[Serusers] setflag question
Java Rockx
javarockx at yahoo.com
Thu Sep 30 04:12:17 CEST 2004
Zues,
Thanks. t_check_status was the method I've overlooked.
Paul
--- Zeus Ng <zeus.ng at isquare.com.au> wrote:
> In your failure route, add:
>
> if (t_check_status("408|48[0|6]" /* or whatever status your want */)) {
> voicemail routine
> }
>
> This will enable voicemail and account CANCEL request in your missed call.
>
> Zeus
>
> > -----Original Message-----
> > From: serusers-bounces at lists.iptel.org
> > [mailto:serusers-bounces at lists.iptel.org] On Behalf Of Java Rockx
> > Sent: Thursday, 30 September 2004 12:01 AM
> > To: 'ser users'
> > Subject: RE: [Serusers] setflag question
> >
> >
> > Hello All.
> >
> > I may have stumbled upon the problem. It has to do with not
> > properly handling CANCEL messages.
> >
> > In my ser.cfg I have:
> >
> > modparam("acc", "db_flag", 1)
> > modparam("acc", "db_missed_flag", 1)
> > modparam("acc", "log_fmt", "fisum")
> > modparam("acc", "report_ack", 1)
> > modparam("acc", "log_level", 1)
> > modparam("acc", "failed_transactions", 1)
> > modparam("acc", "report_cancels", 1)
> > modparam("acc|auth_db|domain|group|uri_db|usrloc", "db_url",
> > "mysql://ser:pwd@sipdb01.mycompany.com/ser")
> >
> >
> >
> >
> > route {
> > ... the usual stuff here
> >
> > # BUSY and NOANSWER go to voice mail
> > t_on_failure("1");
> >
> > # Make sure we handle nathelper stuff
> > t_on_reply("1");
> >
> > # Enable accounting for this transaction
> >
> >
> >
> > setflag(1);
> >
> > if (!t_relay()) {
> > sl_reply_error();
> > };
> > }
> >
> > failure_route[1] {
> >
> >
> >
> > log(1, "Failover to voicemail");
> >
> >
> >
> > rewritehostport("vm01.mycompany.com:5060");
> > append_branch();
> > append_urihf("CC-Diversion: ", "\r\n");
> >
> >
> >
> > t_relay(); }
> >
> > onreply_route[1] {
> >
> >
> >
> > if (isflagset(6) && status =~ "(183)|2[0-9][0-9]") {
> > fix_nated_contact();
> > force_rtp_proxy();
> > } else if (nat_uac_test("1")) {
> > fix_nated_contact();
> > };
> > }
> >
> >
> > I'm thinking that the problem is that a BUSY and CANCEL will
> > both follow the same logic and execute failure_route[1]. I
> > don't want CANCEL messages to do this because
> > failure_route[1] is intended for sending BUSY or NOANSWER
> > replies to the voicemail server.
> >
> > How can I determine if the negative value returned by
> > t_relay() is caused by a CANCEL versus a BUSY message?
> >
> > Regards,
> > Paul
> >
>
>
__________________________________
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!
http://promotions.yahoo.com/new_mail
More information about the sr-users
mailing list