NSF01: Unscheduled downtime
The recent disk failure will be repaired tonight on the large temp storage. This requires a reboot. Sorry for the short notice.
Also, there will be an (overseen) Point of Use (POU) upgrade. When the system comes back up again,
/temporary/saswork
will be (hopefully) functional again as RAID0. The long-term solution involves
- having an on-site spare disk provided by SGI
- having an implementation plan for RAID5 available.
The next time a disk fails, RAID5 will be implemented, using the available on-site spare disk. This should reduce the duration of the next downtime, and prevent future downtimes.
