Storage Boards Cleanup
To make it easier to find information about HPE Storage products and solutions, we are doing spring cleaning. This includes consolidation of some older boards, and a simpler structure that more accurately reflects how people use HPE Storage.
HPE StoreVirtual Storage / LeftHand
cancel
Showing results for 
Search instead for 
Did you mean: 

P4500 Cluster Dissolve

Jurbul
Occasional Visitor

P4500 Cluster Dissolve

Hi Everyone,

 

I have a P4500G2 cluster that has a mirrored pair of storage nodes in a cluster. The cluster serves a single volume that consumes all of the available space.

 

I need to provision a second volume, but do not have the money to provision additional storage. The decision has been made to remove the network RAID from the clustered volume so that the second node can provide the additional storage. Both are configured as RAID6, so whilst there is no longer the network resilience, there will still be disk resilience.

 

How do I remove one of the storage nodes whilst retaining the integrity of my volume?

 

Many thanks in advance,

Stu

1 REPLY
oikjn
Honored Contributor

Re: P4500 Cluster Dissolve

do you actually use that amount of space or is it just provisioned to that size?  thin provisioning might be your answer, If you really are filling it all with 200% of your specified sforage, then you CAN just change the provisioning to network raid0 and you will gain the space you need while keeping hardware redundancy.  I would suggest that if the only thing on these systems are two equal volumes with network raid0, that you remove one system from the current cluster and create a new cluster for that so you put one volume on one cluster with one node and the other on another cluster with the other node.  The reason is that a two node raid0 cluster will be less reliable... when one node goes offline, you would lose access to both LUNs until both come back online.  If you split the LUNs into two clusters, then each LUN is independent and would only go offline when that specific node went offline.

 

I would say that unless you require shared storage for the LUNs, that you should absolutely NOT use NR0 for your entire system.  I know budgets sometimes trump all, but you should be raising hell that this is a bad idea and if they need that extra storage, they should just buy some cheap DAS to put the new data on over putting it on your SAN and screwing up the SAN reliability.