Manage a Service Issue

Overview to understand and manage a service issue

When Sonexion nodes experience problems, the Dashboard tab heat map will indicate by color coding which nodes are experiencing degradation or critical/failed issues. For example, degraded nodes will be shown in amber on the heat map; critical or failed nodes will be shown in red. An email alert will also be sent to system administrators who are set up to receive them (see the Node Status Notifications topic).

When a problem appears on the heat map or an alert notification is received, the system administrator will acknowledge the alert, and determine the issue and possible cause.

Here are some suggested steps to review the service problem details to identify the issue and possibly take corrective action.

  1. With the Dashboard tab selected, click a node square that indicates degradation or failure.
    In the Node Details box, make note of the hostname of the node that is experiencing problems.
  2. Navigate to the Health tab.
  3. Review event logs see if there is any information that will help diagnose the problem.
  4. Click Hostgroup Overview in the left panel menu to display the Status Overview for All Hosts view.
  5. Find the entry for the node whose hostname was identified in step 1. The same node may appear under more than one host group.
    Tip: Look for the same amber or red color that was displayed in the heat map. The color may appear in either the Status or Services column.
    1. Click the blue document icon (View Extended Information for this Host) if the Status column is colored. This will display host state information, which may help identify the problem.
    2. Click the purple page document icon (View Service Details for this Host) if the Services column is colored. This will display the service status details for the host, which may help identify the problem.
    3. Click the gear icon (Perform Extra Host Actions) to view the host service overview with graphs and metrics.
The preceding suggested steps describe only one (1) possible method to identify a reported issue and diagnose the problem. There are typically multiple options and pathways to access the same data and analyze the health and status of the system. Use whatever approach makes the most sense in a given situation.

If a hardware component needs replacement, contact Hewlett Packard Enterprise support to schedule a service appointment.