We have 5 blade enclosures with the integrated switches that we connect to Cisco 6006 switches via etherchanneling the gigabit ports (Compaq nomenclature is trunk). They've run OK for almost 2 years. We had a problem last week where one of the blade enclosures began spewing the following error message in the Integrated Admin log: crit mgmt: Integrated Administrator was reset by ESR / Watchdog Timer. This was happening every 30 seconds or so. The other four enclosures then began showing the same message; then the Cisco switch ports went into errdisable mode and began reporting a spanning tree loop on the ports connected to the enclosures. I had to unplug the gigabit ports on all five enclosures, leaving only the port 24 management port on switch A active. Then I was able to plug the gigabit ports back in. Resetting the first enclosure had no effect. What would cause this error message? I've found no reference to it in any documentation.
This may sound like I'm stating the obvious but... apparently the enclosure couldn't refresh the watchdog timer before it expired. So - I would normally suspect a firmware or hardware issue with the enclosure.
You would want to have the latest firmware on the enclosure and then look at the health LEDs for other signs of problems.
You said 4 enclosures did this around the same time so it doesn't make sense that you would have so many hardware failures.
Perhaps something happened on the network that exposed a firmware bug? Why would it go away all of a sudden?
