In my previous blog post, (
OpenShift Virtualization Backup and Restore with Trilio in AWS ROSA Baremetal ) I discussed how to protect VMs running in OpenShift Virtualization, specifically in AWS ROSA Baremetal environments.
In this post, I want to focus on how you can recover a VM in less than 3 minutes. When it comes to recovery in Kubernetes environments, many of us feel relatively at ease, especially if the Persistent Volumes (PVs) are not particularly large.
However, with virtualization now part of Kubernetes, we’re dealing with much larger data sizes.
Consider VMs with data that range from 300GB to 4TB in size — quite common in enterprise settings. How long do you think it would take to transfer 4TB of data between your backup storage and the recovery cluster?
Now, imagine needing to migrate or recover not just one VM, but 30 or even 300 VMs. This can pose a significant challenge.
In this blog post, you’ll learn about Trilio’s Continuous Restore technology, which allows for disaster recovery of VMs in under 3 minutes. We’ll explore how this innovative approach can streamline your recovery process and enhance your overall disaster recovery strategy – saving you time, resources and potentially people’s jobs.
How to enable Continuous Restore
In the previous blog post we configured everything in the OpenShift console under the Trilio Backups tab. This is really helpful for those end users or departments that operationally don’t want to leave the OpenShift console. In this occasion we are going to provide rapid recovery through the Trilio UI, so we can enable Trilio’s Continuous Restore feature at a Backup Plan level.
Certainly we can have multiple backup plans, and we don’t need to enable Continuous Restore in all of them. This is a question I get a lot when I explain this feature to customers. You can replicate many backup plans, but usually our customers enable that for their most critical workloads only, as obviously it consumes more storage and computing resources to enable this feature.
I recommend our customers to have as many backups as they wish in the backup target, and then have the latest 3 points in time, “pre-staged”, in the disaster recovery cluster.
If you want to read a bit more about this feature, go to this link 
Continuous Recovery & Restore , but we are not exactly what you would expect, because Trilio is cloud native, we don’t need to be “point to point”, and can provide flexibility for “one to many”, or “many to one” as well. For example, several active primary clusters, could be prestaging VMs to a single Disaster Recovery cluster. Or, you could be prestaging VMs to two different cloud providers, from your onprem cluster. TRUST NO ONE! 
So first things first, you need to do something at the target level.
Enable Event Target. This will enable some services from our side, that will implement the whole Continuous Restore strategy. Once that is done, in both clusters that we are going to “link”, then you can progress to the next step.
Second, once the Event Target enablement is done, you will head to Backup Plans, and in the Backup Plan where you want to enable Continuous Restore, the final page when you edit it, it’s the Continuous Restore section. It is optional, and if you don’t need it, just remove it, or directly click on Skip & Create.
On this occasion we will select the name of the Trilio instance in the cluster we are going to use as a DR cluster. In this screenshot the instance is called “Trilio”. In the right, we can set the policy. When we create a policy, we can establish up to 10 Consistent Sets, a set of PVs created as part of the restore process.
After saving this, and after some seconds, in the Backup Plan we will see this green icon, that means that CR is enabled for that Backup Plan.
How to check that we can restore from ConsistentSets
As soon as we enable a Continuous Restore policy, for example to “pre-stage” latest 3 backups, if we create a new backup, once the backup is completed, it will start “pre-staging” that data to the remote cluster. We could go to the Continuous Restore section in the Trilio UI, select the destination cluster, and see the “Inbounds”.
When this is completed, you will be able to see the ConsistentSets in the destination cluster.
How to restore from a ConsistentSet
Please note, you will always be able to restore from any of the backups you have in the backup target.
But, if you want to restore as fast as possible, you should restore from a ConsistentSet from Trilio. This is definitely the quickest option in the industry as of now.
And by the way, YOU CAN use transforms also when restoring from a ConsistentSet.
All progressing good.
Just in a little more than 3 minutes we have our VM restored.
And our VM is up and running on OpenShift Virtualization in AWS.
Conclusion
With the advent of large Persistent Volumes in Kubernetes/OpenShift, specially for Virtualization, recovery time from backups becomes a critical issue for everyone.
With Trilio’s Continuous Restore feature, you can leverage the lowest RTO in the market when restoring from backups.
