This article describes the following topics
- Ceph pool has reach the full ratio at ~=60%, by default it will automatically flush the data into HDD pool.
- But, in some production case, the incoming data/IO and the process of flushing data are stacked togather; It will cause a loop at 60% flushing data and accepting incoming data.
- It will create a consistent IOPS for a long period, So we must manually flush the Data with command.
Connect to console
$ ssh [email protected] Warning: Permanently added '192.168.1.x' (ECDSA) to the list of known hosts. Password:
Flush SSD data to HDD pool:
storage > cache > flush
Welcome to the Cube Appliance Enter "help" for a list of available commands HOSTNAME> storage HOSTNAME:storage> cache HOSTNAME:cache> flush