FreeBSD: fix sendfile shim, again

[exim.git] / doc / doc-txt / experimental-spec.txt

diff --git a/doc/doc-txt/experimental-spec.txt b/doc/doc-txt/experimental-spec.txt
index 8c9a866a20740cb764cd2ddfcd33de963301cd57..5e0b044279828c9999a52010885c1e36a9abfeb4 100644 (file)
--- a/doc/doc-txt/experimental-spec.txt
+++ b/doc/doc-txt/experimental-spec.txt
@@ -639,7 +639,7 @@ used via the transport in question.
 
 
 Dovecot authenticator via inet socket
-------------------------------------
+--------------------------------------------------------------
 If Dovecot is configured similar to :-
 
 service auth {
@@ -666,28 +666,54 @@ and a whitespace-separated port number must be given.
 
 
 
-Twophase queue run fast ramp
-----------------------------
-To include this feature, add to Local/Makefile:
-  EXPERIMENTAL_QUEUE_RAMP=yes
 
-If the (added for this feature) main-section option "queue_fast_ramp" (boolean)
-is set, and a two-phase ("-qq") queue run finds, during the first phase, a
-suitably large number of message routed for a given host - then (subject to
-the usual queue-runner resource limits) delivery for that host is initiated
-immediately, overlapping with the remainder of the first phase.
+Logging protocol unusual states
+---------------------------------------------------------------
+An extra log_selector, "protocol_detail" has been added in the default build.
+The name may change in future, hence the Experimenal status.
+
+Currrently the only effect is to enable logging, under TLS,
+of a TCP RST received directly after a QUIT (in server mode).
+
+Outlook is consistently doing this; not waiting for the SMTP response
+to its QUIT, not properly closing the TLS session and not properly closing
+the TCP connection.  Previously this resulted is an error from SSL_write
+being logged.
+
+
+
+Limits ESMTP extension
+---------------------------------------------------------------
+Per https://datatracker.ietf.org/doc/html/draft-freed-smtp-limits-01
+
+If compiled with EXPERIMENTAL_ESMTP_LIMITS=yes :-
+
+As a server, Exim will advertise, in the EHLO response, the limit for RCPT
+commands set by the recipients_max main-section config option (if it is set),
+and the limit for MAIL commands set by the smtp_accept_max_per_connection
+option.
+
+Note that as of writing, smtp_accept_max_per_connection is expanded but
+recipients_max is not.
+
+A new main-section option "limits_advertise_hosts" controls whether
+the limits are advertised; the default for the option is "*".
+
+As a client, Exim will:
 
-This is incompatible with queue_run_in_order.
+ - note an advertised MAILMAX; the lower of the value given and the
+ value from the transport connection_max_messages option is used.
 
-The result should be a faster startup of deliveries when a large queue is
-present and reasonable numbers of messages are routed to common hosts; this
-could be a smarthost case, or delivery onto the Internet where a large proportion
-of recipients hapen to be on a Gorilla-sized provider.
+ - note an advertised RCPTMAX; the lower of the
+ value given and the value from the transport max_rcpt option is used.
+ Parallisation of transactions is not done if due to a RCPTMAX, unlike
+ max_rcpt.
 
-As usual, the presence of a configuration option is associated with a
-predefined macro, making it possible to write portable configurations.
-For this one, the macro is _OPT_MAIN_QUEUE_FAST_RAMP.
+ - note an advertised RCPTDOMAINMAX, and behave as if the transport
+ multi_domains option was set to false.  The value advertised is ignored.
 
+Values advertised are only noted for TLS connections and ones for which
+the server does not advertise TLS support.
 
 
 --------------------------------------------------------------