Alerts

To ensure the continuous availability of the HCP repository:

A strict majority of the HCP storage nodes must be running and healthy. For example, a six-node system requires at least four running nodes.

If this condition is not true, namespaces cannot accept write requests, including requests to store new data or change object metadata.

At any given time, the unused HCP storage space must be sufficient to allow the creation of new objects and to ensure proper behavior in the event that a node fails.

The top section of the Overview page serves as an early warning system for violations of these requirements. It also reports system configuration issues and provides other system status information, such as notification that the system is rebalancing metadata. When the system is functioning normally, this section indicates that no problems exist.

The information in this section is represented by icons that are associated with text. These are called alerts. To see the text that accompanies an alert icon, mouse over the icon. The text appears below the icon.

This section contains three categories of alerts:

Hardware Status — Alerts in this category identify nodes that are reporting problems relating to drives, network interface cards (NICs), HCP S Series Node tiering, and other internal components.

For more information on the types of problems reported, see Hardware status.

System Status — Alerts in this category report on storage usage and indicate systemwide conditions such as a high rate of page swapping.

Note: When the amount of primary storage that’s used reaches 90 percent of primary storage capacity, HCP will not store any new objects or any new or replacement custom metadata or ACLs. HCP will, however, continue to allow changes to system metadata.

In a SAIN system with primary spindown storage, the 90 percent capacity limit applies to primary running storage usage only.

Tip: If the amount of primary storage that’s used is approaching 75 percent of primary storage capacity, consider adding more storage capacity to HCP. By adding capacity before the 75 percent threshold is reached, you can prevent interruptions in namespace availability due to insufficient space.

In a SAIN system with primary spindown storage, this tip applies to primary running storage usage only.

Object Integrity — Alerts in this category report on the state of object data and metadata. They also identify situations in which the protection or content verification service is fixing errors or has found errors it cannot fix.

The background color in an alert icon indicates the status of the relevant component:

A green background indicates that the component is within normal operating parameters.

An orange background indicates that the component is disabled or needs attention in some way. The text that accompanies icons with an orange background appears automatically below the icon area. This text may contain links to Console pages that you can use to view more information about the problem and, in some cases, change one or more specific configuration settings in order to resolve the problem. For example, the text may contain a network name that’s a link to the page on which you can view and change the IP configuration settings for that network.

A red background indicates that the component is malfunctioning. The text that accompanies icons with a red background appears automatically below the icon area. This text may contain links to more information about the problem and, in some cases, change one or more specific configuration settings in order to resolve the problem. For example, the text may contain a node number that’s a link to more detailed information about a problem with the hardware on that node.

For more information on the alerts that can appear on the Overview page, see Overview page alerts.

Note: If the entire HCP system is read-only due to metadata unavailability, a message about the situation appears at the top of every System Management Console page. While the system is read-only, configuration changes and changes to namespace content are not allowed. Additionally, in this situation, statistics that describe repository content may be inaccurate.

If the message says that the system has limited functionality and the situation persists, contact your HCP support center for help.

Trademarks and Legal Disclaimer

© 2017 Hitachi Data Systems Corporation. All rights reserved.