redundant switch supervisors

sarahr202 · ‎01-25-2009

Hi every body!

I was reading about redundancy in switch supervisors.

There are different modes:

1) route processor redundancy(RPR)

2)Route processor redundancy(RPR+)

3)stateful switchover (SSO)

q1)I understand how these modes but my question is why one want to use any other mode but sso as fail over time is less than a second compared to other modes ?

q2) Should both supervisors be same?

Thanks a lot!

Giuseppe Larosa · ‎01-26-2009

Hello Sarah,

q1) SSO requires identical supervisors, identical ios image and newer IOS image (it has more tight requirements) so there can be cases where RPR or RPR+ can be used but SSO cannot

q2) for SSO they need to be exactly the same model, RPR and RPR+ have lesser requirements

see

http://www.cisco.com/en/US/docs/switches/lan/catalyst4500/12.2/46sg/configuration/guide/RPR.html#wp1116237

Hope to help

Giuseppe

sarahr202 · ‎01-26-2009

thanks a lot Giuseppe!

My book says

For RPR+ ios must be of same release. How about sup? they should be same as well?

Amit Singh · ‎01-26-2009

The sups' have to be the same as well.

regards,

-amit singh

Kerem Gursu · ‎01-26-2009

Hi ,

There are several modes , but every mode has its own unique circumstances. The fastest mode is not always the best mode

1)SSO enables subsecond convergence , but some state tables like HSRP encounters problems during the switchover. For example if your supervisors or RPs switch faster than your HSRP timeout , you will definetely experience a switchover in the HSRP state as well.

2) The Supervisors / RPs must be the same.

HTH