Feature Description

VMs in OpenShift should be protected by existing RH disaster recovery solutions that already protect container and pod workloads.

• DFPLAN-9 OpenShift Application Disaster Recovery

Customer run standard IT workloads, such as databases, middleware, and web based applications on both Windows and Linux.

There should be multiple levels of protection available, based on the application and workload business needs (RPO and RTO).

The two solutions available today with OCP + ODF + ACM are:

• Metro-DR (synchronous replication)
• Regional-DR (asynchronous replication) 

Why is this important

As customers consider migration VM workloads to OpenShift, their current production systems have various DR capabilities, whether using VMware Site Recovery Manager, or a third-party solution like IBM Spectrum Protect, Veeam, Dell, Veritas NetBackup.

Having equivalent capabilities for some of the customer workloads is an initial requirement for most customer to consider migrating virtual machines to OpenShift.

The challenge is that VMs can show up in an OpenShift cluster at any time, and the expectations is that whatever RH solution is protecting the OpenShift cluster will work as well for virtual machine workloads.

There are also Red Hat Virtualization (RHV) customers who have similar needs, and use the disaster recovery capabilities built into RHV (custom Ansible playbooks) or a third party ecosystem solution. This need is accelerated by the current Red Hat Virtualization Life Cycle (maintenance phase until Aug 2024) and ultimate end-of-life (Aug 2026).

NB: Many of the current OpenShift Virtualization success story and reference customers do not require HA for their applications, or use separate database and application DR strategies, in place of a proper infrastructure DR solution.

Acceptance Criteria

• Recover multiple VM machine workloads in a Metro-DR config
• Recover multiple VM workloads in an Regional-DR config

Out of scope

• DR for non ODF/ACM solutions, Like NetApp Astra, Dell, HPE
  • VMs on OpenShift are storage agnostic, separate features will be created for this future work.
• Complex disaster recovery architectures, such as chain or star
• DR for Highly Available applications (Pacemaker, Oracle RAC, Microsoft WSFC)