Cisco core and data center migration - Cisco 6509


- A customer is replacing a a single EoL Cisco 6509 with a new 6509-E core (VSS) pair.  
- All servers in the existing data center are directly attached via GE to the old 6509.
- They are also implementing a new L2 data center layer (implementing 10-GE Nexus 5000/2000 devices) - these are hanging off the new 6509 pair, and the servers will eventually all be migrated from the existing 6509 to this new L2 fabric.
- The server transition from the old 6509 to the new L2 fabric will happen over the course of 2-3 weeks, and it needs to be plug/play - meaning, unplug a server from the old switch, plug it into the new fabric, and it works w/o any reconfiguration of the server or the switch infrastructure.

So, they wish to implement the new 6509 pair, along w/ the data center, but the old system will remain operational alongside until all of the servers are migrated.

Question: I'm looking for the best way to implement this.  Thinking out loud: implement the new core pair and DC fabric, and hang the old/existing 6509 off of one/both of these new switches.  Convert the existing 6509 to a purely L2 device, so that all of its ports, uplink and down to the servers, are on the same VLAN as the new Nexus 5000 L2 switches.  Thus, the servers will be on the same subnet whether they are connected to the existing switch or to the new fabric.

I'd love to hear some different ideas/perspectives on this.

Thanks in advance!
Who is Participating?
Seems like a very good plan to me.  Since they do not want to reconfigure the servers you really don't have a lot of options.  You mention having to "convert" the existing 6509 to layer2 only, so I'm guessing it has an MSFC and is doing routing currently for those/that subnet.  I suppose downtime is hard to come by since this is a DC, so my thought there is if you are running HSRP or another client facing redundancy protocol you could add the new devices to the group in order to migrate the subnet.  Chances are you are not however, although some early dual sup/msfc configs did run HSRP between them, so I suppose it's possible that you are as well.  The only thing that I'd say is make sure you are prepared to clear the CAM table on the switch(es) in question when you move the servers, but you've probably already considered that.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.