[c-nsp] Arp Input Process Causing Spike in CPU

Christian Koch ckoch at qualitytech.com
Thu Mar 15 09:40:22 EST 2007


This is an NSP network.. All the arp entries are customers

Also log errors

Mar 15 07:38:35: %SYS-3-CPUHOG: Task ran for 6072 msec (72/20), process
= ARP Input, PC = 4025C948.
-Traceback= 4025C950 403CA9A0 4024340C 402433F8
Mar 15 07:39:13: %FIB-2-FIBDOWN: CEF has been disabled due to a low
memory condition.
 It can be re-enabled by configuring "ip cef [distributed]"
edge1-gw1.SJC1#



-Christian 

|-----Original Message-----
|From: Gert Doering [mailto:gert at greenie.muc.de] 
|Sent: Thursday, March 15, 2007 10:38 AM
|To: Christian Koch
|Cc: cisco-nsp at puck.nether.net
|Subject: Re: [c-nsp] Arp Input Process Causing Spike in CPU
|
|Hi,
|
|On Thu, Mar 15, 2007 at 09:44:23AM -0400, Christian Koch wrote:
|> I'm seeing cpu spikes due to  the arp input process running high ..
|> 
|> Currently every route is pointed to next hop and the arp table looks 
|> like..
|> 
|> #sh ip arp sum
|> 1743 IP ARP entries, with 21 of them incomplete
|
|1743 ARP entries?
|
|Over *thousand seven hundred* entries?  What sort of networks 
|have you connected to this router...?
|
|> Can anyone suggest how to reduce this for a short term fix... I know 
|> the long term is to upgrade the mem
|
|Memory doesn't play a role in this.  Topology does, 
|reasonably-sized subnets, CoPP rate-limiting ARP storms, etc.
|
|gert
|--
|USENET is *not* the non-clickable part of WWW!
|                                                           
|//www.muc.de/~gert/
|Gert Doering - Munich, Germany                             
|gert at greenie.muc.de
|fax: +49-89-35655025                        
|gert at net.informatik.tu-muenchen.de
|
|



More information about the cisco-nsp mailing list