Can't Cluster 2950-24 switches with identical base MAC addresses

Hi

I have four new Cisco 2950-24 switches running with the Standard
Image.

They are cabled together (I've tried cascading and star topologies)
and traffic flows neatly where it should. Each switch has a unique IP
address assigned to VLAN1.

The first problem I found was that I can only telnet to a given switch
from a system specifically connected to that switch. From any given
workstation I can only ping the switch to which that workstation is
connected. The other three switches just don't reply.

The second problem is that having used CMS to configure any one of the
switches as the cluster commander it is unable to find any candidate
members on the network. CDP does identify all 4 switches.

I have found that all four switches have the identical base MAC
addresses. It seems that this is the best reason why they can't be
seen as different devices by CMS and why they can't forward packets
intended for each other.

Have I missed something in setting up my switches? Is there a way to
change the base MAC address? Are they supposed to have identical MAC
addresses? Could it possibly be true that these cluster capable, STP
capable switches are designed so that there can only be one of them on
any even LAN at a time?

Please help. I'm tearing my hair out.

Thanks

Colin

 

I wish I could help. But it sounds to me like the the arp request is going
unanswered. In addition, I bet if you arp -a <ipaddress_of_switch> - (while
connected directly to each one) you will see that each switch does have a
unique MAC address for VLAN1. Other than that, I can offer nothing else,
save to say that I would expect someone in this NG to know the answer - I
myself have experienced this problem before on several Catalyst 2924's, but
I can't remember what I did to remedy it!

good luck,