This probably should be coordinated between the source and target cluster, such that when the source cluster completes uploading a new snapshot, the target cluster is instructed to apply the new snap to it's filesystem.
Brainstorm on what might be needed:
Architecture could be either 2 instances of SnapTool that talk to each other to coordinate OR
A single instance that can talk to both cluster's APIs
Initial throughts are that (1) would be more extensible such that a single source cluster could be replicated to multiple target clusters. (ie: filesystem A is replicated to India, filesystem B to China, or a single filesystem replicated to multiple locations?)
We need to add support for Synchronous Snaps (https://docs.weka.io/fs/snap-to-obj/snap-to-obj#sync-a-filesystem-from-a-snapshot)
This probably should be coordinated between the source and target cluster, such that when the source cluster completes uploading a new snapshot, the target cluster is instructed to apply the new snap to it's filesystem.
Brainstorm on what might be needed:
Initial throughts are that (1) would be more extensible such that a single source cluster could be replicated to multiple target clusters. (ie: filesystem A is replicated to India, filesystem B to China, or a single filesystem replicated to multiple locations?)