[f-nsp] Unexpected CPU load on NetIron CER

[Eldon Koyle]

Are you using PIM or some other multicast routing, by chance?  We had
some nasty issues where even small amounts of multicast would send the
LP cpu to 50% on our MLX routers.  You might want to try the latest 5.5
release and see if that helps.

-- 
Eldon Koyle
-- 
The solution of problems is the most characteristic and peculiar sort
of voluntary thinking.
		-- William James

On  Oct 20 10:28+0200, Bardo Cornelissen wrote:
> Hi folks,
> 
>  
> 
> I have recently installed a NetIron CER 2024F-4X and at implementation I saw
> relatively high CPU levels when using "route-only" on specifically an 10G
> (AMS-IX Exchange) port. "sh cpu lp" showed values at around 50% and icmp
> traffic on the router ip showed packetloss. When I turn of "route-only" the
> cpu utilization drops to 9-10% and stays stable.
> 
>  
> 
> I still find these values high. We previously used an old jetcore router
> with the same settings and that had a stable 1% cpu-utilization. It feels
> some sort of traffic is causing load on the cpu but so far I've not been
> able to track down the issue. Has anyone seen similar behavior?
> 
>  
> 
> IronWare : Version 5.5.0cT183 Copyright (c) 1996-2013 Brocade Communications
> Systems, Inc.
> 
>  
> 
> #sh cpu lp
> 
>  
> 
> 10:27:24 GMT+01 Mon Oct 20 2014
> 
>  
> 
> SLOT  #:          LP CPU UTILIZATION in  %:
> 
>                in 1 second:  in 5 seconds:  in 60 seconds: in 300 seconds:
> 
>      1:        10            9              9               9
> 
>      2:        10            9              9               9
> 
>  
> 
> Kind regards,
> 
>  
> 
> Bardo Cornelissen
> 
>  
> 
> Caveo Internet BV
> 
>  <http://www.caveo.nl/> http://www.caveo.nl
> 

> _______________________________________________
> foundry-nsp mailing list
> foundry-nsp at puck.nether.net
> http://puck.nether.net/mailman/listinfo/foundry-nsp