3550 switch cluster

[aweise]

We are running a cluster of three 3550-24 switches, where one is the Commander, one is standby, and the third is a member. Using the CLI, I can "rcommand" into the member, but not the standby (get a connection refused, a la an ACL blocking a telnet session). Is this typical? I can't see any other reason that we are to be denied access, without being able to examine the config.

Since it's an active switch supporting users, I don't see why I shouldn't be able to manage it.

[thomas.chen]

I think these are the option that may provide you with the solution, Remove the command "cluster enable " then add the cluster command Mac address .Now the switch may become part of the cluster If standby is not able to rcommand, try downloading the latest IOS version to your switch. The final option is to reconfigure the entire cluster

[n1ckpaul77]

I am having difficulties with replacing a failed command-switch in a cluster of C3550 switches. After what appeared to be a successful re-configuration of MAC-addresses on the member switches, and the neighbours on the command switch, a loop was detected and a GBIC interface kept disabling (even though the cabling was reconnected EXACTLY as before).

Is there a documented process in replacing a failed cluster switch which includes configuration and cabling (and I don't mean using CNA, but pure CLI)?