The Storage Health Monitor examines FC-AL, disk shelf, and disk events and provides two kinds of syslog messages about the events: urgent and nonurgent. Each message has a corresponding message that is sent through autosupport and is available through SNMP. One set of message strings is provided for each event. There is also an informational message.
A syslog message about a predicted failure event is an urgent message and should be acted on immediately.
shm: disk has reported a predicted failure (PFA) event: disk XX, serial_number XXXX
DISK_FAILURE_PREDICTED!!!
The Autosupport priority is alert.
The disk has reported a Predicted Failure (PFA) Event. The disk's internal error processing and logging algorithm computation results are exceeding an internally set threshold. The disk is bound to fail in a matter of hours.
If you experience a predicted failure event, complete the following step.
| 1 | Replace the disk, as described in Chapter 5, Disk Management. |
A syslog message about an FC-AL link failure is an urgent message and should be acted on immediately.
shm: link failure detected, upstream from disk: id XX, serial_number XXXXX
FC-AL LINK_FAILURE
The Autosupport priority is err.
A Fibre Channel disk or cable, if disks are in different disk shelves, might be malfunctioning and might have caused an open loop condition. This results in a loss of sync over 100 milliseconds for a downstream disk that reported it as a link failure.
If you experience an FC-AL link failure, complete the following steps.
An "FC-AL unreliable" message is a nonurgent message and does not require immediate action. However, you should take the actions listed in the Action column of the following table. If you need assistance in performing the actions, contact Network Appliance Technical Support.
shm: possible link errors on disk: id XX, serial number XXXXX.
FC-AL UNRELIABLE
The Autosupport priority is warning.
One of a group of four Fibre Channel disks in a disk shelf or any connecting cable might be malfunctioning.
This results in a large number of invalid_CRC frames and data underruns on the loop. The invalid_CRC and underrun count has crossed its set threshold several times.
If you receive an "FC-AL unreliable message," complete the following steps.
A disk degraded message is a nonurgent message and does not require immediate action. However, you should take the action listed in the Action column of the following table. If you need assistance in performing the action, contact Network Appliance Technical Support.
shm: disk I/O Completion times too long: disk XX, serial number XXXXX.
DISK_DEGRADED-I/O
The Autosupport priority is warning.
Either the disk is old and slow, or it is internally recovering errors, thereby taking too long to complete an I/O. The message also indicates that there are too many I/O timeouts and retries on a disk. The disk might also be returning a Command Aborted status quite often. All these can result in a low data-throughput rate for that disk, and lowered overall system performance.
If you receive a disk degraded message, complete the following step.
| 1 | The disk is failure-prone; you should replace it as described in Chapter 5, Disk Management. |
A disk recovered errors message is a nonurgent message and does not require immediate action. However, you should take the action listed in the Action column of the following table. If you need assistance in performing the actions, contact Network Appliance Technical Support.
shm: disk returns excessive recovered errors: disk XX, serial number XXXXX.
DISK_RECOVERED_ERRORS
The Autosupport priority is warning.
Either the disk has found medium or hardware errors (unrecovered errors), or it has internally recovered numerous errors. The disk might also be returning a Command Aborted status. Quite often the errors returned exceed by many times the disk-vendor-specified bit error rate.
If you receive a disk degraded message, complete the following step.
| 1 | The disk is failure-prone; you should replace it as described in Chapter 5, Disk Management. |
A loop intermittent instability message is an informational message and does not require any action.
shm: intermittent instability on the loop that is attached to Fibre channel adapter: id XXX, name XXXXX. NOTE: no immediate action needed.
None
A Fibre Channel adapter, attached disk shelf, disk, cable, or connectors might cause instability on the FC-AL loop. It results in I/O completion rates that are below a set threshold.
Copyright © 1994-2003, Network Appliance, Inc. All rights
reserved.
Part No. 210-00472
File: appa4.htm
Last updated: 11/17/03 16:58:04