Learn how to set parameter thresholds as the baseline for cluster health.
Cluster Health parameters
Each container cluster is identified in the Clusters table and is designated healthy or unhealthy. This designation is derived from whether any of the health parameter values have exceeded a defined health threshold. The parameters are predefined as follows:
Node health
: Displays how many nodes are deemed unhealthy. Default threshold is 50%.
Critical alerts
: Displays how many critical alerts have been issued. The default threshold is 1.
Memory overcommitted
: Displays the amount of memory that could be consumed by individual containers as a percentage of the set memory limit for that cluster. Default threshold is 90%.
CPU overcommitted
: Displays the amount of CPU that could be consumed by individual containers as a percentage of the set CPU limit for that cluster. The default threshold is 90%.
As an administrator, you can tailor the default parameter thresholds to the specific requirements of your organization.
Setting parameter thresholds
Container Cluster Management provides a Cluster Health Settings page for setting health parameter thresholds. To access the Cluster Health Settings page, click the
Settings
button located at the top right of the page. The service navigates to the Cluster Health Settings page.
The Cluster Health Settings page contains a form for setting the thresholds for each of the four health parameters. The following caveats apply:
Users can enter values using the number input fields.
Each rule can be enabled or disabled using the toggle switch.
Click
Save
to apply changes.
Click
Reset
to revert to the previously saved values before saving.
Click
Cancel
to abort your changes and return to the Container Cluster Management landing page.