Recovering from disk failures, About raid types, Detecting and monitoring disk failures – Dell PowerVault DL2000 User Manual
Page 31
Recovering From Disk Failures
About RAID Types
The internal disk storage containing the operating system and applications in the DL2
x
00 system is preconfigured as
RAID 1. RAID 1 is a mirrored disk group that provides uninterrupted operation of the appliance if a single disk fails within
that disk group.
CAUTION: A failed system disk must be replaced immediately to avoid data loss or complete failure of the
appliance.
The optional data disks in the PowerVault DL2100 and DL2200 systems and storage enclosures are automatically
configured by the backup application in RAID 5 or RAID 6 disk groups. See
.
NOTE: Automatic configuration only occurs if the disk configuration state is Standard and if you select one of the
automatic disk configuration policies. See
.
RAID 5 provides uninterrupted operation of the appliance if a single disk fails within that disk group. The RAID 5 standard
configuration includes a global hot spare in each storage enclosure and PowerVault DL2200 systems. If a single disk
failure occurs, the hot spare is automatically assigned and rebuilt into that disk group. After the disk group is rebuilt, the
RAID 5 disk group continues to operate normally.
CAUTION: If another disk fails while the internal or external disk group is rebuilding, all data in that disk group is
lost. Always monitor the status of the appliance's disk rebuild to ensure that the rebuild operation completes.
When the failed disk is replaced, the data residing on the global hot spare is automatically copied back to the
replacement disk. After the copy-back operation is complete, the global hot spare is once again available if another disk
fails. For more information about using global hot spares, see the
Dell OpenManage Server Administrator User's Guide
at support.dell.com/manuals.
NOTE: Only one disk group per storage enclosure can utilize a hot spare at a time.
NOTE: There are no dedicated hot spares for internal disks in the DL2000, Dl2100 and DL2300 systems.
NOTE: The DL2200 systems have dedicated hot spares for internal disks.
RAID 6 provides uninterrupted operation of the appliance if up to two disks fail within that disk group. Replacement of
the failed disk(s) causes the disk group to rebuild. If a third disk fails before the disk group rebuild is complete, data loss
occurs. The standard RAID 6 configurations do not include a hot spare.
RAID 10 provides high redundancy and throughput by combining RAID 1 and RAID 0 to stripe across four or more even
numbered mirrored disks. The entire capacity of the internal auto-provisioned RAID10 disk group(s) is assigned for use
as the CommVault Deduplication Database.
Detecting And Monitoring Disk Failures
To optimize the protection of RAID configured storage on the appliance, it is essential that you detect hard drive failures
immediately. Replacing failed disks immediately reduces the risk of unplanned downtime and loss of critical data.
You can use the following methods to monitor the health of the disks:
•
Hard drive LED indicators — Each drive in the server and storage enclosure contains status LEDs to indicate
drive activity and certain failure status. See the
Hardware Owner’s Manual
for the servers and storage
enclosures at support.dell.com/manuals.
•
Dell OpenManage Server Administrator — Server Administrator can monitor the health of both internal and
external direct-attached disks. See
.
31