> > > Thanks Paul! I tried the new code. Squirrelamil is definitely slower > > with these options added. > > If you have metrics or profiling traces that quantify this, I'd love to > see that. The added code does (should) not add all that much - it's > just a single extra and very simple IMAP ID command. Did you use the > configuration variable that disables response parsing? > I have these logs, notice it takes over a minute to destroy the connection but perhaps that's normal? Apr 12 10:07:46 auth: Debug: http-client[1]: conn 0.0.0.0:8084 [1]: Idle connection timed out Apr 12 10:07:46 auth: Debug: http-client[1]: conn 0.0.0.0:8084 [1]: Connection close Apr 12 10:07:46 auth: Debug: http-client[1]: conn 0.0.0.0:8084 [1]: Connection disconnect Apr 12 10:07:46 auth: Debug: http-client[1]: conn 0.0.0.0:8084 [1]: ( 0.0.0.0:8084): Disconnected: Connection closed (fd=20) Apr 12 10:07:46 auth: Debug: http-client[1]: conn 0.0.0.0:8084 [1]: Detached peer Apr 12 10:07:46 auth: Debug: http-client[1]: peer 0.0.0.0:8084: Lost a connection (1 queues linked, 0 connections left, 0 connections pending, 0 requests pending, 0 requests urgent) Apr 12 10:07:46 auth: Debug: http-client[1]: conn 0.0.0.0:8084 [1]: Connection destroy Apr 12 10:07:46 auth: Debug: http-client[1]: peer 0.0.0.0:8084: No requests to service for this peer (0 connections exist, 0 pending) Apr 12 10:07:46 auth: Debug: http-client[1]: peer 0.0.0.0:8084: Dropping peer now Apr 12 10:07:46 auth: Debug: http-client[1]: peer 0.0.0.0:8084: Peer close Apr 12 10:07:46 auth: Debug: http-client[1]: peer 0.0.0.0:8084: Peer disconnect Apr 12 10:07:46 auth: Debug: http-client[1]: peer 0.0.0.0:8084: Peer destroy Apr 12 10:07:46 auth: Debug: http-client: peer 0.0.0.0:8084: Peer pool destroy Apr 12 10:07:46 auth: Debug: http-client: peer 0.0.0.0:8084 (shared): Peer destroy Apr 12 10:08:37 auth: Debug: http-client: host 0.0.0.0: Host session destroy Apr 12 10:08:37 auth: Debug: http-client[1]: queue http://0.0.0.0:8084: Destroy Apr 12 10:08:37 auth: Debug: http-client: host 0.0.0.0: Host is idle (timeout = 100 msecs) Apr 12 10:08:37 auth: Debug: http-client: host 0.0.0.0: Host destroy >>>Did you use the configuration variable that disables response parsing? Yes in /etc/squirrelmail/config_local.php I added: $imap_id_command_args = array('remote-host' => '###REMOTE ADDRESS###'); $do_not_parse_imap_id_command_response = TRUE; > And still no source IP of end user but perhaps > > there logs may shed a light? Or what should auth_policy_server_url be? > > The rest of your question is about Dovecot, it is not relevant to this > list. I suggest first sniffing the IMAP connection to make sure the ID > command with the IP address is being sent and acknowledged by Dovecot, > then synchronizing your configuration and debugging that. Any tips on how to do that? Sniffing as in Wireshark? > The Dovecot > list is the best place to continue asking for help as long as the ID > command is working as it should. > I'll try again thanks. ----- squirrelmail-users mailing list Posting guidelines: http://squirrelmail.org/postingguidelines List address: squirrelmail-users@xxxxxxxxxxxxxxxxxxxxx List archives: http://news.gmane.org/gmane.mail.squirrelmail.user List info (subscribe/unsubscribe/change options): https://lists.sourceforge.net/lists/listinfo/squirrelmail-users
