Skip to main content
Version: 3.5

Asynchronous Disaster Recovery

In an asynchronous disaster recovery setup, you can replicate your applications and their data between two Kubernetes or OpenShift clusters. A separate Portworx Enterprise cluster runs on each Kubernetes or OpenShift cluster.

important
  • Cluster-wide operators are not migrated during a DR migration unless they are installed in the same namespace as the applications you want to migrate.
    • For OpenShift, the operator installation defaults to the openshift-operators namespace.
    • For Kubernetes, ensure operators are installed in the same namespace as your applications. As a result, after migration, you will not be able to scale up or down your applications on the destination cluster using storkctl.
  • Asynchronous DR is supported for Kubevirt VMs with PX RWX Block Volumes on the following platforms:
    • OpenShift Container Platform
  • Disaster recovery is not supported for FlashArray Direct Access and FlashBlade Direct Access volumes. Do not configure migration schedules for namespaces containing these volumes.
  • In an asynchronous disaster recovery setup, if a MigrationSchedule is created with a 100 MB object size, failover or failback to a cluster configured with a 10 MB object size fails. Portworx by Pure Storage recommends configuring both the source and destination clusters with the same object size to ensure successful MigrationSchedule and failover/failback operations.

The following diagram shows an asynchronous DR setup involving two clusters that are geographically apart:

Detailed Async setup

  • Application data and supported Kubernetes resources are asynchronously replicated from a source to a destination cluster, which means there is a delay between data changes occurring on the source cluster and their replication to the destination cluster.
  • Incremental changes from Kubernetes or OpenShift applications and Portworx data are continuously sent to the destination cluster.
  • If the source cluster becomes unavailable, you can activate the applications in the destination cluster.

Perform the following steps to set up asynchronous DR: