Fibre Channel (SAN)

Reply
New Contributor
Posts: 2
Registered: ‎09-13-2010

3200 with zero access and strange behavior

I know the bricked 3200 has been covered many times.

As an enthusiast wanting to gain experience in fabric environments, I purchased a used 3200 for my home server to feed four 11 disk arrays.  The device is completely locked down: serial will display messages but not take any input, telnet and web are completely non-functional.  In fact, nmap states all ports are closed.  IP is not the default, so I think I can assume the passwords have also been changed, not that it matters because I never have the opportunity to enter a password.

I am just going to assume that there is no hope of resetting the switch.  Before I declare futility, I intend to open the switch and see if there is a reset jumper or button (unlikely).  In addition, it occurs to me that there must be some obscure way to reset the credentials via some API in the FC connections.  Otherwise, how did the previous owner expect to maintain the switch?  I will look into the FC API approach and see what I can come up with.

Here's my question.  The switch came with 5 SFPs and I am assuming it was intended to perform in a single zone with one (or two?) ISL ports and the remaining four feeding four storage devices.  If so, then the switch config probably is set mostly to defaults.  I cannot seem to attach more than one array without a bunch of errors.

I have a single port from a dual-port QLogic HBA under OpenSolaris b128a attached to the first port of the switch, one other port attached to one loop of an array, and all is well.  Effectively, the switch is an extension cord.

As soon as I attach a second array to the switch, /var/log/messages records a bunch of reset and hardware errors, the second port goes yellow and all I/O stops.  It doesn't seem to matter which switch port I attach to or what array I attach to: the second loop of the already attached array or another array entirely.  Since I lack visibility into the switch, I have no idea what is going on, or if this is even a switch, HBA or OS issue.  Swapping HBA ports, cables, GBICs or switch ports changes nothing, so the issue is when the second array is attached to the switch.

The whole purpose of the switch was to become familiar with fabric environments, so I have very little experience here.

What do you think is going on?

Many thanks in advance,

Marty

New Contributor
Posts: 2
Registered: ‎09-13-2010

Re: 3200 with zero access and strange behavior

It was a bad GBIC after all.

Join the Community

Get quick and easy access to valuable resource designed to help you manage your Brocade Network.