Skip to main content

Disk Capacity Alert

The Arvancloud Cloud Database continuously monitors your cluster's disk usage and sends alerts at two specific thresholds to let you know the disk is filling up. If the disk reaches full capacity, the system automatically switches the cluster to read-only mode to prevent data corruption and service failure.

How It Works​

When the cluster's disk usage reaches 80% of capacity, the system issues an alert with Warning severity and notifies you of this threshold by email, SMS, and panel notification at the same time. At this stage, the cluster keeps working normally, and both read and write operations are available.

At this stage, before reaching the critical threshold, it is recommended that you increase the disk capacity or delete unnecessary data.

If you take no action after receiving the warning and disk usage reaches 100%, the system automatically puts the cluster in read-only mode. In this mode, read operations are available, but no write operations can run. To return the cluster to normal, you need to increase the disk capacity.

Note that the occupied disk space includes the database's data and the temporary files the system needs. These temporary files mostly consist of incremental backups, which, besides being transferred immediately to Object Storage, are kept on the disk for a while. This process is vital for preventing data corruption in critical situations.

Notifications​

At both stages, messages are sent to you by email, SMS, and panel notification.

StageEventSeverity
Stage 1Disk usage ≥ 80%Warning
Stage 2Disk usage = 100%Critical

Cluster Status​

When the cluster is at the warning stage, besides the notifications, a new status is shown as a warning indicator in the Disk column of the Database List page.

If disk usage reaches 100% and the cluster is switched to read-only mode, the status indicator changes to show the critical, read-only state.

Restoring the Service​

After reaching the critical state, the only way to return the cluster to normal is to increase the disk capacity.

The guide to resizing a database cluster helps you with this.

After the capacity increase is complete, the read-only state is removed automatically and the cluster returns to normal.