squirrley controller behavior not a good omen

[Loki]

I sat down today and tried to configure my promise storage. I have 2 controller chassis and one expansion chassis. The standalone controller chassis configuration when smoothly, but the controller+expansion config was a mess. Here's what happened, as best I can tell:

ran the recommended Apple configuration script
controller 1 went offline during the build for some reason
drives 26 and 29 entered some state where they could not be incorporated into the logical disks I was trying to configure, so 2 logical disks immediately started rebuilding with the designated hot spares.

What a mess.

The event log shows messages about losing the heartbeat from controller 1, and failing over to controller 2.

This is brand new hardware, barely burned in.

What sort of things should I look for?

I should mention that I left the equipment running overnight a few days ago, and the chassis that is giving me trouble now got hot enough to trip the alarm. Did I cook the controller?

[abstractrude]

try swapping the controllers. see what happens.

the temp threshold is pretty low on the promise units, you can change it. look at your logs and look at the threshold values and then determine its temp.

[Loki]

abstract-

I powered everything down and swapped the controllers as you suggested. Powered everything back up in the proper order, and the logical drives were there, and seemed fine. The only glitch was that 2 of the 3 hot spares I had configured weren't there. I configured them in the GUI.

So...it seems to be happy now. I can't say my confidence is very high, but we'll see.

thanks for the suggestion.