Re: [Isms] comments on draft-nelson-isms-extended-vacm-00

[d.b.nelson at comcast.net]

Juergen Schoenwaelder writes...

> I think Randy Presuhn's model was that the RADIUS
> client acts pretty much like an "embedded" SNMP
> manager doing a local set to modify the VACM table.

I think that's the right generic model, except that it isn't the RADIUS client that does this work.  (Another modularity discussion.)  The RADIUS client is typically isolated from the SNMP engine by PAM and the SSH server.  So, something external to the RADIUS client, that is triggered by the actions of the SSH server or the SSH Transport Model, will take the information from the tnStateReference and affect the VACM tables.