storage disk firmware revert

Revert disk firmware

Availability: This command is available to cluster administrators at the advanced privilege level.

Description

The storage disk firmware revert command reverts firmware on all disks or a specified list of disks on a node.

You can specify a list of one or more disks whose firmware is to be reverted by using the -disk parameter.

You can revert the firmware on all the disks owned by a node by using the -node parameter.

This command can make the disks inaccessible for up to five minutes after the start of its execution. Therefore, the network sessions that use the concerned node must be terminated before running the storage disk firmware revert command. This is particularly true for CIFS sessions that might be terminated when this command is executed.

If you need to view the current firmware versions, use the storage disk show -fields firmware-revision command. The following example displays partial output from the storage disk show -fields firmware-revision command, where the firmware version for the disks is NA02:

cluster1::> storage disk show -fields firmware-revision
		disk     firmware-revision
		-------- -----------------
		1.0.0    NA02
		1.0.1    NA02
		1.0.2    NA02
		1.0.3    NA02
		1.0.4    NA02
		1.0.5    NA02

The firmware files are stored in the /mroot/etc/disk_fw directory on the node. The firmware file name is in the form of "product-ID.revision.LOD". For example, if the firmware file is for Seagate disks with product ID X225_ST336704FC and the firmware version is NA01, the file name is X225_ST336704FC.NA01.LOD. If the node in this example contains disks with firmware version NA02, the /mroot/etc/disk_fw/X225_ST336704FC.NA01.LOD file is downloaded to every disk when you execute this command.

How to Revert the Firmware for an HA Pair in a Cluster

Use the following procedure to perform a revert on the disks in an HA environment:

Make sure that the nodes are not in takeover or giveback mode.
Download the latest firmware on both nodes by using the storage firmware download command.
Revert the disk firmware on Node A's disks by entering the storage disk firmware revert -node node-A command.
Wait until the storage disk firmware revert command completes on Node A, and then revert the firmware on Node B's disks by entering the storage disk firmware revert -node node-B command.

Parameters

{ -disk <disk path name>, ... - Disk Name: Specifies the disk or disks whose firmware is to be reverted.
| -node {<nodename>|local}} - Node Name: Specifies the node name. The disk firmware will be reverted on all the disks owned by the node specified by this parameter.

Examples

cluster1::> storage disk show -fields firmware-revision
		disk     firmware-revision
		-------- -----------------
		1.0.0    NA02
		1.0.1    NA02
		1.0.2    NA02
		1.0.3    NA02
		1.0.4    NA02
		1.0.5    NA02

The following example reverts the firmware on all disks owned by cluster-node-01:

cluster1::*> storage disk firmware revert -node cluster-node-01

Warning: Disk firmware reverts can be disruptive to the system. Reverts involve
        power cycling all of the affected disks, as well as suspending disk
        I/O to the disks being reverted. This delay can cause client
        disruption. Takeover/giveback operations on a high-availability (HA)
        group will be delayed until the firmware revert process is complete.
        Disk firmware reverts should only be done one node at a time. Disk
        firmware reverts can only be performed when the HA group is healthy;
        they cannot be performed if the group is in takeover mode.

Do you want to continue with disk firmware reverts? {y|n}: y

Info: Reverting disk firmware for disks on cluster-node-01.

The following example reverts the firmware on disk 1.5.0 which is owned by node cluster-node-04:

cluster1::*> storage disk firmware revert -disk 1.5.0

Warning: Disk firmware reverts can be disruptive to the system. Reverts involve
        power cycling all of the affected disks, as well as suspending disk
        I/O to the disks being reverted. This delay can cause client
        disruption. Takeover/giveback operations on a high-availability (HA)
        group will be delayed until the firmware revert process is complete.
        Disk firmware reverts should only be done one node at a time. Disk
        firmware reverts can only be performed when the HA group is healthy;
        they cannot be performed if the group is in takeover mode.

Do you want to continue with disk firmware reverts? {y|n}: y

Info: Reverting disk firmware for disks on cluster-node-04.