1 $Cambridge: exim/doc/doc-txt/NewStuff,v 1.4 2004/10/21 15:24:42 ph10 Exp $
6 This file contains descriptions of new features that have been added to Exim,
7 but have not yet made it into the main manual (which is most conveniently
8 updated when there is a relatively large batch of changes). The doc/ChangeLog
9 file contains a listing of all changes, including bug fixes.
15 1. There is a new build-time option called CONFIGURE_GROUP which works like
16 CONFIGURE_OWNER. It specifies one additional group that is permitted for
17 the runtime configuration file when the group write permission is set.
19 2. The "control=submission" facility has a new option /retain_sender. This
20 has the effect of setting local_sender_retain true and local_from_check
21 false for the incoming message in which it is encountered.
23 3. $recipients is now available in the predata ACL (oversight).
25 4. The value of address_data from a sender verification is now available in
26 $sender_address_data in subsequent conditions in the ACL statement. Note:
27 this is just like $address_data. The value does not persist after the end
28 of the current ACL statement. If you want to preserve it, you can use one
35 1. There is a new Boolean global option called mua_wrapper, defaulting false.
36 This causes Exim to run an a restricted mode, in order to provide a very
39 Background: On a personal computer, it is a common requirement for all
40 email to be sent to a smarthost. There are plenty of MUAs that can be
41 configured to operate that way, for all the popular operating systems.
42 However, there are MUAs for Unix-like systems that cannot be so configured:
43 they submit messages using the command line interface of
44 /usr/sbin/sendmail. In addition, utility programs such as cron submit
47 Requirement: The requirement is for something that can provide the
48 /usr/sbin/sendmail interface and deliver messages to a smarthost, but not
49 provide any queueing or retrying facilities. Furthermore, the delivery to
50 the smarthost should be synchronous, so that if it fails, the sending MUA
51 is immediately informed. In other words, we want something that in effect
52 converts a command-line MUA into a TCP/SMTP MUA.
54 Solutions: There are a number of applications (for example, ssmtp) that do
55 this job. However, people have found them to be lacking in various ways.
56 For instance, some sites want to allow aliasing and forwarding before
57 sending to the smarthost.
59 Using Exim: Exim already had the necessary infrastructure for doing this
60 job. Just a few tweaks were needed to make it behave as required, though it
61 is somewhat of an overkill to use a fully-featured MTA for this purpose.
63 Setting mua_wrapper=true causes Exim to run in a special mode where it
64 assumes that it is being used to "wrap" a command-line MUA in the manner
67 If you set mua_wrapper=true, you also need to provide a compatible router
68 and transport configuration. Typically there will be just one router and
69 one transport, sending everything to a smarthost.
71 When run in MUA wrapping mode, the behaviour of Exim changes in the
74 (a) A daemon cannot be run, nor will Exim accept incoming messages from
75 inetd. In other words, the only way to submit messages is via the
78 (b) Each message is synchonously delivered as soon as it is received (-odi
79 is assumed). All queueing options (queue_only, queue_smtp_domains,
80 control=queue, control=freeze in an ACL etc.) are quietly ignored. The
81 Exim reception process does not finish until the delivery attempt is
82 complete. If the delivery was successful, a zero return code is given.
84 (c) Address redirection is permitted, but the final routing for all
85 addresses must be to the same remote transport, and to the same list of
86 hosts. Furthermore, the return_address must be the same for all
87 recipients, as must any added or deleted header lines. In other words,
88 it must be possible to deliver the message in a single SMTP
89 transaction, however many recipients there are.
91 (d) If the conditions in (c) are not met, or if routing any address results
92 in a failure or defer status, or if Exim is unable to deliver all the
93 recipients successfully to one of the hosts immediately, delivery of
94 the entire message fails.
96 (e) Because no queueing is allowed, all failures are treated as permanent;
97 there is no distinction between 4xx and 5xx SMTP response codes from
98 the smarthost. Furthermore, because only a single yes/no response can
99 be given to the caller, it is not possible to deliver to some
100 recipients and not others. If there is an error (temporary or
101 permanent) for any recipient, all are failed.
103 (f) If more than one host is listed, Exim will try another host after a
104 connection failure or a timeout, in the normal way. However, if this
105 kind of failure happens for all the hosts, the delivery fails.
107 (g) When delivery fails, an error message is written to the standard error
108 stream (as well as to Exim's log), and Exim exits to the caller with a
109 return code value 1. The message is expunged from Exim's spool files.
110 No bounce messages are ever generated.
112 (h) No retry data is maintained, and any retry rules are ignored.
114 (i) A number of Exim options are overridden: deliver_drop_privilege is
115 forced true, max_rcpt in the smtp transport is forced to "unlimited",
116 remote_max_parallel is forced to one, and fallback hosts are ignored.
118 The overall effect is that Exim makes a single synchronous attempt to
119 deliver the message, failing if there is any kind of problem. Because no
120 local deliveries are done and no daemon can be run, Exim does not need root
121 privilege. It should be possible to run it setuid=exim instead of
122 setuid=root. See section 48.3 in the 4.40 manual for a general discussion
123 about the advantages and disadvantages of running without root privilege.
125 2. There have been problems with DNS servers when SRV records are looked up.
126 Some mis-behaving servers return a DNS error or timeout when a non-existent
127 SRV record is sought. Similar problems have in the past been reported for
128 MX records. The global dns_again_means_nonexist option can help with this
129 problem, but it is heavy-handed because it is a global option. There are
130 now two new options for the dnslookup router. They are called
131 srv_fail_domains and mx_fail_domains. In each case, the value is a domain
132 list. If an attempt to look up an SRV or MX record results in a DNS failure
133 or "try again" response, and the domain matches the relevant list, Exim
134 behaves as if the DNS had responded "no such record". In the case of an SRV
135 lookup, this means that the router proceeds to look for MX records; in the
136 case of an MX lookup, it proceeds to look for A or AAAA records, unless the
137 domain matches mx_domains.
139 3. The following functions are now available in the local_scan() API:
141 (a) void header_remove(int occurrence, uschar *name)
143 This function removes header lines. If "occurrence" is zero or negative,
144 all occurrences of the header are removed. If occurrence is greater
145 than zero, that particular instance of the header is removed. If no
146 header(s) can be found that match the specification, the function does
149 (b) BOOL header_testname(header_line *hdr, uschar *name, int length,
152 This function tests whether the given header has the given name. It
153 is not just a string comparison, because whitespace is permitted
154 between the name and the colon. If the "notdel" argument is TRUE, a
155 FALSE return is forced for all "deleted" headers; otherwise they are
156 not treated specially. For example:
158 if (header_testname(h, US"X-Spam", 6, TRUE)) ...
160 (c) void header_add_at_position(BOOL after, uschar *name, BOOL topnot,
161 int type, char *format, ...)
163 This function adds a new header line at a specified point in the header
164 chain. If "name" is NULL, the new header is added at the end of the
165 chain if "after" is TRUE, or at the start if "after" is FALSE. If
166 "name" is not NULL, the headers are searched for the first non-deleted
167 header that matches the name. If one is found, the new header is added
168 before it if "after" is FALSE. If "after" is true, the new header is
169 added after the found header and any adjacent subsequent ones with the
170 same name (even if marked "deleted"). If no matching non-deleted header
171 is found, the "topnot" option controls where the header is added. If it
172 is TRUE, addition is at the top; otherwise at the bottom. Thus, to add
173 a header after all the Received: headers, or at the top if there are no
174 Received: headers, you could use
176 header_add_at_position(TRUE, US"Received", TRUE, ' ', "X-xxx: ...");
178 Normally, there is always at least one non-deleted Received: header,
179 but there may not be if received_header_text expands to an empty
182 (d) BOOL receive_remove_recipient(uschar *recipient)
184 This is a convenience function to remove a named recipient from the
185 list of recipients. It returns TRUE if a recipient was removed, and
186 FALSE if no matching recipient could be found. The argument must be a
187 complete email address.
189 4. When an ACL "warn" statement adds one or more header lines to a message,
190 they are added at the end of the existing header lines by default. It is
191 now possible to specify that any particular header line should be added
192 right at the start (before all the Received: lines) or immediately after
193 the first block of Received: lines in the message. This is done by
194 specifying :at_start: or :after_received: (or, for completeness, :at_end:)
195 before the text of the header line. (Header text cannot start with a colon,
196 as there has to be a header name first.) For example:
198 warn message = :after_received:X-My-Header: something or other...
