Re: M3UA notify message
"Brian F. G. Bidulock" <[email protected]>
| Newsgroups | gmane.ietf.sigtran |
|---|---|
| Organization | http://www.openss7.org/ |
| Message-ID | <[email protected]> |
Tolga, Tolga Asveren wrote: (Wed, 22 Feb 2006 14:54:42) > Barry, > > > -----Original Message----- > > From: Barry Nagelberg [mailto:[email protected]] > > Sent: Wednesday, February 22, 2006 3:06 PM > > To: [email protected] > > Subject: RE: [Sigtran] M3UA notify message > > > > > > Brian, > > > > The AS state is just the combinatorial sum of the ASP states. > > > > So I don't think that it makes sense to say that "each SGP > > maintains its own ASP state, but that AS state is shared > > across the SGPs making up an SG" - this falls apart as soon as > > the ASP states differ on the SGPs. > > > > Futhermore, "ASP state" doesn't exist in a vaccuum. Each ASP has > > a separate state in each AS to which it belongs. So I > > think that the management of both AS and ASP states must be done > > at the same level - either at the SG or the SGP level. > > > > So the next logical question is: "At which level"? I think that > > we have a big problem here. There are conflicting > > answers in the RFC (draft-ietf-sigtran-rfc3332bis-06.doc). I > > think that we urgently need to get rid of these > > contradictions. > > > > Those in favor of maintenance at the SG level can point to the following: > > > > >From section 1.4.1 > > <snip> > > Where an SG contains more than one SGP, the MTP3 routeset, SPMC and > > remote AS/ASP states of each SGP SHOULD be coordinated across all the > > SGPs. > > <snip> > [TOLGA]This section is about Signalling Point Code Representation. To give > the SG-wide unique view for the SPMC, one needs to coordinate AS/ASP states, > but IMO this does not mean running a single ASP/AS state machine from M3UA > procedures point of view, e.g. AS:PC=1-1-1, and there are two SGPs in the > SG. Unless AS is INACTIVE in both of them, SPMC will be AVAILABLE but each > SGP is running its own ASP state machine and following M3UA procedures > independently. The result of a such coordination could be considered as the > "global ASP state" but I think it shouldn't affect M3UA procedures between > SGP and ASP. Read the thread "AS state machine sharing between SGPs". --brian -- Brian F. G. Bidulock [email protected] http://www.openss7.org/