CONTEXT
The notes below are based on work done in Nov-08 with deployment of an HPC Cluster (26 nodes, pe1950 dual-quadcore pe1950 servers) running Rocks 4.5 (RHEL 4.X based linux environment)
The particular issues summarized here relate to:
- raid health monitoring for OS disks (internal disks - hardware raid mirrors)
- raid health monitoring for data disks (perc6e attached external MD1000 sata disk arrays)
- configuration of IPMI to permit remote control of power state (power on, power off) of server.
Most of this material was simply located online, but scattered across various sites. Hopefully by presenting all these resources here / in one location, it may be easier for others to use / locate in the future.