Re: M3UA AS override mode from SGP perspective

Hi Brian,

Thanks for the response, I found it quite helpful even if you
found it futile :)

However, I am still confused about the recent thread on override
mode. Some of the statements seems contradictory to me and I'd
like to be sure I'm drawing the correct conclusions.

Suppose we have an SG with 5 SGPs and 3 SCTP Associations
connecting to each SGP. As per the discussion we just had, each
SCTP Association is considered by the SG as a separate ASP.
From the SG perspective then, there are 15 ASPs connecting to 5
SGPs, regardless of whether any of the associations terminate at
the "same remote ASP process".

In this configuration, how many ASPs can be active at once for
an AS in Override mode?

(1) At most one ASP, for the entire SG
(2) At most one ASP on each SGP (at most 5 ASPs)
(3) It depends on some other details of the configuration
(4) Other

Andrew

[email protected] wrote: -----

To: Andrew Booth <[email protected]>
From: "Brian F. G. Bidulock"
Sent by: [email protected]
Date: 10/12/2010 05:34PM
Cc: [email protected], [email protected], [email protected]
Subject: Re: [Sigtran] M3UA AS override mode from SGP perspective

Andrew,

Andrew Booth wrote: (Tue, 12 Oct 2010 16:49:55)
>
> Could the ASP vendor legitimately claim that an ASPAC message sent
> from the ASP process to the SG on either of its two associations
> should make the ASP active on both associations (assuming the SG
> accepts the ASPAC)? After all, the language in RFC4666 is all about
> messages exchanged between ASPs and SGPs, not specifically about
> associations.

No, they cannot. If you look in Figure 3, SCTP CDI/RI
causes transition from the ASP-ACTIVE and ASP-INACTIVE
state to the ASP-DOWN state. So, loss of an association
is considered loss of communication with the ASP. They
might try to convinve themselves that the model supports
it, but it does not.

Also, as I said, the SG can enforce one ASP per association
for these fancy-pants ASP vendors, whoever they might be,
by requiring an ASP Id thus enforcing its uniqueness. This
is clearly laid out in the RFC, so if they want their ASP
to work with SGs they should certainly support one ASP per
association.

But this discussion is futile: I give ASP designers more
credit than that.

--brian

--
Brian F. G. Bidulock
[email protected]
http://www.openss7.org/
_______________________________________________
Sigtran mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sigtran

_______________________________________________
Sigtran mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sigtran