Pausing and resuming a cluster

Suggest edits

Pausing and resuming clusters

Pausing a cluster allows you to save on compute costs without losing data or cluster configuration settings. There's no added cost for paused clusters, although you must still pay for any disk and object storage that the paused cluster requires.

Pausing a cluster drops all existing connections, blocks future actions such as editing the cluster, and stops monitoring. While the cluster is paused, you can't make or maintain connections or read, write, or edit cluster settings. Automatic backups are paused. However, backups taken while the cluster was running remain until their retention period expires.

Note

If you defined the backup retention period for a cluster to be less that the duration of the pause, it's possible for write-ahead logs (WALs) to be deleted from the backup object store, and resumption might fail.

While paused, clusters aren't upgraded or patched, but upgrades are applied when the cluster resumes. Pausing a high availability or Postgres Distributed cluster shuts down all cluster nodes.

After seven days, single-node and high-availability clusters automatically resume. Resuming a cluster applies any pending maintenance upgrades. Monitoring begins again.

With CLI 3.7.0 and later, you can pause and resume a cluster using the CLI.

You can enable in-app inbox or email notifications to get alerted when the paused cluster is or will be reactivated. For more information, see managing notifications.

Pausing a cluster

  1. Go to the Clusters page in the Console.
  2. Do one of the following:
    • In the row for the cluster, at right, from the ellipsis menu, select Pause Cluster.
    • Select the cluster you want to pause. From Quick Actions on the cluster details page, select Pause Cluster.
  3. Confirm that you want to pause the cluster. When the process finishes, the cluster status appears as Paused. It can take up to 15 minutes for the compute instance in your CSP to scale down.

Resuming a cluster

  1. Go to the Clusters page in the Console.
  2. Do one of the following:
    • In the row for the cluster, at right, from the ellipsis menu, select Resume Cluster.
    • Select the cluster you want to resume. From Quick Actions on the cluster details page, select Resume Cluster.
  3. Confirm that you want to resume the cluster. The process might take a few minutes. When it finishes, the cluster status appears as Healthy.
Note

A TDE-enabled cluster resumes only if the TDE key status is ready or available. Clusters are automatically paused if there is any issue with the TDE key. You need to resolve/give permissions to the key in your respective cloud region. Resume the cluster manually after resolving the issues.

← Prev

Modifying database configuration parameters

↑ Up

Managing your cluster

Next →

Performing a major version upgrade of Postgres on Cloud Service

Could this page be better? Report a problem or suggest an addition!