About alerts

The appliance uses alert messages to report issues with the resources it manages and monitors. The resources generate alerts to notify you that some meaningful event occurred and that an action might be required.

An event describes a single problem or change that occurred on a resource. For example, an event might be an SNMP trap received from a server's (iLO) management processor.

Each alert includes the following information about the event it reports: severity, state, description, and urgency. You can clear alerts, assign owners to alerts, and add notes to alerts.

While alerts have an active or locked state, they contribute to the overall displayed status for a resource. After you change their state to Cleared, they no longer affect the displayed status.

You can filter the display of the activity list to help you locate and view specific activities. However, you cannot customize filters to otherwise manage the display of the activities.

  • The appliance keeps a running count of incoming alerts. At intervals of 500 alert messages, the appliance determines if the number of alerts has reached 75,000. When it does, an auto-cleanup occurs, which deletes alert messages until the total number is fewer than 74,200. When the auto-cleanup runs, it first removes the oldest cleared alerts. Then it deletes the oldest alerts by severity starting with the least severe.

  • While many alerts are automatically cleared when the issue resolves, the original alert must be cleared manually. The original message is retained so that the administrator is made aware of the incident; otherwise events might occur, and go undetected.

    For example, let us say that someone unplugs a cable in the data center and plugs it back in. In this event, HPE OneView displays an alert that the connectivity on the uplink was lost and then that it was restored. However, the original alert "Connectivity lost on uplink X" stays in the Activity log until an administrator manually clears the alert.