Re: RFC 4666 M3UA - Reg Request Message

Jitin,

The advantage of LOADSEL is that it does not require the SG to have any
knowledge of ISUP beyond the ability to syntactically extract the CIC
(which is trivial), and thus provides a basis for interoperability.

Also, you have not considered the multiple SG as STP scenario where any
stateful approach at the SG is not workable across both SGs.

Because LOADSEL does not require any ISUP knowledge at the SG, it
automatically and transparently handles the multiple SG as STP scenario.

--brian

Bhandari, Jitin (Jitin) wrote:                                   (Mon, 06 Nov 2006 15:51:46)
> 
> Lincoln,
> 
> Just to complete the discussion on our first point, I am referring to ISUP
> specs (ITU or any other variant). You cannot send UPU but as ISUP treats
> unknown CICs with UCIC (for us at sigtran, if you get IAM for an unknown CIC
> at Routing Key) or locally blocked cics with either Group blocking or BLO
> (for us at sigtran, If you get an IAM with AS being down). Again, I am
> avoiding this discussion as this is ISUP variant specific and can get into
> very detailed analysis message by message. But every ISUP variant defines
> such scenarios. 
> 
> Also, I don't think that once we regain connectivity, we need to sync up
> with SG for CIC states because if you look at various ISUP procedures, after
> AS regains connectivity with SGP, it should kick-in its standard ISUP
> procedure at ISUP level to broadcast local and remote CIC status.
> 
> On ETS specs, we know that both Dynamic registration and Routing Key at CIC
> level is not considered (even though then provided by RFC-3332).
> 
> Thanks,
> -Jitin

-- 
Brian F. G. Bidulock
[email protected]
http://www.openss7.org/