This topic describes how to use the container monitoring module to view the network topologies for the Services and workloads in a Kubernetes cluster.

Go to the pages on which you can view network topologies

  1. Log on to the ARMS console.
  2. In the left-side navigation pane, click Container Monitoring.
  3. In the top navigation bar of the MNS console, select the region where your cluster is deployed.
  4. On the Container Monitoring page, click the name of the Kubernetes cluster whose network topologies you want to view.
  5. In the left-side navigation pane, click Cluster network topology.
    By default, the network topology of Services in all namespaces appears.
  6. Optional:Click the Workload tab to view the network topology of Deployments, StatefulSets, and DaemonSets.

Supported operations

In the upper part of a network topology, you can perform the following operations:
  • In the date and time picker in the upper-right corner, select a time range such as last 30 minutes, last 1 hour, or last 12 hours.
  • From the Node Filtering drop-down list, select a namespace and a label to filter the resources. You can click the name of a resource to identify the resource in the topology and view the resource details in the right-side details panel. For more information, see Resource details.
On the canvas of a network topology, you can perform the following operations:
  • Click the Expand icon icon in the upper-right corner of the topology box for a namespace to show all the Services in this namespace. Then, you can click the Fold icon icon to hide the Services.
  • Move the pointer over a resource to view the number of requests, number of errors, and response time of this resource in the lower part of the canvas.
  • Move the pointer over the line between two resources to view the number of requests, number of errors, and response time between these resources in the lower part of the canvas.
    Note If an error occurs between two resources or the response time between them is longer than 500 ms, the connection between the two resources is abnormal. In this case, the line is displayed in red.
  • Click a resource to view its details in the right-side details panel. For more information, see Resource details.
  • Click the line between two resources to view its details in the right-side details panel. For more information, see Details of a line between resources.
  • Right-click a resource and select View container layer (3D) to view the 3D topology at the container layer for this resource. For more information, see View the 3D topology at the container layer.
  • Right-click a resource and select Only upstream, Only downstream, or Only upstream and downstream to view the topology between the current resource and its parent resource, the topology between the current resource and its child resource, or the topology among the current resource, its parent resource, and its child resource.
  • On the canvas, right-click the blank area and select Fold all or Expand all to hide or show the resources in all namespaces.
In the lower part of a network topology, you can perform the following operations:
  • Click the Zoom In icon icon or scroll the mouse wheel up to zoom in the topology.
  • Click the Zoom Out icon icon or scroll the mouse wheel down to zoom out the topology.
  • Click the Expand Topology icon icon to show the resources in all namespaces.
  • Click the Fold Topology icon icon to hide the resources in all namespaces.
  • Click the Adjust Topology icon icon to adjust the topology size to fit the page.
  • Click the Refresh icon icon to refresh the topology.

Network topology of Services

The network topology of Services displays the topological relationships among the Services in different namespaces.

Container monitoring - Service topology

After you hide the Services in a namespace, the box for this namespace shows the number of Services in the namespace. The circle that encloses the number is colored to show the distribution of normal and abnormal Services. Yellow indicates abnormal Services, whereas blue indicates normal Services.

After you show the Services in a namespace, the box for this namespace displays these Services and the number of requests for each Service. If an error occurs within a Service or the response time of the Service is longer than 500 ms, the Service is considered abnormal. In this case, this Service is displayed in yellow.

Network topology of workloads

The network topology of workloads displays the topological relationships among the deployments, StatefulSets, and DaemonSets in different namespaces.

Container monitoring - workload topology

After you hide the workloads in a namespace, the box for this namespace shows the number of workloads in the namespace. The circle that encloses the number is colored to show the distribution of normal and abnormal workloads. Yellow indicates abnormal workloads, whereas blue indicates normal workloads.

After you show the workloads in a namespace, the box for this namespace displays these workloads and the number of requests for each workload. If an error occurs within a workload or the response time of the workload is longer than 500 ms, the workload is considered abnormal. In this case, this workload is displayed in yellow.

Resource details

Click a resource in a network topology to view the details of the resource in the right-side details panel.

The right-side details panel displays the statistics on and the time series curves of the number of requests, number of errors, average response time, and number of instances. The right-side details panel also displays the time series curve of the number of slow calls.

container monitoring allows you to view the following protocol-related metrics: the number of requests, response time, number of errors, and number of slow calls.

Click the name of a protocol to view the time series curves of the preceding metrics related to the protocol. For more information about protocol-related metrics, see Protocol-related metrics.

Details of a line between resources

Click a line between two resources in a network topology to view the details of the line in the right-side details panel.

The right-side details panel displays the statistics on and the time series curves of the number of requests, number of errors, and average response time. The right-side details panel also displays the time series curve of the number of slow calls.

container monitoring allows you to view the following protocol-related metrics: the number of requests, response time, number of errors, and number of slow calls.

Click the name of a protocol to view the time series curves of the preceding metrics related to the protocol. For more information about protocol-related metrics, see Protocol-related metrics.