How to rebalance disk and adapter instance(s) in Aria Operations

Products

• VCF Operations/Automation (formerly VMware Aria Suite)

Issue/Introduction

After adding or replacing a node, it's important to rebalance disk and adapters across all cluster nodes to maintain optimal performance.

Environment

Aria Operations 8.x

Cause

• After adding or replacing an analytics node, disk usage across the cluster may become unbalanced. Although the rebalancing process will begin automatically after a short delay, in some cases it may be beneficial to initiate it manually to redistribute data more quickly.
• Adapter instances may not be evenly distributed across the cluster following node changes. To ensure optimal performance, it's recommended to manually rebalance the adapter load across all collectors.

Resolution

After successfully adding a new data node or replacing a node in Aria Operations you will need to:

• Ensure that all analytics nodes have equal amount of disk space for /storage/db filesystem. New nodes start with 250GB disks space for /storage/db, /storage/log, and /storage/core.
• You may be required to add additional storage, before rebalancing disks. Failure to verify and add disk space if required will result in cluster going down due to node out of storage.
• Rebalance the disk for all analytics nodes, follow the steps from the documentation: Cluster and Node Maintenance
• Rebalance all adapter instances equally across all analytics nodes, follow the steps from the documentation: Configuring Collector Groups

Note: Analytics nodes are: Primary, Primary-Replica, and Data nodes.

Additional Information

Aria Operations require that all Analytics nodes have close to the same amount of disk space. This is due to collected data being shared across the analytics nodes. If one node has less disk space available for /storage/db the cluster will go down when the node with the least amount of disk space will fill up before the nodes with more disk space available for the FSDB database.

 

The FSDB database will balance the data across all the analytics nodes in the cluster.

 

Run command df -h in console or SSH session when logged in with root account, to determine how much disk space is required.