In the SAN environment, the SCSI ID of a device may get changed due to a HBA failover or if the a hardware component in the SAN is reset. In such a situation, read/write requests to the media in the device may fail, resulting in job failures. However, if this option is enabled, the system will automatically scan the device and update the SCSI ID and the job will be re-tried in the established retry interval.
This option also includes the following sub-options:
-
SCSI/SAN device scan interval n hour(s)
This option specifies the time interval that must be used for successive retry operations for SCSI ID updates. For example, if a SCSI ID update is performed after a read/write failure and a second read/write failure occurs almost immediately or within the specified time interval, the system will not perform the SCSI ID update. The system will scan for SCSI ID updates only when a error occurs after the specified time interval. It is recommended that the time interval be set to 30 minutes or more, as running jobs could go into pending, when the SCSI ID is updated.
-
Mark drive inaccessible on error and perform status check according to device scan interval
When enabled, this option marks the drive controller offline when the SCSI ID for a device is changed. (This can be viewed from the Drive Properties - Drive Controller tab, where the Drive Accessible will be displayed as No.)
The device detection will be performed once every 24 hours. This property can be modified using the SCSI/SAN device scan interval n hours(s) option in the MediaAgent Properties. Note that if the value established in MediaAgent Properties is less than 24 hours, the system automatically defaults to 24 hours.
If you want to perform this operation more frequently create the nDisabledDriveDetectIntervalMin additional setting. Once created, this additional setting will override the setting established in the MediaAgent Properties.
The device detection will be repeated for 7 days. If these drives are not accessible even after 7 days, the drives will be marked as accessible. (If the drive continues to have a problem, the subsequent job accessing the drive will once again cause it to be marked as inaccessible.) If necessary the nDisabledDriveMaxDetectionRetryHours additional settings can be created to change the number of days the device detection will be repeated.
Procedure
-
From the CommCell Browser, expand Storage Resources | MediaAgent.
-
Right click the appropriate MediaAgent, and then click Properties.
-
Click the Control tab.
-
Select Auto Update SCSI ID during live operation to automatically scan the device and update the SCSI ID.
-
In the SCSI/SAN device scan interval specify number of hours after which successive SCSI updates can be run on the MediaAgent
-
Select Mark drive inaccessible on error and perform status check according to device scan interval to mark drive controlled offline when the SCSI ID for a device is changed.
-
Click OK to save the configuration.