With the release of version 8.1.0, the replication engine has been completely replaced. Replication 2.0, generally referred to as ‘Resumable Replication’ is a total overhaul to the way replication is performed, providing both a more robust transfer mechanism and a 3x-5x improvement in end-to-end replication performance.
What happens after 8.1.0 is installed on both the appliance and vault?
- Once 8.1.0+ is installed on both your appliance and vault, Resumable Replication mode will be enabled.
- In order to complete the transition from Legacy mode to Resumable Replication, all of the existing snapshots for each protected system that have not yet been ingested have to be completed.
- Once this is done, replication will proceed directly to the ZFS pool, with no more intermediate staging step in the aristosbay on the vault.
- In order to facilitate this, once Resumable Replication is enabled, replication of snapshots in Legacy mode will pause until ingestion is completed on the vault. This can make it appear that replication is broken on the appliance, as the ‘spinning wheels’ will continue to show up in the Jobs menu and no new replication jobs will be sent until ingestion is complete.
- Once a given protected system is caught up, replication will proceed directly to ZFS pool storage, and no more ‘spinning wheel’ states will be encountered.
- Monitor the progress of overall ingestion by using the Health Report and/or Trouble Report pages on the Global Management Portal.
- Ingestion should move forward for at least some of your systems every day, and once each protected system is caught up, it will replicate in parallel with all the other systems. There will be no more single-threaded ingestion bottleneck.
How does Resumable Replication work when fully enabled?
Once Resumable Replication is fully enabled, the only time that Legacy replication will occur is when adding a new protected system via USB seeding.
In the case of a USB seed of the base image, the incremental backups will be staged to the aristosbay in real-time until the base image is fully ingested. At that point, replication will be paused while ingestion completes, then will automatically resume with direct ingestion to ZFS once all of the staged incremental backups are ingested.
What happens if a system fails to upgrade from 8.1.x?
There are a number of systems in the field that are failing to upgrade from previous versions to 8.1.x for various reasons.
Once the next hotfix release (8.1.2) is fully released, we will begin a concerted effort to address all remaining devices that have not been upgraded. Hotfix Release 8.1.2 contains a number of fixes and remediations that should address the bulk of the troublesome systems, but where necessary Axcient support will be reaching out to assist.
All systems should be on version 8.0.9 or newer at this point. If you have systems that have not yet upgraded, contact Axcient support for assistance at any time.