Head swap in 2-node MCC

[aborzenkov]

Are there any concerns? My understanding is that it basically should be same as motherboard replacement which is supported. Of course it is necessary to adjust ports, licenses etc for new head, but that is not different to standard ARL head swap procedure.

I could not find any definitive answer whether this is "officially" supported or not.

TIA

P.S. was too fast. The obvious problem will be NVRAM mirroring. No chance to do it non-disruptively?

[akiendl]

Hi,

yes this is supported. This should be in the documentation here:

http://docs.netapp.com/platstor/topic/com.netapp.doc.hw-upgrade-controller/home.html?cp=8_4 

For finding out which guide to use.

here is the guide using ARL:

http://docs.netapp.com/platstor/topic/com.netapp.doc.hw-upgrade-controller/GUID-6F9CE102-FA28-4649-AB03-76B076C1BC75.html?cp=8_4_5_0

You will find MetroCluster specific steps in that guide like this one here as an example:

1. If the original nodes are in a two-node MetroCluster configuration, switch over storage and client access and heal the root aggregates:
   1. Check the MetroCluster status by using the metrocluster show command to determine whether automatic switchover has occurred.
   2. If automatic switchover has not occurred, perform a planned switchover operation from the healthy node by using the metrocluster switchover command.
   3. Resynchronize the data aggregates by using the metrocluster heal -phase aggregates command from the surviving cluster.
   4. Heal the root aggregates by using the metrocluster heal -phase root-aggregates command.
   5. Verify that the heal operation is complete by using the metrocluster operation show command on the destination cluster.

Andy

[aborzenkov]

---

@akiendl wrote:

yes this is supported. This should be in the documentation here:

http://docs.netapp.com/platstor/topic/com.netapp.doc.hw-upgrade-controller/home.html?cp=8_4 

---

Thanks for the link, but just wanted to make sure we are speaking about the same things.

After head swap new controller most likely has different NVRAM size so NVRAM mirroring is disabled. Actually for 4-node cluster with ARL head swap MC FC FAQ explicitly warns "NVRAM mirroring between the sites must be disabled for the entire process" which logically also applies to 2-node cluster. And it goes on to add "Planned switchover is not possible while the upgrade is in process".

But for non-disruptive 2-node head swap we must do planned switchover (and switchback). It is clear that initial switchover with old heads is possible. But the main question is - will switchback and subsequent switchover in opposite direction with one old and one new head be possible too? If yes, is it done by standard commands or some diagnostic mode magic is necessary?

Documentation rather sidesteps it unfortunately ...

[akiendl]

hi,

there should be no special settings/options needed. I know the procedure has been tested recently and updated so following the setup should work.

Andy