You can use the Resource dashboard to monitor your environment by viewing resource metrics across primary resources with the option to drill down further to view the related, secondary resources. When you encounter an issue with your primary resource, you can troubleshoot the related resources to rule out issues at that level.
Table 1 includes a description of the standard monitoring widgets that are displayed for all resource types.
Depending on the resource type, other widgets are shown that are not common across all the dashboards but are specific to a resource.
Table 2 includes a description of some of the monitoring widgets that are displayed for other specific resource types such as Kubernetes.
|User interface item||Description|
|Events timeline and time span||The default display of the events timeline and chart views presents the past 12 hours. You can adjust the time span to show as few as 3 hours or as much as 1 week. If your Monitoring installation is configured for a data retention value of
2 days or more, 32 days at most, the time span options reflect the value that was set. For more information, see About data retention and summarization.
For Tivoli Monitoring data providers, a text indicator with the message data provider is online or data provider is offline is displayed on this timeline to show the status of Tivoli Monitoring resource data provider.
Red markers are a way to group events that occur in the same vicinity. The number indicates the number of events of the same type that are in close succession. Hover the mouse over an event marker to see when the event was opened and what triggered it. You can click the event to open the corresponding incident.
Drag the pin to move across time intervals and view metrics then. For example, if you want to see the metrics at the time an event (or events) occurred, drag the pin to that time.
|Related Resources widget||View the details and metrics of the secondary or related resources that are associated with your primary resources. Sort any of the following columns in ascending or descending order.
Enter any text to search for a resource, for example, enter LINUX and the LINUX resource is shown.
The status of the resource: Critical, Major, Minor, Warning, Indeterminate, Normal.
The relationship this related resource has with the one you are monitoring.
Click the resource link to drill down further to the resource dashboard to look at the metrics more closely.
Predefined groups are type System Defined. You have a predefined group for every type of agent that you installed in your environment. Custom groups that you or others in your environment define are type User Defined.
|Resource Properties widget||You can view the properties of the object in the monitoring topology service. Scroll through the properties and their values or type in the Filter box to locate a specific property, such as a node's osImage or a pod's qosClass.|
|Custom metrics widget||Click the Custom Metrics twistie and filter the metrics to view the Custom Metrics widgets that are common for all dashboards.
Use the Custom Metrics widget to explore the available collected metrics that aren't already reported in the dashboard line charts. You can display six additional metrics at most in one or two line charts:
Click the Custom Metrics twistie to expand the widget. The widget is typically the last one in the dashboard and shows two views side by side.
Select an Aggregation function. The aggregations that are available are: Average, Minimum, Maximum,Sum,Deviation.
Select a metric from the Filter metric list. If the metric has dependencies, a Filter dimension list is displayed. If the Filter dimension has a dependency, another Filter dimension list is displayed or a Dimension value list. The line chart is rendered after you select the required metric and dimension values.
If a Filter dimension list is displayed, select a dimension from the list. If the metric has multiple dimensions, a second Filter dimension list is displayed for you to select from.
If a Dimension value list is displayed, select one or more values.
After the required metric type, dimensions, values are selected, the line chart is rendered in the widget space.
Example: You might want to view other metrics that relate to the event and correlate these metrics with the standard dashboard metrics. When you view metrics side by side, you can correlate these two sets of metrics.
Depending on the resource type, other widgets are shown that are not common across all the dashboards but they are specific to a resource. For example; if a Linux server has a high CPU usage that caused the incident, you can choose to view a graph that shows the history and trend of CPU utilzation, or attributes about the server, or details of the processes that run on the system across various metric widgets.
|User interface item||Description|
|Line charts and the golden signals||Line charts plot metrics from the past three hours or as selected. Hover the mouse pointer over a plot point to see the value and time stamp. All the line charts in the dashboard are synchronized to show the same point in time as you move
the mouse pointer across one of the charts.
Some of the Kubernetes dashboards have a set of widgets for monitoring the four golden signals: Latency, Errors, Traffic, and Saturation. Latency and Errors typically indicate the symptoms that users are most likely to perceive. The causes behind them are usually Traffic and Saturation.
The Latency chart plots the latency in milliseconds. Drag the pin on the timeline or drag the vertical line on the chart to open the hover display for that time. : how long 99% of requests took to complete, how long 50% of requests took to complete, and how long 95% of requests took to complete. For example, a latency of 492 ms in percentile 99 means that 99% of requests took fewer than 492 ms to complete, 189 ms in percentile 50 means that 50% of requests took fewer than 189 ms to complete, and 492 ms in percentile 95 means that 95% of requests took fewer than 492 ms to complete.
There is a Path widget for filtering paths. If you have multiple end points and only one is performing badly, you can show the signals for only the requests in that path by clearing the check boxes of the other paths.
|Service Dependencies||The Service Dependency view in the Kubernetes Service dashboard shows what application is calling this service and what this service is calling, one step at a time. This view shows service-to-service relationships to help you debug issues
across the dependency tree. For example, if the symptoms presented in the Latency and Error line charts are bad but the Traffic and Saturation did not change, you can search this view to find out what is being called. Click a service to
open its dashboard.
If a Kubernetes service has dependent services, you can click on Expand/Collapse to open a richer topology view. This service dependency view embeds Netcool Agile Service Manager functionality. For more information on using this view, see Managing Applications.
|Kubernetes topology||Hover over the Kubernetes topology to see a pop-up note with status information about an object; click an object to open its dashboard. The topology widget in a Kubernetes Service dashboard displays an icon if the service provides ingress, and you can click the icon to open the associated Kubernetes Ingress dashboard.
In the Kubernetes Cluster dashboard, click the node to open the associated dashboard. You can drill down to each level in the cluster from the node dashboard to the pod or container dashboard, and from the pod dashboard to the container dashboard. To return to the node dashboard from the pod or container, click inside the hexagon.
From the Kubernetes service, node, pod, or container, you can click the area inside the outermost circle to jump to the cluster dashboard.