So the actual goal is not the OOB vs PFE-forwarded traffic separation but
rather the opposite?
Even if you are able to achieve what you are aiming at (forward traffic
between PFE interfaces and fxp0 interface) then at certain rate it would get
throttled by PFE CPU. And this behaviour (PFE CPU throttling) is not
configurable AFAIK.
Rgds
Alex
----- Original Message -----
From: "Alfred Schweder" <[email protected]>
To: "Alex" <[email protected]>
Cc: <[email protected]>; "Alfred Schweder" <[email protected]>
Sent: Saturday, November 28, 2009 5:49 PM
Subject: Re: [j-nsp] Routing Instance for fxp0
Hello
This is correct fxp0 behaviour.
Just pick any other PFE interface for outband access and put it into
routing-instance.
In this fashion, with dedicated OOB PFE interface, if your M7i RE fails,
you still have OOB access to the node.
I tought fxp0 should be "the" dedicated OOB Interface.
If the fxp0 would be a real freebsd local interface all would be fine.
The problem I have is, that the fxp0 Network(s) are black holed for
the normal traffic in the global context.
I think switching off this side effect will be easy by turning off the
internal bridge. Perhaps there is already a (hidden) switch ?
--
Mit freundlichen Gruessen
Dipl. Ing. A. Schweder
Gutachten & Consulting
Mobil +49.177.2194627 (+49.160.97639300)
_______________________________________________
juniper-nsp mailing list [email protected]
https://puck.nether.net/mailman/listinfo/juniper-nsp
