Promoting a read replica to primary, for Redis OSS (cluster mode disabled) replication groups - Amazon ElastiCache (Redis OSS)
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Promoting a read replica to primary, for Redis OSS (cluster mode disabled) replication groups

Information in the following topic applies to only Redis OSS (cluster mode disabled) replication groups.

You can promote a Redis OSS (cluster mode disabled) read replica to primary using the Amazon Web Services Management Console, the Amazon CLI, or the ElastiCache API. You can't promote a read replica to primary while Multi-AZ with Automatic Failover is enabled on the Redis OSS (cluster mode disabled) replication group. To promote a Redis OSS (cluster mode disabled) replica to primary on a Multi-AZ enabled replication group, do the following:

  1. Modify the replication group to disable Multi-AZ (doing this doesn't require that all your clusters be in the same Availability Zone). For more information, see Modifying a replication group.

  2. Promote the read replica to primary.

  3. Modify the replication group to re-enable Multi-AZ.

Multi-AZ is not available on replication groups running Redis OSS 2.6.13 or earlier.

Using the Amazon Web Services Management Console

The following procedure uses the console to promote a replica node to primary.

To promote a read replica to primary (console)
  1. Sign in to the Amazon Web Services Management Console and open the ElastiCache console at https://console.amazonaws.cn/elasticache/.

  2. If the replica you want to promote is a member of a Redis OSS (cluster mode disabled) replication group where Multi-AZ is enabled, modify the replication group to disable Multi-AZ before you proceed. For more information, see Modifying a replication group.

  3. Choose Redis OSS, then from the list of clusters, choose the replication group that you want to modify. This replication group must be running the "Redis" engine, not the "Clustered Redis" engine, and must have two or more nodes.

  4. From the list of nodes, choose the replica node you want to promote to primary, then for Actions, choose Promote.

  5. In the Promote Read Replica dialog box, do the following:

    1. For Apply Immediately, choose Yes to promote the read replica immediately, or No to promote it at the cluster's next maintenance window.

    2. Choose Promote to promote the read replica or Cancel to cancel the operation.

  6. If the cluster was Multi-AZ enabled before you began the promotion process, wait until the replication group's status is available, then modify the cluster to re-enable Multi-AZ. For more information, see Modifying a replication group.

Using the Amazon CLI

You can't promote a read replica to primary if the replication group is Multi-AZ enabled. In some cases, the replica that you want to promote might be a member of a replication group where Multi-AZ is enabled. In these cases, you must modify the replication group to disable Multi-AZ before you proceed. Doing this doesn't require that all your clusters be in the same Availability Zone. For more information on modifying a replication group, see Modifying a replication group.

The following Amazon CLI command modifies the replication group sample-repl-group, making the read replica my-replica-1 the primary in the replication group.

For Linux, OS X, or Unix:

aws elasticache modify-replication-group \ --replication-group-id sample-repl-group \ --primary-cluster-id my-replica-1

For Windows:

aws elasticache modify-replication-group ^ --replication-group-id sample-repl-group ^ --primary-cluster-id my-replica-1

For more information on modifying a replication group, see modify-replication-group in the Amazon ElastiCache Command Line Reference.

Using the ElastiCache API

You can't promote a read replica to primary if the replication group is Multi-AZ enabled. In some cases, the replica that you want to promote might be a member of a replication group where Multi-AZ is enabled. In these cases, you must modify the replication group to disable Multi-AZ before you proceed. Doing this doesn't require that all your clusters be in the same Availability Zone. For more information on modifying a replication group, see Modifying a replication group.

The following ElastiCache API action modifies the replication group myReplGroup, making the read replica myReplica-1 the primary in the replication group.

https://elasticache.us-west-2.amazonaws.com/ ?Action=ModifyReplicationGroup &ReplicationGroupId=myReplGroup &PrimaryClusterId=myReplica-1 &Version=2014-12-01 &SignatureVersion=4 &SignatureMethod=HmacSHA256 &Timestamp=20141201T220302Z &X-Amz-Algorithm=&AWS;4-HMAC-SHA256 &X-Amz-Date=20141201T220302Z &X-Amz-SignedHeaders=Host &X-Amz-Expires=20141201T220302Z &X-Amz-Credential=<credential> &X-Amz-Signature=<signature>

For more information on modifying a replication group, see ModifyReplicationGroup in the Amazon ElastiCache API Reference.