On 17/04/2008, B Venkat S.R Swamy wrote:
> While going through the RFC 3413 and RFC 3584 for Proxy Forward
> application, there are certain procedures
> that are not clearly defined for proxy. Can some one provide insight
> into the proxy behaviour with respect to following items:
> > Implications on engine discovery. Intermediate or pass-through ?
> > USM and VACM implications.

One thing to realise is that the Net-SNMP agent does not actually
act as a Proxy Forwarder as defined in RFC 3413. That document
described the forwarding of entire requests, based purely on the
administrative header information.
As such, there are indeed issues relating to which of the assorted
SNMP processing engines plays exactly what role in various situations.

Instead the Net-SNMP engine performs what I tend term "proxy
delegation", forwarding selected varbinds from within the request.
In terms of the RFC 3413 terminology, it acts as both Command
Responder, and Command Generator.

This simplifies the interpretation of RFC 3414, since you've
effectively got two separate paths of communication:
client <-> main agent and main agent <-> proxied agent

Engine discovery, access control, etc are performed independently
for these two connections.

As regards the correct behaviour for a "pure Proxy Forwarder",
this is probably a question that is best addressed to the IETF
SNMP working group, who defined the protocols in the first place.


This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
Net-snmp-users mailing list
Please see the following page to unsubscribe or change other options: