RBE Slow Performance

[jmoraes]

I have a 7507 with RSP4 and IOS 12.2(26) Service Provider. This router are terminate 1600 ATM VCs with RBE.

On my standard configuration the sub-interface has the following configuration:

interface ATM1/0.32 point-to-point

ip unnumbered Loopback0

ip helper-address XXX.XXX.XXX.XXX

atm route-bridged ip

pvc 10/32

encapsulation aal5snap

All works fine until I configure some (300) point to point interfaces for users with range of fixed IP address. This subscribers hava the following configuration template:

interface ATM1/0.1572 point-to-point

ip address XXX.XXX.XXX.XXX 255.255.255.248

atm route-bridged ip

pvc 6/823

encapsulation aal5snap

This router have only one static default route to border router to Internet.

My problem is the performance. The customer are experience slow access to Internet, and the ping to the customer PC show slow response time (600ms).

After reload, the router return to the normal performance, but after 6 hours the customers experience slow performance again. I allways disable CEF but without results. The router show CPU utilizatin under 30%.

Anyone can help me ?

Thank's in advance

[b.hsu]

When the bridging logic runs in a loop to replicate the broadcasts, it relinquishes the CPU intermittently. Relinquishing the CPU is also called suspending the CPU.

Note: Configuring many subinterfaces in the same bridge group can also overwhelm the CPU.

Typical Symptoms

If your bridged PVCs result in high CPU utilization on the router, the first thing to look for would be a high number of broadcasts on your interface. This is shown in the output below:

ATM_Router# show interface atm1/0

---------> Received 463236 broadcasts, 0 runts, 0 giants, 0 throttles

You can see the above output while received more broadcast traffic.

For more information: http://www.cisco.com/en/US/tech/tk39/tk371/technologies_tech_note09186a0080093c91.shtml