Skip to main content
Use these guides to view cluster health, inspect GPU and Slurm signals, understand alerts, and gather what you need for Andromeda Support.

​
Start here

​
Common tasks

View dashboards

Find GPU Nodes, Job Analysis, and Tenant Dashboard panels.

Check metrics

Review GPU, node, container, Slurm, InfiniBand, Weka, and Kubernetes metrics available in your dashboards.

Review alerts

Understand alert tiers, Slack behavior, alert families, and expected alert chains.

Troubleshoot symptoms

Use scoped diagnostic queries for GPU, node, InfiniBand, Slurm, and container symptoms.