Re: SUA: RCs and DAUDs

Mark,

There are also provisions for trickling the responses back.  It does
not take multiple RCs with multiple Destinations to cause congestion
in this regard, just DAUD 1(*) is enough if there are many nodes in
a network.  So that's not really a reason, you need congestion avoidance
on responses anyway.

--brian

Erickson, Mark wrote:                                            (Thu, 16 Feb 2006 12:52:30)
> The problem is not returning the status - the problem is if there are
> more than one network referenced by the RCs, what status do you return?
> The SG  must provide the status for each network, whether the auditing
> ASP actually cares about the destination in a network or not, since the
> ASP hasn't given the SG a clear indication of his/her intention.  So,
> potentially, for the one message, there are ((# RCs in DAUD) * (#
> destinations in DAUD) * 2) messages going back if congestion and
> availability status are included.  If the ASP audits frequently, as we
> have seen in some cases, a fair amount of overhead traffic can be
> generated.
> 
> Mark
> 
> -----Original Message-----
> From: Brian F. G. Bidulock [mailto:[email protected]] 
> Sent: Thursday, February 16, 2006 12:51 PM
> To: Erickson, Mark
> Cc: [email protected]
> Subject: Re: [Sigtran] SUA: RCs and DAUDs
> 
> Mark,
> 
> Well, section 4.5.3 says:
> 
>    An SG MAY refuse to provide the availability or congestion status of
>    a destination or subsystem if, for example, the ASP is not authorized
>    to know the status of the destination or subsystem.  The SG MAY
>    respond with an Error Message (Error Code = "Destination Status
>    Unknown") or Error Message (Error Code = "Subsystem Status Unknown").
> 
> which you are pretty much welcome to use in any circumstance where the
> the SG will not or cannot provide the status.
> 
> But, why don't you just return the status?
> 
> --brian
> 
> Erickson, Mark wrote:                          (Thu, 16 Feb 2006
> 09:19:03)
> > 
> >    All:
> > 
> > 
> >    In  the  SUA  RFC,  DAUDs  are  permitted  to  not  only have
> multiple
> >    destinations,  but  also  multiple  RCs.  Since one determines
> network
> >    domain  from  the RC, it's conceivable that if a DAUD has multiple
> RCs
> >    they   could   be   from  different  networks  -  i.e.,  the  DAUD
> is
> >    network-ambiguous.  Since there is no correlation of destination to
> RC
> >    specified  by  the RFC, one would then have to generate a response
> for
> >    each  destination's  appearance in each network referenced by the
> list
> >    of RCs.
> > 
> > 
> >    So,  as  a  suggestion/proposal,  if the SG receives a DAUD
> containing
> >    multiple RCs, if:
> > 
> >    a)        the  RCs  all  reference  one  network, process the DAUD
> and
> >    return status for the destinations (each status response would
> contain
> >    one destination but all the applicable RCs from the DAUD)
> > 
> >    b)       the  RCs reference multiple networks, reject the DAUD with
> an
> >    error, say maybe "Invalid RC".
> > 
> > 
> >    Comments, rebuttals?  Or maybe I'm just off base here?
> > 
> > 
> >    Mark
> 
> > _______________________________________________
> > Sigtran mailing list
> > [email protected]
> > https://www1.ietf.org/mailman/listinfo/sigtran
> 
> 
> -- 
> Brian F. G. Bidulock
> [email protected]
> http://www.openss7.org/
> 

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