[SIPForum-discussion] VB: 5 minute cut-off

[Binan AL Halabi]

Hi,

To keep the SIP session alive, the terminal sends periodic re-INVITE or UPDATE requests (this is called session refresh requests).


For those terminals which do not send session refresh requests, try to configure these terminals to use high value for session interval (session-Expires header). This value should be accepted by the softswitch. You might need to change the timer on the switch also to be the same or larger.

Regards.
// Binan.



----- Vidarebefordrat meddelande ----
Från: Jorge <jwalcantara at hotmail.com>
Till: 'Tristan Bulandus' <tristanbulandus at yahoo.com>; smraney at comcast.net 
Kopia: 'Binan AL Halabi' <binanalhalabi at yahoo.com>; murali87ece at gmail.com; discussion at sipforum.org 
Skickat: tisdag, 3 september 2013 14:14
Ämne: RE: 5 minute cut-off
 


ptime has to do with media/RTP packetization time , in other words a ptime of 20 says that the media will be sent in 20 millisecond carrying  payload/voice packets of RTP or 50 packets per second.  This is not the cause of the you hang up problem.
 
The problem has to do with re-INVITES and how often they are sent by the varying end points.
An INVITE with a tag in the TO: field is a re-INVITE.
 
 
Jorge Alcantara
 
From:Tristan Bulandus [mailto:tristanbulandus at yahoo.com] 
Sent: Tuesday, September 03, 2013 1:50 AM
To: smraney at comcast.net
Cc: Jorge; Binan AL Halabi; murali87ece at gmail.com; discussion at sipforum.org
Subject: Re: 5 minute cut-off
 
Hi.
 
Huawei softswitch sends BYE with reason: SESSION TIMER EXPIRES. Our Huawei Softswitch has Timer of 300s which is why calls for IOS and Android gets cut off at around 5 minutes.
 
My only concern is that 3 of the other SIP Clients do not experience a call cut-off at around 5 minutes while simulating the exact customer scenario.
 
Would you be familiar on the function of this 'a=ptime' attributes? Is the ptime value in seconds? Would appreciate an immediate reply please.
 
Thanks.
 
From:"smraney at comcast.net" <smraney at comcast.net>
To: Tristan Bulandus <tristanbulandus at yahoo.com> 
Cc: Jorge <jwalcantara at hotmail.com>; Binan AL Halabi <binanalhalabi at yahoo.com>; murali87ece at gmail.com; discussion at sipforum.org 
Sent: Monday, September 2, 2013 8:50 PM
Subject: Re: 5 minute cut-off
 
Tristan,
 
I admittedly do have have working experience with all these particular endpoints, so my suggestions are merely based on observations of the Invite messages.
 
The Invite's associated to the IOS and Android endpoints do not appear to have 'ptime' attributes, whereas the Invites associated to the other three endpoints do.  Issues with ptime attributes typically effect quality of service.
 
PTimer Attributes:
 a. Genband IP Phone - a=ptime: 20
 b. Grandstream IP Phone - a=ptime: 20
 c. IOS - Not Configured
 d. Android - Not Configured
 e. PC Client - a=ptime: 30
 
My other thought was whether or not this could be associated to a 'session refresh' issue, if the session is seen as inactive and requires refresh to remain established?
 
Thanks,
 
Steve.....
 
From: "Tristan Bulandus" <tristanbulandus at yahoo.com>
To: "Jorge" <jwalcantara at hotmail.com>, "Binan AL Halabi" <binanalhalabi at yahoo.com>, smraney at comcast.net, murali87ece at gmail.com
Cc: discussion at sipforum.org
Sent: Sunday, September 1, 2013 11:33:43 PM
Subject: 5 minute cut-off
 
Hi guys.
 
Genband A2 is currently peered to Huawei Softswitch.
 
Genband A2 has the following terminals that register to it:
 a. Genband IP Phone
 b. Grandstream IP Phone
 c. IOS
 d. Android
 e. PC Client
 
Now, I try to call to Numbers routed at the Huawei Softswitch from the following terminals and experience:
a. Genband IP Phone calling Huawei - Call NOT cut-off
b. Grandstream IP Phone calling Huawei - Call NOT cut-off
c. IOS calling Huawei - Call cut-off at 5 minutes
d. Android calling Huawei - Call cut-off at 5 minutes
e. PC Client calling Huawei - Call NOT cut-off
 
May I request what has to be adjusted on the IOS and Android clients please? I've attached a summary of the SIP Invites being generated by terminals.
 
Thanks! 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sipforum.org/pipermail/discussion/attachments/20130907/738c4fd2/attachment-0002.html>