[c-nsp] Cat 4500 series High CPU

[Yann GAUTERON]

Hi Terry,

If you haven't done an "no ip redirects" on all your routed interfaces, this
could be an issue.

We faced this problem recently. In fact on a subnet shared by 2 routers, one
host used my 4507 as its default gateway. As the 4507 knew that the best
path to reach the destination was through the second router on the same
subnet, it created and sent back an IGMP Redirect. This was CPU consuming.

If you don't need your routers/L3 switches to send back this IGMP Redirect
message, you'd better to put "no ip redirects" on all your routed interfaces
(no switchport physical interfaces and vlan interfaces).

Hope this helps. Keep us informed.

Yann

2011/8/31 Terry Rupeni <rupeni_t at usp.ac.fj>

> Hi All,
>
>
>
> We have a Cat4506 facing high CPU. Investigating further we have narrowed
> it
> down to the following on the Cisco Bug Report:
>
>
>
> High CPU due to Mgmt LoPri process, MAC addresses being added/deleted
> Symptom:
> A Catalyst 4500 switch might experience high CPU utilization due to the
> Cat4k Mgmt LoPri process and the K2CpuMan and K2L2 Address Table reviews
> (show platform health. High CPU utilization does not impact the traffic
> switched in hardware.
>
>
>
> Conditions:
> The problem is seen when a large MAC address table exists and when the
> switch is frequently relearning MAC addresses on multiple VLANs. Enabling
> service internal followed by debug platform log feature k2l2addresstable
> will show output similar to the following. Do not enable these commands on
> a
> production switch unless instructed by Cisco TAC.
>
>
>
> However no solution is recommended by Cisco. As this is a production switch
> don't really want to do a debug right now.
>
>
>
> Has anyone faced a similar issue?
>
>
>
> Thks
>
> Terry
>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>