Operational considerations for third‑party backup
When using NetBackup with MongoDB Ops Manager, consider the following operational guidelines:
Use only one backup solution per cluster. Do not run native Ops Manager backup and third‑party backup on the same cluster.
Do not initiate snapshot backups during a restore operation. Start snapshot backups only after the restore is complete and the cluster is running normally.
At least one node should be selected as a preferred node within a replica set/shard.
Do not use multiple third‑party backup tools on the same cluster.
Monitor disk usage for the transaction log directory to ensure that sufficient space is available.
If the transaction log directory is misconfigured or becomes unavailable, errors such as missing transaction logs may occur, which can reduce or break the point‑in‑time recovery window.
Granular recovery is not supported using third party APIs.
Following the recovery, MongoDB cluster is reverted to the state of the chosen backup images.
For restore operations, install the NetBackup client on all nodes of the cluster where the restore will be performed. This applies to both the original cluster and any alternate cluster used for restore.
Ensure that the NetBackup clients are installed on all the nodes of the target cluster.
If TLOG backups are not configured or have gaps, PIT restore cannot be performed and recovery is limited to the available snapshot restore points (FULL/INCR).