10.5 HP StoreVirtual Storage User Guide (AX696-96269, March 2013)

1. In the Edit Cluster window, select a storage system from the list.
2. Click Remove Systems.
The storage system moves out of the cluster, but remains in the management group.
3. Click OK when you are finished.
NOTE: Removing a storage system causes a full cluster restripe.
Troubleshooting a cluster
Auto Performance Protection monitors individual storage system health related to performance
issues that affect the volumes in the cluster.
Repairing a storage system provides a way to replace a failed disk in a storage system and
minimize the time required to bring the storage system back to normal operation in the cluster with
fully synchronized data.
Auto Performance Protection
If you notice performance issues in a cluster, a particular storage system may be experiencing slow
I/O performance, overload, or latency issues. Identify whether Auto Performance Protection is
operating by checking the storage system status on the storage system Details tab.
Auto Performance Protection is indicated by two unique statuses reported on the Details tab. You
will also receive event notifications on these statuses. For information about setting up event
notification, see “Configuring events” (page 91).
Storage System Overloaded—Indicates that operations to the storage system are completing
too slowly. In this state, volume availability is maintained while the storage system is
quarantined in the cluster. While the storage system is quarantined it does not participate in
I/O, which should relieve the performance degradation.
After the operations return to normal (in 10 minutes), the storage system is returned to active
duty and resynchronized with the data that has changed since its quarantine. Volumes that
depend on this storage system will then show “Resyncing” on the volume Details tab.
Storage System Inoperable—Indicates that the storage system is unable to repair the slow
I/Os, which may indicate a potential hardware problem. Volumes that depend on this storage
system are unavailable. For information about how to determine volume availability, see the
section “Determining volume and snapshot availability” (page 25).
Rebooting the storage system may return the status to Normal.
Auto Performance Protection and the VSA
The VSA will not report Storage System Overloaded, because there is no way to determine what
may be affecting I/O on the underlying hardware. However, the VSA can accurately report when
I/Os are not completing, and can return the Inoperable status.
Auto Performance Protection and other clusters
Auto Performance Protection operating on a storage system in one cluster will not affect performance
of other clusters in the management group.
To identify whether Auto Performance Protection is active on a storage system in a cluster with
performance issues:
Troubleshooting a cluster 141