asterisk users - May 2004 - "Glare" condition - How well does asteriskhandle?

You are correct... No glare on a PRI

W. Kevin Hunt

CCIE #11841
www.huntbrothers.com
 
-----Original Message-----
From: asterisk-users-admin@lists.digium.com
[mailto:asterisk-users-admin@lists.digium.com] On Behalf Of Steven
Critchfield
Sent: Tuesday, May 25, 2004 3:26 PM
To: asterisk-users@lists.digium.com
Subject: Re: [Asterisk-Users] "Glare" condition - How well does
asteriskhandle?

On Tue, 2004-05-25 at 13:53, Scott Stingel wrote:
> Hi-
> 
> I have an upcoming application that requires use of PRI channels that 
> are primarily used for high-volume incoming traffic, but that are to 
> be used for outbound calling as well.  Of course, one option is to 
> have dedicated outbound channels reserved, but this is an inefficient 
> use of channel resources.
> 
> Normally PBX's are designed to have the CPE yield to an incoming call 
> if a particular channel is seized by both ends at the same time (a 
> condition known as "glare"), but I'm wondering if anyone has 
> real-world experience with asterisk to say how well this is handled.
While I may be wrong, I don't think "glare" happens on PRI. The
difference being that the call isn't sent over a channel until there had
been communications on the D channel. This means a send and a receive.
"Glare" would happen on a channelized T1 where it is possible for each
end to try and seize the channel at the same time, since there isn't any
out of band communications.
--
Steven Critchfield  <critch@basesys.com>

_______________________________________________
Asterisk-Users mailing list
Asterisk-Users@lists.digium.com
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

> You are correct... No glare on a PRI
Are you sure?

What about the scenario where the CO switch offers you a call at the same
time you try to set up an outbound call on the same channel?

We dealt with this at my previous company (developing with VOS/CT-ADE and
Dialogic hardware connected to PRIs) and I'm pretty sure there was some
documentation from Intel that said how to deal with it. I think incoming
calls have to take precendence, so if you try and make an outbound call at
the same time an incoming call is presented, you answer the inbound call and
give it regular incoming call treatment, then failover your outbound call to
use a different channel.

Ultimately, PRI is all about exchanging messages in the right order, so it's
totally feasible that you could get glare, no?

very very sure.  pri's don't experience glare.  channelized t1's do
(they are actually CAS lines, or carrier associated signalling)

pri's use out of band signalling (the D channel,which is usually the
24th channel)
Q.921 and Q.931 handle layer 2 and 3 call setup and tear down.  Q.931
would not allow a simultaneous negotiation of both ends attempt to use a
channel.  See http://www.freesoft.org/CIE/Topics/126.htm for more info
on Q.931 or, if you have a cisco box w/ a PRI on it, do a debug isdn
q.931 to see the setup and tear down messages...

W. Kevin Hunt

CCIE #11841
www.huntbrothers.com
 
-----Original Message-----
From: asterisk-users-admin@lists.digium.com
[mailto:asterisk-users-admin@lists.digium.com] On Behalf Of Paul Crick
Sent: Tuesday, May 25, 2004 7:32 PM
To: asterisk-users@lists.digium.com
Subject: RE: [Asterisk-Users] "Glare" condition - How well does
asteriskhandle?
> You are correct... No glare on a PRI
Are you sure?

What about the scenario where the CO switch offers you a call at the
same time you try to set up an outbound call on the same channel?

We dealt with this at my previous company (developing with VOS/CT-ADE
and Dialogic hardware connected to PRIs) and I'm pretty sure there was
some documentation from Intel that said how to deal with it. I think
incoming calls have to take precendence, so if you try and make an
outbound call at the same time an incoming call is presented, you answer
the inbound call and give it regular incoming call treatment, then
failover your outbound call to use a different channel.

Ultimately, PRI is all about exchanging messages in the right order, so
it's totally feasible that you could get glare, no?

_______________________________________________
Asterisk-Users mailing list
Asterisk-Users@lists.digium.com
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Hi,

WKH> You are correct... No glare on a PRI

Really? I followed some of the regular advice that's dispensed on this list
and tried to RTFG. Interestingly it transpires that several hundred hits on
Google seem to imply that you're both wrong:

http://tinyurl.com/2vmrh

..and here are two PRI/Glare scenarios nicely documented by Intel (for their
Linux stack before someone mentions M$):

http://tinyurl.com/27her

and

http://tinyurl.com/27her

Perhaps we are disagreeing over use of terminology rather than an event that
can obviously occur. I understand why  Scott raises the issue especially
with the aggressive services that he supports using Asterisk. Perhaps Scott
could use his call loop-back stress tester code to model the problem and let
us know how Asterisk behaves in a test environment. (Although he might need
two * machines, back to back, to recreate real circuit contention problems.)

Just my 2c

Darren
--
Comgate
Telco>Internet<Broadcast

-----Original Message-----
From: asterisk-users-admin@lists.digium.com
[mailto:asterisk-users-admin@lists.digium.com]On Behalf Of W. Kevin Hunt
Sent: 25 May 2004 23:27
To: asterisk-users@lists.digium.com
Subject: RE: [Asterisk-Users] "Glare" condition - How well does
asteriskhandle?


You are correct... No glare on a PRI

W. Kevin Hunt

CCIE #11841
www.huntbrothers.com

-----Original Message-----
From: asterisk-users-admin@lists.digium.com
[mailto:asterisk-users-admin@lists.digium.com] On Behalf Of Steven
Critchfield
Sent: Tuesday, May 25, 2004 3:26 PM
To: asterisk-users@lists.digium.com
Subject: Re: [Asterisk-Users] "Glare" condition - How well does
asteriskhandle?

On Tue, 2004-05-25 at 13:53, Scott Stingel wrote:
> Hi-
>
> I have an upcoming application that requires use of PRI channels that
> are primarily used for high-volume incoming traffic, but that are to
> be used for outbound calling as well.  Of course, one option is to
> have dedicated outbound channels reserved, but this is an inefficient
> use of channel resources.
>
> Normally PBX's are designed to have the CPE yield to an incoming call
> if a particular channel is seized by both ends at the same time (a
> condition known as "glare"), but I'm wondering if anyone has
> real-world experience with asterisk to say how well this is handled.
While I may be wrong, I don't think "glare" happens on PRI. The
difference being that the call isn't sent over a channel until there had
been communications on the D channel. This means a send and a receive.
"Glare" would happen on a channelized T1 where it is possible for each
end to try and seize the channel at the same time, since there isn't any
out of band communications.
--
Steven Critchfield  <critch@basesys.com>

_______________________________________________
Asterisk-Users mailing list
Asterisk-Users@lists.digium.com
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


_______________________________________________
Asterisk-Users mailing list
Asterisk-Users@lists.digium.com
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users