Diagnosing the Cluster Network

This section describes how to manually diagnose the network health status so that you can proactively detect and quickly locate network problems. In addition, the system displays information about the detected subhealthy links.

Prerequisites

The network subhealth check function has been enabled. By default, this function is enabled for the system.

To query and enable the network subhealth check function, perform the following steps:

  1. Log in to the management node as user dsware and run the following command to check whether the network subhealth check function is enabled:
    /opt/dsware/client/bin/dswareTool.sh --op queryNetworkSubhealthSwitch
  2. If the network subhealth check function is disabled, run the following command to enable it:
    /opt/dsware/client/bin/dswareTool.sh --op setNetworkSubhealthSwitch -s open

Context

Cluster network diagnosis complies with different protocols, as listed in Table 1.

Table 1 Cluster network diagnosis protocol table

Diagnosis Type

Diagnosis Protocol

Subhealthy Links

Perform diagnosis based on the actual network protocols, such as RoCE, IB, and TCP.

In-Depth Diagnosis

E2E Diagnosis

Perform diagnosis based on the TCP ICMP protocol.

Plane Diagnosis

Management plane

Perform diagnosis based on the TCP ICMP protocol.

Control plane

Perform diagnosis based on the TCP ICMP protocol.

Storage plane

Perform diagnosis based on the actual network protocols, such as RoCE, IB, and TCP.