If a cluster stops working, you can restart it to restore normal running. Only clusters in the Available or Abnormal status can be restarted.
- During the restart, the cluster is unavailable. If the restart fails, data may be lost or the cluster may become unavailable. Therefore, exercise caution when performing this operation.
- If the cluster you want to restart is available, you are advised to stop all data processing tasks on the cluster before restarting it. If there are running tasks, for example, importing data, searching for data, backing up or restoring snapshots, then the snapshots or transmitted data may be lost upon the restart.
- The cluster is in the Available or Abnormal status.
- There are no running tasks, such as importing data, searching for data, creating snapshots, and restoring snapshots, on the cluster.
- Log in to the CSS management console.
- Click Cluster Management to switch to the Cluster Management page. In the row where the cluster you want to restart is located, click More > Restart in the Operation column.
- It takes several minutes for the cluster to restart. Refresh the page and check the cluster status. During the restart, the cluster status is Processing, and the task status is Restarting. If the cluster status changes to Available, the cluster has been restarted successfully. If the cluster status changes to Abnormal, you are advised to contact customer service for troubleshooting.