This link is to release notes for 15.0(2)SE shows support for REP in 3560 and 3750 but not in 2960.
I looked in the Cisco Feature Navigator and had difficulty in finding anything that shows support for REP on the 2960. But when I looked at 2960X in the most recent release the Feature Navigator does show support for REP. It does look like this is a flaw in Feature Navigator that Cisco should address.
The switch is onsite now with no REP configured.
I can't remember the exact version it was running, but we definitely checked it with the Feature navigator.
Nowhere could I find how to exactly configure it on the 2960X so we did not configure it, luckely it was not required since it was the only switch in the Ring that did not support it (the others were IE).
I have configured it before on a 3750X with no problem, but that commands do not work on a 2960X.
What happens is we have a REP ring of industrial switches (2x 1Gb uplink, 8x 100Mb downlink) and need to connect our servers to that ring.
So now, if we want to put a 24 ports 1Gb switch in that REP ring with full REP protocol, we need to buy high range 3750X to get a 1Gb link to our servers.
I raised the issue with the team that maintains the Feature Navigator. Here is the response that I received:
I got the confirmation from the platform manager and will do necessary changes in CFN.
It will reflect in CFN by the end of tomorrow.
So Cisco agrees that it was a mistake in Feature Navigator and will correct it.
It is good to know that you did solve your problem by using RSTP.
Cisco has the Feature Navigator as a way to help customers find features and where they are supported. Like any tool there will be some flaws and mistakes (especially when you consider the extremely large number of features and the wide variety of platforms and versions of code). Feedback from customers in a situation like this helps Cisco find the things that need to be fixed. And they generally do respond to customer feedback - as they have in this issue. They have made changes about this entry in Feature Navigator and so now the question for those of you who discovered the flaw - do you agree that the current information is more accurate?