All luns within the same RAID Group use the same set of drives so you need to take care that one lun doesn't affect other systems...

Backups typically occur after hours so your impact to the other Luns should hopefully be limited (unless your doing after-hours processing.) Spreading your luns across the two RG's would be preferred method as you would end up with more spindles to absorb IO for production and Backup activities.

If you are doing after-hours processing on those Luns then you may be better off isolating your backup and production RG's from one another by adding more drives (500GB SATA?) and creating a new RG just for backup activity or using the Lun Migration option Aran suggested.

If you are able to create 2 Luns of the same size within each RG, you can create yourself a new Striped Meta and life should be good.

If the two Luns are of different sizes you will have to create a Concantenated Meta. Since you are performing backup (sequential writes) you may be ok, but your performance will be impacted a bit since you will be using less spindles concurrently.

If the two RAID groups that contain the LUNs you want to reallocate also contain LUNs that are in use then unbinding the LUNs will not help much. There are a couple of different approaches you can take to achieve your goals, and I will refer you to the Naivpshere Manager documentation for the detailed steps.

The document that will describe these steps in more detail is the Navisphere Manager Administrators Guide at:
Home > Support > Technical Documentation and Advisories > Software ~ J-O ~ Documentation > Navisphere Management Suite > Maintenance/Administration

1. Create a MetaLUN with all the available LUNs. This will be the quickest option and will provide you with a large LUN for backups, but it is probably not the most ideal solution.

2. Rearrange your RAID groups. You can use the "Migrate" feature in Navisphere to move your production LUNs to one RAID group. Unbind the unused LUNs first to free up space. And then create one large LUN in the second RAID Group for your backups. This would be more ideal as you are separating out your production from backup LUNs.

Unbinding a LUN does not affect other LUNs in the same RAID Group, so don't worry about that. You can always view in Navisphere the LUNs under each RAID Group and the LUNs in each Storage Group to determine which LUNs are in use in which RAID Groups.

Hope this helps,
Aran

Thanks for your help on this, the comments have been useful and hopefully I will be able to sort out the configuration. Thanks again.

nord,

Have you put the MetaLUN into a Storage Group in Navisphere? That's where you join the disks (LUNS) with the hosts. Just making a MetaLUN on a Raid Group isn't enough, you must put them together and Storage Groups are where that happens. Good Luck.

Harold

If you are still having problems, here are some high-level things you can look at within Navisphere Manager...

In Navisphere, you can right click the array and check "Connectivity Status"...

You should be able to see each of your host connections:
The HBA icon on the left should be green
Logged In should be "Yes"
Registered should be "Yes"

If this isn't the case, then we need to look at troubleshooting host connectivity issues: HBA Settings, Firmware, Navisphere Agent, Zoning, etc.

If these are all correct, then we need to look at Navisphere configuration issues:

As Aran stated, you should have created a Storage Group:
Right Click the Storage Group Folder and select "Create Storage Group"

Assign Host:
Once the Storage Group is created, right click the storage group you created and select "Connect Hosts" - Add the host from the list of available hosts (if not already added)

Assign LUN(s):
Select the LUNs Tab (right click the storage group and select "Select LUNs" and add the appropriate lun(s) from the list.

You can then rescan the disks from Disk Management and you 'should' be good to go.

If all of these steps are already completed, you may need to reboot your host (I've had a few instances where a reboot has resolved this kind of issue.)

Aran,

Thanks for your help, I have created a MetaLUN and it was simple to do, however the host is not able to see the disk.

When adding LUNs to a host previously I have simply gone onto the host server, done a recan in disk management and up pops a new drive, however this has failed to work in this instance.

The MetaLUN shows against the right host in navishphere, am I missing something else? Do I need to add the new MetaLUN to a RAID group?

I have tried the admin manual but it does not offer much help.

Thanks

Nord,
When you have tried the advice from Harold and Bowling, let us know if that helped. They covered the basics on what to look for if the LUN does not show up on the host.

At a lower level, after you have checked what they suggested, you could also look in PowerPath and you will be able to see the LUN listed if it is visible to the HBAs.

Aran

Thank you all for the help. The MetaLUN was already part of the correct storage group, however due to the mention of Power Path I checked it as it was indicating an error. In PowerPath it showed as several disks not being availalble, I removed these from the config, on doing so the MetaLUN then appeared in the list and subsequently appeared as a disk in windows. So all sorted!

So previously used LUNS that had been removed from the host were still showing in the PowerPath monitor and for whatever reason this prevented new LUNs from appearing.

Thank you all for the help.

Whenever you remove LUNs from a host you should clear the powerpath configuration of those paths before doing anything else. Depending on your host OS the commands will be different, refer to the notes for the appropriate version of powerpath.