you can deploy another 6509 with same configuration as current one which will avoid single point of failure( collapse core) and use HSRP/GLBP/VRRP as first hop redundancy. For choosing among HSRP/VRRP/GLBP is ur decision based on your requirement.
You can definitely have another switch with the same configuration to have redundancy running HSRP. I won't go for VSS becasue it requires a SUP change in your scenario.
However we'll need to understand the current network topology. For example if you have a server directly terminating on 6500 switch you might have to move it to a discribution switch which in turn will be connected to two 6500's or else do NIC teaming on the server.
Similarlty if you have a collapsed core, then each access switch should have two uplinks, one to each 6500. Hence you need to check the interface availability on the access switches.
NSF/SSO supervisor redundancy is supported on the Supervisor 32.
However you are really talking about 2 different things
1) Intra-chassis failover. With this you have dual supervisors in the same chassis. This is where NSF/SSO is used.
Benfit is you can lose a supervisor and the chassis and all the linecards will still work. But if any of your linecards fail you have lost connectivity to all devices on that linecard.
2) Inter-chassis failover. This is where you have 2 6500 chassis connected together either via a L2 trunk or a L3 link depending on your design. Within each chassis you could have single or dual supervisors. For end devices you can run HSRP/VRRP/GLBP to provide a redundant gateway.
Thanks Jon. Your post has clarified a lot of doubts.
Adding on -
Each SUP has 8 fiber ports. To have proper redundancy I should then be connecting SUP1 Chassis 1 (S1C1) to S1C2 and S2C2 and similarly other ports as well. Or can I use Etherchannel for L2 connectivity and HSRP for L3.
Most designs use an etherchannel to connect the 2 switches together. That can be a L3 or a L2 etherchannel and this really does depend on your design and network topology.
Which connections you use to create the etherchannel depends on how may sups you have in each chassis. We ran one supervisor per chassis and used 2 of the GBIC ports on the supervisor bundled with 2 10/100/1000 ethernet ports on a ws-x6748-ge-tx module. You can mix copper and fibre as long as they are the same speed/duplex etc..
I am using the same thread so that I dont have to explain the scenario again. To keep things simple can I stick to the SRM on the two 6509 switches and use HSRP for L3 configuration. This way I dont have to upgrade the IOS or CATOS to support NSF/SSO.
We are pleased to announce availability of Beta software for 16.6.3. 16.6.3 will be the second rebuild on the 16.6 release train targeted towards Catalyst 9500/9400/9300/3850/3650 switching platforms. We are looking for early feedback from custome...