HPE StoreVirtual Storage / LeftHand
cancel
Showing results for 
Search instead for 
Did you mean: 

Upgrade 9.0 to 9.5

ArjanA
Advisor

Upgrade 9.0 to 9.5

Hi,

 

I want to upgrade two Lefthand Management Groups (one with 2 and one with 4 storage systems) from 9.0 to 9.5. 

I will be performing the upgrade when there are no users active, so there will be minimal disk i/o.

Do I have to bring down all the attached servers?

Or is the upgrade a rolling upgrade in which the volumes stay accessible?

 

Kind regards,


Arjan

4 REPLIES

Re: Upgrade 9.0 to 9.5

Hey Arjan,

 

If your setup was a Fail Over Manager (FOM) you don't have to bring your servers down. The firmware upgrade process has a lot off checks built in so it a solid and straight forward process. Is there async replication between the 2 groups? In that case you should first upgrade the group that receives the replication and afterwards the group that sends the traffic. But this check will also be done by the firmware upgrade.

 

Kind regards,

 

Joris Vliegen

ArjanA
Advisor

Re: Upgrade 9.0 to 9.5

To clarify things:

The two management groups are independent, they are on seperate sites.

 

One group consists of two P4300G2 nodes.

The other group is four P4500G2 nodes.

 

I run a virtual fail-over manager on a node in the group, not a FOM on a server.

If I install FOM on a server, will that make it possible to upgrade without downtime?

 

Kind regards,

Arjan

 

cheazell
Advisor

Re: Upgrade 9.0 to 9.5

My experience of upgrading to 9.5 has not been positive. I have twice attempted to upgrade from 9.0 to 9.5. I'm running 2 node VSA + FOM. In the last attempt I got 29% through - then it "Failed" and brought down the SAN. Not pleased. I was able to recover but I'm certainly not terribly confident in the process as a result.

Jamester
Occasional Advisor

Re: Upgrade 9.0 to 9.5

I upgrade a 12 node mangement group without problems (kind of took 36hrs with all reboots and resyncing). users noticed a lilttle slowness at times.

But it was sucessfull.