the open succeeded,
3206 causing mysterious errors if for some reason the permissions on /dev/null
3207 got screwed. Now Exim writes a message to the main and panic logs, and
3208 bombs out if it can't open /dev/null.
3210 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3211 interact so that it is all more flexible. It is supposed to remain
3212 backwards compatible. Also added extra_local_interfaces.
3214 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3215 to bomb out with an assertion failure - to the client this appears as a
3216 connection drop. This problem occurs in the part of the code that was taken
3217 from the Samba project. Fortunately, the assertion is in a very simple
3218 function, so I have fixed this by reproducing the function inline in the
3219 one place where it is called, and arranging for authentication to fail
3220 instead of killing the process with assert().
3222 75. The SPA client code was not working when the server requested OEM rather
3223 than Unicode encoding.
3225 76. Added code to make require_files with a specific uid setting more usable in
3226 the case where statting the file as root fails - usually a non-root-mounted
3227 NFS file system. When this happens and the failure is EACCES, Exim now
3228 forks a subprocess and does the per-uid checking as the relevant uid.
3230 77. Added process_log_path.
3232 78. If log_file_path was not explicitly set, a setting of check_log_space or
3233 check_log_inodes was ignored.
3235 79. If a space check for the spool or log partitions fails, the incident is now
3236 logged. Of course, in the latter case the data may get lost...
3238 80. Added the %p formatting code to string_format() so that it can be used to
3239 print addresses in debug_print(). Adjusted all the address printing in the
3240 debugging in store.c to use %p rather than %d.
3242 81. There was a concern that a line of code in smtp_in.c could overflow a
3243 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3244 initially expressed, the concern was not well-founded, because trailing
3245 spaces are removed early. However, if the trailing spaces were followed by
3246 a NULL, they did not get removed, so the overflow was possible. Two fixes
3249 (a) I re-wrote the offending code in a cleaner fashion.
3250 (b) If an incoming SMTP command contains a NULL character, it is rejected
3253 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3254 runs initgroups(), so that if the Exim user is in any additional groups,
3255 they will be used during message reception.
3261 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob