Self-monitoring with topology
You can create the topology visualization of your Cloud Pak for AIOps instance to monitor the errors and warnings and to view all the resources, which make up the Cloud Pak for AIOps installation.
With the topology, see how different aspects of the estate are related to one another. Also, use it for more advanced analytics, such as topology-based alert correlation, probable cause determination, inventory management, and more.
Since Cloud Pak for AIOps runs on Red Hat OpenShift Container Platform, you need a single observer to get started, namely the Kubernetes observer. The Kubernetes observer works with many of the Kubernetes-based platforms, such as Red Hat OpenShift and Google Kubernetes Engine.
To configure a Kubernetes integration, you need the following information:
- Kubernetes cluster or data center name
- Namespace where Cloud Pak for AIOps is installed
- Kubernetes master IP address
- Kubernetes API port
- Kubernetes token
Complete the following steps to configure the self-monitoring by using the Kubernetes integration:
1. Create and use a Service Account.
To create and use a Service Account, follow the steps that are listed under the topic Kubernetes Observer Load job prerequisites.
2. Configure the observer.
-
Log in to IBM Cloud Pak for AIOps console.
-
Expand the navigation menu (four horizontal bars). Then, click Define > Integrations.
-
On the Integrations page, click Add integration.
-
From the list of available integrations, search Kubernetes and click the Kubernetes tile.
-
Click Get started.
-
Select Load as job type.
-
Enter the values for the Unique ID, data_center, Kubernetes Master IP Address, Kubernetes API port, and Kubernetes token fields.
Figure. Kubernetes observer job -
Under the additional parameters (optional) tab, enter the namespace where Cloud Pak for AIOps is installed so you observe only the namespace and not the other applications installed on the cluster.
Figure. Additional parameters -
Under the Job schedule (optional) tab, indicate how frequently you want to refresh the information. For example, refresh every 5 minutes.
For more information about all the fields, see Configuring Kubernetes Observer jobs.
3. View the topology.
When the observer job runs for the first time, it creates a resource group by using the name that is provided in the observer data_center field.
The observer brings in the alerts, errors, and warnings, which are displayed under the Resource groups tab.

When you click the resource group name that is created by the observer job, you can view all the resources, which make up the Cloud Pak for AIOps installation. If the cluster is updated, or the Cloud Pak for AIOps is updated, the changes are shown as displayed in the following example.

4. Create topology resource groups.
The view of the entire set of resources for Cloud Pak for AIOps is useful, but it might be useful to look at specific portions. For example, you can view each node to see the workload of the node and any issues associated with that node.
To create groups for each of the nodes, you can use the dynamic resource group template. For more information, see Building resource group templates.
Use the following steps to create topology resource groups:
-
From the main navigation of the Cloud Pak for AIOps console, click Resource management.
-
Click Resource group templates icon.
-
Click Create a new template > Dynamic template.
-
Click Start. The Template builder page appears.
-
In the search box, enter the resource name, for example,
worker1.data-center1.example.com
, and click the desired resource.Figure. Template builder -
To see both the node and the associated workload, right-click the large node image as shown in the following example, and select Get Neighbors.
Figure. Render template When you hover over the large node image, you see the full name of the resource and the type of the resource.
-
Click All. This option shows the pods that are running on the node and the information such as the IP address.
Figure. Workload of node -
Provide a Template name and description.
-
Select the resource group type. For example, compute.
-
Optionally, add a tag name. For example, self-monitoring. Adding a tag helps with the filtering of results that are listed under the resource groups.
-
Click Save template & generate resource groups.
Cloud Pak for AIOps creates the template and automatically creates the resource groups for all the nodes, each showing the workload as seen from the following example.
