Hardware Problems
2-30
HP Servers Troubleshooting Guide
HP CONFIDENTIAL
Writer: Ted Weiman File Name: c-ch2 Hardware Problems.doc
Codename: Generic TSG Part Number: 161759-007 Last Saved On: 11/19/02 9:31 AM
Predictive Failure Alert
The predictive failure alert is a powerful problem-prevention tool that warns you
when the system has determined that a drive failure is imminent. This alert allows
you to proactively schedule downtime for maintenance and not interrupt critical
business operations that rely on your server. In addition, with hot-pluggable drives
connected to Smart Array Controllers you can remove and replace one or several
drives within a server while the system is online. This minimizes the interruption of
the network, server downtime, and data loss. Refer to your Insight Manager 7 (or
previous version) and Management Agents documentation, found on the
Management CD, for instructions on implementing this function.
CAUTION:
Failure to follow these guidelines can result in data loss.
IMPORTANT:
It is recommended that some level of fault tolerance be utilized in your RAID
configuration. Refer to the Smart Array Controller user guide for information on fault-tolerance
options.
IMPORTANT:
You must use Insight Manager 7 (or previous version) and a Smart Array
Controller to manage the drive array on your server if you wish to implement the predictive
failure alert.
Predictive Failure Replacement Guidelines
To minimize server downtime and data loss, use these guidelines when Insight
Manager 7 (or previous version) issues a predictive failure alert. The alert indicates
that a drive is degraded and should be replaced.
•
Be sure that all physical drives in the affected array are present and that the
online LEDs are illuminated before removing the degraded hot-plug drive. If any
online LEDs are flashing (indicating a rebuild) or are not illuminated, the
degraded drive should not be removed.
For step-by-step instructions on connecting a hot-plug hard drive, refer to the
user documentation for your server.
•
If you are upgrading to larger drives in the array, follow the previously stated
rules and be sure that each drive has completed its rebuild before adding the next
new drive to the array.