200 If more than one header is supplied in a single warn statement, each one is
201 treated independently and can therefore be placed differently. If you add
202 more than one line at the start, or after the Received: block, they will
203 end up in reverse order.
205 Warning: This facility currently applies only to header lines that are
206 added in an ACL. It does NOT work for header lines that are added in a
207 system filter or in a router or transport.
209 5. There is now a new error code that can be used in retry rules. Its name is
210 "rcpt_4xx", and there are three forms. A literal "rcpt_4xx" matches any 4xx
211 error received for an outgoing SMTP RCPT command; alternatively, either the
212 first or both of the x's can be given as digits, for example: "rcpt_45x" or
213 "rcpt_436". If you want (say) to recognize 452 errors given to RCPT
214 commands by a particular host, and have only a one-hour retry for them, you
215 can set up a retry rule of this form:
217 the.host.name rcpt_452 F,1h,10m
219 Naturally, this rule must come before any others that would match.
221 These new errors apply to both outgoing SMTP (the smtp transport) and
222 outgoing LMTP (either the lmtp transport, or the smtp transport in LMTP
223 mode). Note, however, that they apply only to responses to RCPT commands.
225 6. The "postmaster" option of the callout feature of address verification has
226 been extended to make it possible to use a non-empty MAIL FROM address when
227 checking a postmaster address. The new suboption is called "postmaster_
228 mailfrom", and you use it like this:
230 require verify = sender/callout=postmaster_mailfrom=abc@x.y.z
232 Providing this suboption causes the postmaster check to be done using the
233 given address. The original "postmaster" option is equivalent to
235 require verify = sender/callout=postmaster_mailfrom=
237 If both suboptions are present, the rightmost one overrides.
241 (1) If you use a non-empty sender address for postmaster checking, there is
242 the likelihood that the remote host will itself initiate a callout
243 check back to your host to check that address. As this is a "normal"
244 callout check, the sender will most probably be empty, thus avoiding
245 possible callout loops. However, to be on the safe side it would be
246 best to set up your own ACLs so that they do not do sender verification
247 checks when the recipient is the address you use for postmaster callout
250 (2) The caching arrangements for postmaster checking do NOT take account of
251 the sender address. It is assumed that either the empty address, or a
252 fixed non-empty address will be used. All that Exim remembers is that
253 the postmaster check for the domain succeeded or failed.
255 7. When verifying addresses in header lines using the verify=header_sender
256 option, Exim behaves by default as if the addresses are envelope sender
257 addresses from a message. Callout verification therefore tests to see
258 whether a bounce message could be delivered, by using an empty address in
259 the MAIL FROM command. However, it is arguable that these addresses might
260 never be used as envelope senders, and could therefore justifiably reject
261 bounce messages (empty senders). There is now an additional callout option
262 for verify=header_sender that allows you to specify what address to use in
263 the MAIL FROM command. You use it as in this example:
265 require verify = header_sender/callout=mailfrom=abcd@x.y.z
269 (1) As in the case of postmaster_mailfrom (see above), you should think
270 about possible loops.
272 (2) In this case, as in the case of recipient callouts with non-empty
273 senders (the use_sender option), caching is done on the basis of a
274 recipient/sender pair.
276 8. If you build Exim with USE_READLINE=yes in Local/Makefile, it will try to
277 load libreadline dynamically whenever the -be (test expansion) option is
278 used without command line arguments. If successful, it will then use
279 readline() for reading the test data. A line history is supported. By the
280 time Exim does this, it is running as the calling user, so this should not
281 cause any security problems. Security is the reason why this is NOT
282 supported for -bt or -bv, when Exim is running as root or exim,
283 respectively. Note that this option adds to the size of the Exim binary,
284 because the dynamic loading library is not otherwise included. On my
285 desktop it adds about 2.5K. You may need to add -ldl to EXTRA_LIBS when you
286 set USE_READLINE=yes.
288 9. Added ${str2b64:<string>} to the expansion operators. This operator
289 converts an arbitrary string into one that is base64 encoded.
291 10. A new authenticator, called cyrus_sasl, has been added. This requires
292 the presence of the Cyrus SASL library; it authenticates by calling this
293 library, which supports a number of authentication mechanisms, including
294 PLAIN and LOGIN, but also several others that Exim does not support
295 directly. The code for this authenticator was provided by Matthew
296 Byng-Maddick of A L Digital Ltd (http://www.aldigital.co.uk). Here follows
299 xx. THE CYRUS_SASL AUTHENTICATOR
301 The cyrus_sasl authenticator provides server support for the Cyrus library
302 Implementation of the RFC 2222 "Simple Authentication and Security Layer".
303 It provides a gatewaying mechanism directly to the Cyrus interface, so if
304 your Cyrus library can do, for example, CRAM-MD5, then so can the
305 cyrus_sasl authenticator. By default it uses the public name of the driver
306 to determine which mechanism to support.
308 Where access to some kind of secret file is required, for example in GSSAPI
309 or CRAM-MD5, it is worth noting that the authenticator runs as the exim
310 user, and that the Cyrus SASL library has no way of escalating privileges
311 by default. You may also find you need to set environment variables,
312 depending on the driver you are using.
314 xx.1 Using cyrus_sasl as a server
316 The cyrus_sasl authenticator has four private options. It puts the username
317 (on a successful authentication) into $1.
319 server_hostname Type: string* Default: $primary_hostname
321 This option selects the hostname that is used when communicating with
322 the library. It is up to the underlying SASL plug-in what it does with
325 server_mech Type: string Default: public_name
327 This option selects the authentication mechanism this driver should
328 use. It allows you to use a different underlying mechanism from the
329 advertised name. For example:
333 public_name = X-ANYTHING
334 server_mech = CRAM-MD5
337 server_realm Type: string Default: unset
339 This is the SASL realm that the server is claiming to be in.
341 server_service Type: string Default: "smtp"
343 This is the SASL service that the server claims to implement.
345 For straigthforward cases, you do not need to set any of the
346 authenticator's private options. All you need to do is to specify an
347 appropriate mechanism as the public name. Thus, if you have a SASL library
348 that supports CRAM-MD5 and PLAIN, you might have two authenticators as
353 public_name = CRAM-MD5
361 11. There is a new global option called tls_on_connect_ports. Its value must be
362 a list of port numbers; the most common use is expected to be
364 tls_on_connect_ports = 465
366 Setting this option has the same effect as -tls-on-connect on the command
367 line, but only for the specified ports. It applies to all connections, both
368 via the daemon and via inetd. You still need to specify all the ports for
369 the daemon (using daemon_smtp_ports or local_interfaces or the -X command
370 line option) because this option does not add an extra port -- rather, it
371 specifies different behaviour on a port that is defined elsewhere. The
372 -tls-on-connect command line option overrides tls_on_connect_ports, and
373 forces tls-on-connect for all ports.
375 12. There is a new ACL that is run when a DATA command is received, before the
376 data itself is received. The ACL is defined by acl_smtp_predata. (Compare
377 acl_smtp_data, which is run after the data has been received.)
378 This new ACL allows a negative response to be given to the DATA command
379 itself. Header lines added by MAIL or RCPT ACLs are not visible at this
380 time, but any that are defined here are visible when the acl_smtp_data ACL
383 13. The "control=submission" ACL modifier has an option "/domain=xxx" which
384 specifies the domain to be used when creating From: or Sender: lines using
385 the authenticated id as a local part. If the option is supplied with an
386 empty domain, that is, just "/domain=", Exim assumes that the authenticated
387 id is a complete email address, and it uses it as is when creating From:
390 14. It is now possible to make retry rules that apply only when the failing
391 message has a specific sender. In particular, this can be used to define
392 retry rules that apply only to bounce messages. The syntax is to add a new
393 third item to a retry rule, of the form "senders=<address list>". The retry
394 timings themselves then become the fourth item. For example:
396 * * senders=: F,1h,30m
398 would match all bounce messages. If the address list contains white space,
399 it must be enclosed in quotes. For example:
401 a.domain timeout senders="x@b.dom : y@c.dom" G,8h,10m,1.5
403 When testing retry rules using -brt, you can supply a sender using the -f
404 command line option, like this:
406 exim -f "" -brt user@dom.ain
408 If you do not set -f with -brt, a retry rule that contains a senders list
409 will never be matched.
411 15. Two new control modifiers have been added to ACLs: "control = enforce_sync"
412 and "control = no_enforce_sync". This makes it possible to be selective
413 about when SMTP synchronization is enforced. The global option
414 smtp_enforce_sync now specifies the default state of the switch. These
415 controls can appear in any ACL, but the most obvious place to put them is
416 in the ACL defined by acl_smtp_connect, which is run at the start of an
417 incoming SMTP connection, before the first synchronization check.
419 16. Another two new control modifiers are "control = caseful_local_part" and
420 "control = caselower_local_part". These are permitted only in the ACL
421 specified by acl_smtp_rcpt (i.e. during RCPT processing). By default, the
422 contents of $local_part are lower cased before ACL processing.
423 After "control = caseful_local_part", any uppercase letters in the original
424 local part are restored in $local_part for the rest of the ACL, or until
425 "control = caselower_local_part" is encountered. However, this applies only
426 to local part handling that takes place directly in the ACL (for example,
427 as a key in lookups). If a "verify = recipient" test is obeyed, the
428 case-related handling of the local part during the verification is
429 controlled by the router configuration (see the caseful_local_part generic
432 This facility could be used, for example, to add a spam score to local
433 parts containing upper case letters. For example, using $acl_m4 to
434 accumulate the spam score:
436 warn control = caseful_local_part
437 set acl_m4 = ${eval:\
439 ${if match{$local_part}{[A-Z]}{1}{0}}\
441 control = caselower_local_part
443 Notice that we put back the lower cased version afterwards, assuming that
444 is what is wanted for subsequent tests.
446 17. The option hosts_connection_nolog is provided so that certain hosts can be
447 excepted from logging when the +smtp_connection log selector is set. For
448 example, you might want not to log SMTP connections from local processes,
449 or from 127.0.0.1, or from your local LAN. The option is a host list with
450 an unset default. Because it is consulted in the main loop of the daemon,
451 you should strive to restrict its value to a short inline list of IP
452 addresses and networks. To disable logging SMTP connections from local
453 processes, you must create a host list with an empty item. For example:
455 hosts_connection_nolog = :
457 If the +smtp_connection log selector is not set, this option has no effect.
459 18. There is now an acl called acl_smtp_quit, which is run for the QUIT
460 command. The outcome of the ACL does not affect the response code to QUIT,
461 which is always 221. Thus, the ACL does not in fact control any access.
462 For this reason, the only verbs that are permitted are "accept" and "warn".
464 The ACL can be used for tasks such as custom logging at the end of an SMTP
465 session. For example, you can use ACL variables in other ACLs to count
466 messages, recipients, etc., and log the totals at QUIT time using one or
467 more "logwrite" modifiers on a "warn" command.
469 You do not need to have a final "accept", but if you do, you can use a
470 "message" modifier to specify custom text that is sent as part of the 221
473 This ACL is run only for a "normal" QUIT. For certain kinds of disastrous
474 failure (for example, failure to open a log file, or when Exim is bombing
475 out because it has detected an unrecoverable error), all SMTP commands
476 from the client are given temporary error responses until QUIT is received
477 or the connection is closed. In these special cases, the ACL is not run.
479 19. The appendfile transport has two new options, mailbox_size and mailbox_
480 filecount. If either these options are set, it is expanded, and the result
481 is taken as the current size of the mailbox or the number of files in the
482 mailbox, respectively. This makes it possible to use some external means of
483 maintaining the data about the size of a mailbox for enforcing quota
484 limits. The result of expanding these option values must be a decimal
485 number, optionally followed by "K" or "M".
487 20. It seems that there are broken clients in use that cannot handle multiline
488 SMTP responses. Can't people who implement these braindead programs read?
489 RFC 821 mentions multiline responses, and it is over 20 years old. They
490 must handle multiline responses for EHLO, or do they still use HELO?
491 Anyway, here is YAWFAB (yet another workaround for asinine brokenness).
492 There's a new ACL switch that can be set by
494 control = no_multiline_responses
496 If this is set, it suppresses multiline SMTP responses from ACL rejections.
497 One way of doing this would have been just to put out these responses as
498 one long line. However, RFC 2821 specifies a maximum of 512 bytes per
499 response ("use multiline responses for more" it says), and some of the
500 responses might get close to that. So I have implemented this by doing two
503 (1) Extra information that is normally output as part of a rejection
504 caused by sender verification failure is omitted. Only the final line
505 (typically "sender verification failed") is now sent.
507 (2) If a "message" modifier supplies a multiline response, only the first
510 The setting of the switch can, of course, be made conditional on the
513 21. There is now support for the libradius library that comes with FreeBSD.
514 This is an alternative to the radiusclient library that Exim already
515 supports. To use the FreeBSD library, you need to set
517 RADIUS_LIB_TYPE=RADLIB
519 in Local/Makefile, in addition to RADIUS_CONFIGURE_FILE, and you probably
520 also need -libradius in EXTRALIBS.
526 1. The "personal" filter test is brought up-to-date with recommendations from
527 the Sieve specification: (a) The list of non-personal From: addresses now
528 includes "listserv", "majordomo", and "*-request"; (b) If the message
529 contains any header line starting with "List=-" it is treated as
532 2. The Sieve functionality has been extended to support the "copy" and
533 "vacation" extensions, and comparison tests.
535 3. There is now an overall timeout for performing a callout verification. It
536 defaults to 4 times the callout timeout, which applies to individual SMTP
537 commands during the callout. The overall timeout applies when there is more
538 than one host that can be tried. The timeout is checked before trying the
539 next host. This prevents very long delays if there are a large number of
540 hosts and all are timing out (e.g. when the network connections are timing
541 out). The value of the overall timeout can be changed by specifying an
542 additional sub-option for "callout", called "maxwait". For example:
544 verify = sender/callout=5s,maxwait=20s
546 4. Changes to the "personal" filter test:
548 (1) The list of non-personal local parts in From: addresses has been
549 extended to include "listserv", "majordomo", "*-request", and "owner-*",
550 taken from the Sieve specification recommendations.
552 (2) If the message contains any header line starting with "List-" it is
553 treated as non-personal.
555 (3) The test for "circular" in the Subject: header line has been removed
556 because it now seems ill-conceived.
558 5. The autoreply transport has a new option called never_mail. This is an
559 address list. If any run of the transport creates a message with a
560 recipient that matches any item in the list, that recipient is quietly
561 discarded. If all recipients are discarded, no message is created.
567 The documentation is up-to-date for the 4.40 release. What follows here is a
568 brief list of the new features that have been added since 4.30.
570 1. log_incoming_interface affects more log lines.
572 2. New ACL modifier "control = submission".
574 3. CONFIGURE_OWNER can be set at build time to define an alternative owner for
575 the configuration file, in addition to root and exim.
577 4. Added expansion variables $body_zerocount, $recipient_data, and
580 5. The time of last modification of the "new" subdirectory is now used as the
581 "mailbox time last read" when there is a quota error for a maildir
584 6. The special item "+ignore_unknown" may now appear in host lists.
586 7. The special domain-matching patterns @mx_any, @mx_primary, and
587 @mx_secondary can now be followed by "/ignore=<ip list>".
589 8. New expansion conditions: match_domain, match_address, match_local_part,
590 lt, lti, le, lei, gt, gti, ge, and new expansion operators time_interval,
593 9. New lookup type called "iplsearch".
595 10. New log selectors ident_timeout, tls_certificate_verified, queue_time,
596 deliver_time, outgoing_port, return_path_on_delivery.
598 11. New global options smtp_active_hostname and tls_require_ciphers.
600 12. Exinext has -C and -D options.
602 13. "domainlist_cache" forces caching of an apparently variable list.
604 14. For compatibility with Sendmail, the command line option -prval:sval
605 is equivalent to -oMr rval -oMs sval.
607 15. New callout options use_sender and use_postmaster for use when verifying
610 16. John Jetmore's "exipick" utility has been added to the distribution.
612 17. The TLS code now supports CRLs.
614 18. The dnslookup router and the dnsdb lookup type now support the use of SRV
617 19. The redirect router has a new option called qualify_domain.
619 20. exigrep's output now also includes lines that are not related to any
620 particular message, but which do match the pattern.
622 21. New global option write_rejectlog. If it is set false, Exim no longer
623 writes anything to the reject log.
git://git.exim.org / exim.git / blob