Overview

MindCluster Ascend FaultDiag (fault diagnosis tool) provides the diagnosis function for slow nodes and slow network faults in a cluster.

Before You Start

Before using the slow node and slow network fault diagnosis function, you need to increase the CPU and memory resources in NodeD and modify the resource information in the NodeD startup YAML file.

Original YAML file content:

resources:
            requests:
              memory: 300Mi
              cpu: 500m
            limits:
              memory: 300Mi
              cpu: 500m

Modified YAML file content:

resources:
            requests:
              memory: 10Gi
              cpu: 5000m
            limits:
              memory: 10Gi
              cpu: 5000m

Deployment Mode

ClusterD and the Fault Diagnose Online (FD-OL) framework are deployed in one process on the management node. Once ClusterD is started, FD-OL is automatically started.

Parent topic: Slow Node and Slow Network Faults