Multiple disks reported as failed in same aggregate for specific shelf
Applies to
- DS460C
- DS2246
Issue
- System power-cycled for maintenance
- After reboot of controller and shelves, one shelf appears missing
- All the disks belonging to the shelf show in failed state/missing
Cluster::> system node run -node nodename
Type 'exit' or 'Ctrl-D' to return to the CLI
Nodename> sysconfig -r
Aggregate XXXXXXXXXXXX (failed, raid_tec, partial) (block checksums)
Plex /XXXXXXXXXXXX/plex0 (offline, failed, inactive)
RAID group /XXXXXXXXXXXX/plex0/rg0 (partial, block checksums)
RAID Disk Device HA SHELF BAY CHAN Pool Type RPM Used (MB/blks) Phys (MB/blks)
--------- ------ ------------- ---- ---- ---- ----- -------------- --------------
tparity FAILED N/A 9324290/ -
dparity 0b.01.24 0b 1 24 SA:B 0 FSAS 7200 9324290/19096145920 9342976/19134414848
parity FAILED N/A 9324290/ -
data 0b.01.53 0b 1 53 SA:B 0 FSAS 7200 9324290/19096145920 9342976/19134414848
data FAILED N/A 9324290/ -
data 0b.01.54 0b 1 54 SA:B 0 FSAS 7200 9324290/19096145920 9342976/19134414848
data FAILED N/A 9324290/ -
Raid group is missing 4 disks.
- Missing shelf does not appear in
sysconfig -a
- Both PSU LEDs show amber
- Both PSU work fine when moved to a different shelf
- SAS port link activity LED show as unlit between missing shelf and connected shelves
- IOMs from missing shelf do not work in different shelf