Raid Setup; Raid Management And Operation; Critical And Offline Arrays - Promise Technology SuperTrak SX6000 User Manual

SuperTrak SX6000™ User Manual
This chapter discusses the various types of RAID arrays, their applications, including
performance and reliability considerations, and general maintenance information. Given
the variety of applications for RAID systems, the various RAID levels each have
advantages and disadvantages.
Determining what RAID level to use for your system is an important step. To help guide
you, we have divided this chapter into two parts:
•
RAID Management and Operations covers various fault scenarios,
rebuilding/synchronization of drives, and drive partitioning/formatting, which may
be encountered during typical use.
•
Configuration Ideas cover organizational models of array systems that use the
previously documented array types. It also points out additional advantages of
various options.
This is certainly not intended as a complete, nor definitive reference on RAID
technology applications, but merely serves as a first step to understanding the
principles and proper usage of RAID. Please see Appendix A: Technology Background
for more details on the functionality of RAID systems.

RAID Management and Operation

This segment covers issues pertaining to general RAID system management,
operation, and terminology.

Critical and Offline Arrays

A fault tolerant array goes "critical" when a drive is removed or fails. Due to the fault
tolerance of the array, the data is still available and online. However, once the array
goes critical, the array has lost its fault tolerance, and performance may be adversely
affected. If the fault was caused by a drive that was removed, the drive should be
replaced by another drive, either identical or larger, in order for the RAID system to
rebuild and restore optimal configuration. Only a failed drive should be removed from
the array. Removing any other drive may cause the array to become unavailable.
When notified either by the SuperTrak SX6000 BIOS or the PAM utility that a drive has
failed, note the channel number of the failed drive to determine the physical location of
the failed drive. Figure 1 on page 6 shows the channel numbers for the SuperTrak
SX6000.
In arrays where a hot spare is already installed, the hot spare automatically replaces
the failed drive. When the failed drive is replaced, the new drive becomes the new hot
spare.
A non-fault tolerant array goes "offline" when a drive is removed or fails. Since the array
is not fault tolerant, the data stored in the array is no longer accessible. If the drive was
removed, then it should be replaced to restore accessibility to the array. If the drive
failed, then the entire array must be deleted and re-initialized since all data is
considered lost.

RAID Setup

- 33 -
Chapter 4