beautypg.com

HP ProLiant DL380 G5 Server User Manual

Page 45

background image

IMPORTANT:

Never turn off a storage system when the attached system is still turned on.

Check the physical proximity of the system to other electrical devices. Since electrical noise may cause this

error, check the AC circuit for other electrical devices.
Timeouts can be caused when two or more drives are set to the same SCSI ID. Ensure that the storage

system and system SCSI IDs do not conflict.

On a storage system, check the SCSI ID cable on the drive tray. If the cable is damaged or incorrectly

installed, SCSI Timeouts can occur. See the documentation accompanying the Hot Plug Drive Tray Service

Spare Kit.

Ensure that the system temperature is within specified limits. Ensure that the fans are operating and are not

blocked.
In some instances, drive failure can cause Timeouts. If you continue to receive many of these errors, replace

the drive.

SCSI Bus Faults—Displays the number of times that SCSI bus parity, overrun, or underrun errors have been

detected on the SCSI bus. Since the controller retries the operation, SCSI bus faults can cause a drop in

performance, or, in some cases, data corruption.
If the count is not zero and the drive has failed, the failure might be correctable without replacing the drive.

Verify the status of the drive by checking the following:

Ensure that all system and storage system cables are intact and seated properly. You may need to replace

the cables.

Check the physical proximity of the system to other electrical devices. Since electrical noise may cause a

Bus Fault error, check the AC circuit for other electrical devices.

Ensure that the system temperature is within specified limits. Ensure that fans are operating and are not

blocked.
SCSI Bus Faults can be caused when two or more drives are set to the same SCSI ID. Ensure that storage

system and system SCSI IDs do not conflict.
In some instances, drive failure can cause SCSI Bus Faults. If you continue to receive many of these errors,

replace the drive.

IRQ Deglitch—Displays the number of times that a glitch has been detected on the drive interface cable. Since

the controller retries the operation, problems can cause a drop in performance or, in some cases, data

corruption. Glitches indicate electrical noise on the drive cable or an intermittent failure of the drive electronics.
This item is considered a Problem Indicator that may be correctable without replacing the drive. Verify the status

of the drive by checking the following:

Ensure that all system and storage system cables are intact and seated properly. You may need to replace

cables.

Check the physical proximity of the system to other electrical devices. Since electrical noise may cause a

glitch error, check the AC circuit for other electrical devices.

If you continue to receive many of these errors, replace the drive.

Failure Indicators

Use the Failure Indicators to determine the cause of a drive failure. Typically, the number of failures is zero when the

drive is operating normally. If a counter is not zero and the drive has not failed, there could be an intermittent

problem that may require the drive to be replaced.
The Failure Indicators are:

Spinup Errors—When the physical drive fails due to the failure of a spin-up command, a Spinup Error occurs. If

the failure count is not zero and the drive has failed, replace the drive.
If the counter is not zero and the drive is OK (has not failed), there may be an intermittent problem that requires

drive replacement. If you observe that the count is increasing over time, replace the drive.

Aborted Commands—The Aborted Commands counter records the number of times that a physical SCSI drive

returned an Aborted Command status when a SCSI command was attempted. This error count indicates

unsuccessful termination of the SCSI command. When the physical drive is failed due to aborted commands that

could not be retried successfully, Aborted Commands errors occur. If the number of errors is not zero and the

drive has failed, replace the drive.
If the counter is not zero and the drive is OK (has not failed), there may be an intermittent problem that requires

drive replacement. If you observe that the count is increasing over time, replace the drive.

Agent information 45


This manual is related to the following products: