Purpose

Provide a troubleshooting reference for engineers when a BeeGFS Capacity Critical alert is triggered in the cluster.

This document outlines the meaning of the alert and the actions required by L1 and L2 engineers.

When to Use

Use this KB article when the monitoring system generates the following alert:

Alert Name: BeeGFS Capacity Critical

What the Alert Means

This alert indicates that BeeGFS storage capacity has reached a critical threshold.

This may occur due to:

• High storage consumption

• Large files written to the filesystem

• Stuck or deleted files consuming space

• Storage servers (OSS) being offline

• Metadata or storage thread bottlenecks

If not addressed, this may result in:

• Job failures

• Storage write failures

• Performance degradation

Severity

Critical

Immediate investigation is recommended to prevent storage exhaustion.

L1 Engineer Actions

Follow the steps below when the alert is received.

Step 1 – Acknowledge the Alert

• Record the alert in the Alert Monitoring System

Step 2 – Verify Alert Details

Check the following:

• Affected filesystem

• Node generating the alert

• Current storage usage

Step 3 – Initial Monitoring

• Confirm whether the usage spike is temporary

• Monitor storage usage for a short duration

Step 4 – Escalate if Required

If the alert persists:

• Escalate to L2 Engineer

• Provide alert details and timestamps

L2 Engineer Actions

Perform deeper investigation once the alert is escalated.

Step 1 – Check OSS Server Status

Verify whether any OSS server is down or offline.

Step 2 – Verify Metadata / Storage Threads

Check for issues with:

• Metadata service

• Storage threads

Step 3 – Check Filesystem Usage

Run the following command:

df -h

Check usage on the BeeGFS mount point.

Step 4 – Identify Large Directories

Locate large directories consuming space.

du -sh *

Step 5 – Check for Stuck or Deleted Files

Investigate files that may still be consuming storage after deletion.

Step 6 – Coordinate with Storage Team

If required:

• Coordinate with NetApp team

• Investigate backend storage issues

Step 7 – Remediation

Take corrective actions such as:

• Cleaning unused data

• Moving data

• Expanding storage capacity

Expected Outcome

After corrective action:

• BeeGFS storage usage should fall below the critical threshold

• Alert should automatically clear in the monitoring system