Power off/on testing in MetroCluster
Applies to
- MetroCluster IP configuration
- MetroCluster FC configuration
Issue
- When carrying out a powering off/on test on a MetroCluster in the incorrect steps can cause issues such as:
- Node failing to boot back up after power off/on test
- When the node powers on some disks appear as failed/not responding
SYSCONFIG-R outputs:
RAID group /SSD_mirror_aggr0/plex1/rg2 (partial, block checksums)
RAID Disk Device HA SHELF BAY CHAN Pool Type RPM Used (MB/blks) Phys (MB/blks)
--------- ------ ------------- ---- ---- ---- ----- -------------- --------------
dparity Node:8.126L188 1a 17 5 FC:A 1 SSD N/A 915465/1874872832 915715/1875385008
parity FAILED N/A 915465/ -
data Node:8.126L23 1a 10 22 FC:A 1 SSD N/A 915465/1874872832 915715/1875385008
data FAILED N/A 915465/ -
RAID Disk Device HA SHELF BAY CHAN Pool Type RPM Used (MB/blks) Phys (MB/blks)
--------- ------ ------------- ---- ---- ---- ----- -------------- --------------
failed node:8.126L48 1a 11 21 FC:A 1 SAS 10000 1142352/2339537408 1144641/2344225968
not responding Node:8.126L102 1a 13 23 FC:A 1 SAS 10000 1142352/2339537408 1144641/2344225968
not responding Node:8.126L106 1a 14 1 FC:A 1 SAS 10000 1142352/2339537408 1144641/2344225968
not responding Node:8.126L154 1a 15 23 FC:A 1 SAS 10000 1142352/2339537408 1144641/2344225968
not responding Node:8.126L1 1a 10 0 FC:A 1 SSD N/A 915465/1874872832 915715/1875385008