Silicon Graphics InfiniteStorage 4000 Series, Настройка и обслуживание

Страница: 210 / 276

198 Recovery Operations

The loop cycle count

The operating system (OS) error count

The port connections attempted count

The port connections held off count

The port utilization
The method for collecting error statistics starts by establishing a baseline for the SOC
error statistics. The baseline consists of SOC error statistics that are established at a
set time for each SOC device on the loop. The baseline is set by clearing the error
counters in each SOC device. You can set a device baseline by performing one of
these actions:

Turning on the power to the device or resetting the device

Running the
reset storageArray SOCBaseline
command
In addition, each controller also initializes the SOC error counters in all of the drive
trays that are attached to the controller following a cold boot (power-on or hot
insertion). If you add a drive tray while the power is turned on to the storage array, a
new baseline is established for any device on the drive tray.
After you have established the baseline for the SOC devices, you run the storage array
for a predetermined amount of time (for example, two hours). At the end of the run
time, you collect the SOC error statistics by saving the information to a file. To save
the information, run the
save storageArray SOCCounts file
filename
command. The default name of the file that contains the SOC error
statistics is
socStatistics.csv
. You can use any file name that has the
.csv
extension.
Analyzing the SOC error statistics is beyond the scope of normal storage array
management. After you have collected the SOC error statistics in a file, send the file
to your Technical Support representative.
Recovery
Operations Recovery operations include repairing the storage array and returning it to an
operational state. This might involve replacing a failed canister, a failed controller, or
a drive or restoring data or the storage array to operation. For information about when
it is appropriate to replace a canister, see “
Routine Maintenance
.”
Setting the
Controller
Operational Mode
A controller has three operational modes:

Online

Offline

Service
Placing a controller online sets it to the Optimal state and makes it active and
available for I/O operations. Placing a controller offline makes it unavailable for I/O
operations and moves its volume groups to the other controller if failover protection is
enabled.