[c-nsp] Brief CPU spikes on 6500 Sup 720

Benjamin Lovell belovell at cisco.com
Wed Jul 14 12:27:23 EDT 2010


Depending on the number of connected IP addresses this can be an issue. This is why in most cases it will be better to bring up the MAC table timers as opposed to bring down the ARP timers.

-Ben

On Jul 14, 2010, at 11:16 AM, Jon Lewis wrote:

> On Wed, 14 Jul 2010, Phil Mayers wrote:
> 
>>> mac-address-table aging-time 14400
>>> http://www.ciscopress.com/articles/article.asp?p=336872
>> 
>> Or of course drop the ARP timeout:
>> 
>> int VlanX
>> arp timeout 290
> 
> I haven't tracked down the exact issue, but we've seen issues when using too short an arp timeout on a 6500 vlan interface.  Sometimes customers will ask for really short arp timeouts to cope with failover systems that change an IPs MAC address.
> 
> The vlan in question had an arp timeout of 60s and had a couple of KVM servers with 100 or so virtual machines.  Especially when a large number of VMs started up, we'd see periods of packet loss.  My assumption is that the sup720-3bxl can only handle so much arp activity on a vlan.  There was nothing in the config to artificially limit arp traffic.
> 
> ----------------------------------------------------------------------
> Jon Lewis                   |  I route
> Senior Network Engineer     |  therefore you are
> Atlantic Net                |
> _________ http://www.lewis.org/~jlewis/pgp for PGP public key_________
> _______________________________________________
> 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/




More information about the cisco-nsp mailing list