Okay,
- LAN (ping, default gateway, routes, ...)
- SAN (disk errors, failed paths, IO errors (lvm_io_fail), ..)
- rootvg (disk errors, mirroring)
- errpt (permanent hardware errors, ...)
Monitoring the errpt for permanet hardware errors is a good start
Regards