Manually flush SSD cache

When Ceph tiering is in use and the SSD cache tier is stuck at it's full ratio threshold of (e.g. 60%), automatic flushing to the backing HDD pool may not progress due to incoming I/O and flush operations pile up. This can lead to persistent high IOPS load. To remediate, manually trigger an SSD cache pool flush operation.

Access CubeCOS cluster VIP via SSH

ssh admin@<your-cluster-vip>

Warning: Permanently added '192.168.1.x' (ECDSA) to the list of known hosts.
Password:

Flush SSD data to HDD pool

Enter storage > cache > flush in the terminal to manually flush the cache.

controller> storage
controller:storage> cache
controller:cache> flush