Cloudera Docs

Replication Policies page

The "Replication Policies" page shows the number of replication policies that are active, the number of policies that have been suspended, the number of policies that are in error state, and the total number of replication policies available in Replication Manager. The page also provides a detailed view about the replication policies.

The Replication Policies page shows a replication policy status dashboard which shows the following panels and the number of policies with the status:
  • Error shows the number of replication policies associated with a cluster designated as Error on the Classic Clusters map. Click the number to understand the policy names, the names of the source and destination clusters, and which services are stopped on the source or destination cluster.
  • Active replication policies that are in Submitted or Running state. This item is not actionable.
  • Suspended replication policies that have been suspended by the administrator. This item is not actionable.
  • Total number of running policies.

Click a replication policy to view more details about the policy. Click Actions (The image shows the icon to change timezones.) to perform more actions on a replication policy.

Additionally, you can perform the following tasks on the Replication Policies page:
  • Change the timezone, if required, using The image shows the icon to change timezones..
  • View the list of unreachable clusters using The image shows an icon that you can use to view the list of unreachable clusters..

Replication policy details

You can also view the following policy details on the Replication Policies page:
  • Current policy Status .
  • Policy Type shows HDFS, Hive, or HBase.
  • Replication policy Name .
  • Source cluster name.
  • Destination cluster name.
  • Jobs that were run for the replication policy and its current status.
  • Duration or time taken to run the policy.
  • Last Success timestamp of the last successful run.
  • Next Run timestamp of the next scheduled run.

Optimize Replication Policies page performance

By default, the replication policies are loaded only partially on the Replication Policies page, therefore the page might display incomplete statistics about a job status. This is because the job history is necessary to decide whether a policy failed or succeeded. The replication policies with failed jobs might take a longer time to load.

You can change the page load behavior depending on your requirements using The image shows the icon to choose the load operation type.. Choose one of the following options to load the Replication Policies page faster by delaying to load the job history:
  • Delay loading job history when it takes too long attempts to load the job history, but omits the load operation above a certain threshold. By default, Replication Manager uses this option.
  • Never load job history minimizes the load on Cloudera Manager and maximizes Replication Manager performance.
  • Always load job history ensures that the job history is always loaded for all the displayed replication policies.
Use Case
Sometimes, Replication Manager fails to reach a healthy Cloudera Manager when there is a temporary networking blip or when there is a load spike on Cloudera Manager. When a cluster becomes unreachable for Replication Manager, the cluster is placed in the list of unreachable clusters (the list appears when you click The image shows the icon that you can use to view the list of unreachable clusters.). Replication Manager retries to reach the cluster again after 20 minutes. After you confirm that the Cloudera Manager is healthy and expect it to be reachable by Replication Manager, you can force reload the Replication Policies page using The image shows the reload icon that you can use to force reload the Replication Policies page. to reconnect every cluster.

For more information, see Replication Policies page does not display all the replication policies.