HP Smart Array Cluster Storage System

Troubleshooting
HP Smart Array Cluster Storage System User Guide D-13
HP CONFIDENTAL
Writer: Rob Weaver File Name: j-appd Troubleshooting
Codename: Aurora Part Number: 240333-003 Last Saved On: 11/6/02 1:07 PM
Hardware-Based Fault Tolerance
1. Identify and document which physical drive has failed. (On hot-plug drives in a
ProLiant server or storage system, this is indicated by an amber drive failure
LED on the drive tray.) Note the drive type and capacity.
NOTE: Storage systems using hardware-based fault tolerance in NetWare cannot detect
failure of a single physical drive. In this case, the data will still show as valid and
accessible during the rebuilding process. However, the driver will have registered that a
physical drive has failed, and a message will be displayed notifying the user that a logical
drive is in a degraded state. CPQONLIN will also show the drive has failed.
2. Note which partition and volume, if any has failed. This information is provided
in the error message on the server console. It is also recorded in the server error
log file, which can be viewed using the NWADMIN Utility.
3. Remove the failed drive and replace it with a drive that is of the same type and
capacity. For hot-plug drives, after you secure the drive in the bay, the LEDs on
the drive each flash once in an alternating pattern to indicate that the connection
was successful. The online LED flashes, indicating that the controller recognized
the drive replacement and began the recovery process.
4. Power up the storage system, if it was turned off in step 3.
5. The array controller firmware rebuilds information that was on the failed hard
drive onto the new drive, based on information from the remaining physical
drives in the logical drive. While reconstructing the data on hot-plug drives, the
online LED flashes. When drive rebuild is complete, the online LED is
illuminated.
No Fault Tolerance
If you configured the system for no fault tolerance, data must be recovered from
backup media. Perform the following steps:
1. Record the device number and device name of the failed logical drive. This
information is shown on the server console and recorded in the server error log
file, which may be viewed using the NWADMIN Utility (4.x). For example:
NWPA: [V503-A2-D1:0] Compaq SMART-2 Slot 8 Disk 2 NFT
You will use this information later to create a valid partition.