Monitoring events by using GUI

You can primarily use the Monitoring > Events page to monitor the events that are reported in the system. The status bar that is placed on the upper right of the GUI header, also displays the number of events that are reported in the system.

The events are raised against the respective component, for example, GPFS, NFS, SMB, and others. Some of these events might occur multiple times in the system. Such events are grouped under the Event Groups tab and the number of occurrences of the events are indicated in the Occurrences column. The Individual Events tab lists all the events irrespective of the multiple occurrences.

A graphical view of events that are reported against each component is available. Clicking the graph displays only the relevant events in the grid view. Clicking a section on the graphical view applies the corresponding filter on the search action and fetches only the relevant data in the events table.

You can further filter the events that are listed in the Events page with the help of the following filter options:
  • Current Issues displays all unfixed errors and warnings.
  • Current State lists active events that are generated because of state changes.
  • Notices displays the events that are not caused by a state change of an entity or component. Notices never become inactive on their own. Use the Mark Notices as Read action to make them historical when read.
  • All Events displays all the events irrespective of severity and type. It shows both active and historical events. The historical events are displayed with a grey-colored icon.

The severity icons help to quickly determine whether the event is informational, a warning, or an error. Click an event and select Properties from the Action menu to see detailed information on the event. The event table displays the most recent events first.

The following table describes the severity levels of events.

Table 1. Notification levels
Notification level Description
Error Error notification is sent to indicate a problem that must be corrected as soon as possible.

This notification indicates a serious problem with the system. For example, the event that is being reported might indicate a loss of redundancy in the system, and it is possible that another failure might result in loss of access to data.

Warning A warning notification is sent to indicate a problem or unexpected condition with the system. Always immediately investigate this type of notification to determine the effect that it might have on your operation, and make any necessary corrections.
Information An informational notification is sent to indicate the occurrence of an expected event. For example, a NAS service is started. No remedial action is required when these notifications are sent.
Note: A separate event type with severity Tip is also available. Tips are the recommendations that are given to ensure that you avoid certain issues that might occur in the future. The tip events are monitored separately in the Monitoring > Tips page of the GUI.

Marking events as read

The basic types of events can be categorized in the following manner.
  • State change events - The type of event that is generated when an entity changes its state.
  • The non-state change events - The type of event that is generated when there is no change in the entity's state.
The state change events are managed by the system and such events become inactive as soon as the state changes again. The non-state change events are referred to as notices in the GUI. Notices never become inactive on their own.

You must use the action Mark Selected Notices as Read or Mark All Notices as Read on the non-state change events. By using these actions, you can make them historical because the system displays those events as active events even if the problem or information is not valid anymore.

The result of mark as read operation is stored locally on the GUI node. That is, the changes that are made to the events are not visible through the other GUI nodes.

Resolving Event

Some issues can be resolved manually. To resolve events created for such issues, select the event and then click the Resolve Event option that is available under the Actions menu. On selecting the option, the mmhealth event resolve command is run to resolve the specific event. You can also right-click an event and select Resolve Event option from the drop-down menu that appears. On completion of the task, the status appears in the task window. The complete event thread can be viewed under the detailed view that you can access by using the View Details option.

Running fix procedure

Some issues can be resolved by running a fix procedure. To run a fix procedure, select Run Fix Procedure option that is available in the Actions menu.