Heads Page

The Heads page is available via the left navigation panel under Cluster > Head Nodes. You can also access the Heads page for a specific head node via the Overview page's ClusterWare Disk Usage panel. The panel has a pulldown menu on the upper right where you can select a cluster head node to display.

In example above, there are three headnodes. The selected head node displays details for the selected head node's ICE ClusterWare™ disk usages for each of the various types of ClusterWare entities: ISOs, images, boot configurations, Git repositories, "other", influxdb data, and the etcd database.

Note

These are just the proportional disk usages for ClusterWare entities, not for the node as a whole.

In the ClusterWare Disk Usage panel, click the Manage Head Nodes link in the lower right to open the Heads page, which displays details about every cluster head node.

The Heads page provides a view of basic head node status and allows users to copy information about the ClusterWare version or the public key that the head node uses to connect to compute nodes. The version string is useful when reporting an issue to Penguin Computing support.

Although each head node's pie chart is likely to display small differences in the sizes of the pie components, you can hover the cursor over specific components to see their absolute sizes and note that the shared objects (ISOs, Images, Boot Configs) show the identical sizes across the head nodes.

The "Other" category consists of files in the ClusterWare storage directory that are not recognized by the system. These are usually files left behind during partial uploads, interrupted image cloning, or other failure cases. If the cluster is working as expected though there is space consumed by this "Other" category, those files can be removed via the "Clean storage/" option in the More (...) menu at the top of each head node panel. This action is equivalent to executing the following command:

scyld-clusterctl heads -i <HEADNODE> clean --files