alamo maintenance complete

Submitted by David Gignac on Fri, 27 May 2011, 17:28:08 GMT

FutureGrid Hardware Outage Information

alamo maintenance complete

Status
Resolved
Type
Network
Impacted systems
alamo
Start of outage
Thu, 26 May 2011, 17:27 EDT
Anticipated end of outage
Fri, 27 May 2011, 17:00 EDT

Description

A NFS hang wreaked havoc on the IB fabric, most nodes had to be rebooted to clear the problem.

Resolution

Compute nodes were configured to mount NFS via the GigE interfaces. A config change was also needed for OFED stack.

»