Appendix A

Nondisruptive Hardware Upgrade in a Cluster

About this appendix

This appendix describes the process of changing hardware components by using the nondisruptive cluster upgrade method.

In a nondisruptive upgrade, each node is successively "failed;" that is, partners take over for each other at specific points of the procedure. While the "failed" node is in the waiting for giveback loop, it can be interrupted and brought to the firmware ok prompt or powered off. For more information about nondisruptive upgrades of clustered gFilers, see Planning for upgrading clusters.

When hardware updates can be performed

Hardware maintenance and upgrades that are supported by the nondisruptive process can be performed in conjunction with upgrading Data ONTAP on clustered gFilers if the upgrade is within a release family (from 6.5 and later).

If the upgrade is from a release prior to 6.5, you can upgrade Data ONTAP to 6.5.1 first using the standard upgrade method, thereby causing a small disruption in service. Then you can use the nondisruptive upgrade method for hardware upgrades; this method results in a longer period of disruption than if you were using the standard method.

Note
In these circumstances, the new Data ONTAP software must be downloaded first.

Supported hardware upgrades

When a cluster node is in "failed" mode, you can perform the following hardware upgrades:

Replacing the head (with the same type as before, and with the same adapters)
Replacing the motherboard
Replacing or adding an NVRAM or NIC, either with the same component or with an improved component (for example, you can upgrade from 2-port to 4-port gigabit Ethernet (GbE), or 1-port to 2-port Fibre Channel).
Replacing the Cluster Interconnect card
Installing on-board firmware on various platforms (for example, the BMC on the GF9xx models)

You can also run diagnostics on various components of the "failed" machine (for example, motherboard or NVRAM.)

Note
Running diagnostics on parts of the machine that are still visible to or in use by the partner--notably disks and the Cluster Interconnect card--can present problems. See the diagnostics documentation.

After any of the listed upgrades are complete, the node can be rebooted into the waiting for giveback state. During the rest of the nondisruptive upgrade procedure, the partner node does a cf giveback and the node boots on its new hardware and firmware.

Performing a nondisruptive hardware upgrade

If you are performing a nondisruptive upgrade for hardware separately from a Data ONTAP or system firmware upgrade, complete the following steps at the console of one of the gFilers in the cluster.

Step	Action
1	If...	Then...
	CIFS is not in use on gFiler A	Go to Step 2.
	CIFS is in use on gFiler A	Enter the following command: cifs terminate -t nn nn is a notification period appropriate for your clients. After that period of time, proceed to Step 2.
2	At the console of gFiler B, enter the following command: cf takeover Result This command causes gFiler A to shut down gracefully and leaves gFiler B in takeover mode.
3	Turn off the power on gFiler A. After you complete the upgrade of hardware on gFiler A, turn on the power on gFiler A, which causes it to boot. When the message `Waiting for giveback` appears, go to Step 4.
4	At the console of gFiler B, enter the following command: cf giveback Result This command causes gFiler A to reboot and resume normal operation as a cluster partner.
5	Repeat Step 1 through Step 4 to update the partner gFiler; that is, to shut down and update gFiler B with partner gFiler A in takeover mode.