6

I read the question here Reusing Raid 5 Drive? but I'm not sure it applies 100% to my situation.

What is the standard practice when removing a drive from an array to use as a spare for another array? Is it absolutely necessary to wipe the drive and remove the array, or can I literally just take a drive from a decommissioned server and plug it into another one?

I'd like to break a RAID 1 array on a production server (so I have a backup drive), and if an upgrade goes well return the drive to the array to restore the mirroring function without introducing a brand new drive; however, I can logically see things wrong with doing it this way.

Thoughts?

The server is an HP DL380 G7 running RAID 1 on the OS drive (ESXi is the OS), and a RAID 5 array with 4 300 Gb drives for data.

Community
  • 1
hax0r_n_code
  • 420
  • 2
  • 5
  • 16
  • Can you specify the OS, HP server model and generation, type of disk and any other real details? – ewwhite Dec 04 '14 at 13:50

1 Answers1

5

You don't need to wipe the drive to do what you're describing. You'll find that you won't even have a convenient way to erase a single disk.

I'd urge you not to break the RAID 1 array as a form of backup, though. What happens if the upgrade "goes poorly"?

This practice shouldn't be promoted because it carries some risk, and doesn't scale to other RAID levels or larger groups of disks.

HP does have a sanctioned approach to do this type of operation. It's called Split/Recombine Mirror Array.

Edit:

You're doing this with VMware ESXi. It's unnecessary to go through this trouble for ESXi. When you upgrade VMware, there's an option to revert to an alternate Bootbank.

See: How is the /altbootbank/ partition used in ESXi?

Or the VMware ESXi architecture paper.

The ESXi system has two independent banks of memory, each of which stores a full system image, as a fail-safe for applying updates. When you upgrade the system, the new version is loaded into the inactive bank of memory, and the system is set to use the updated bank when it reboots. If any problem is detected during the boot process, the system automatically boots from the previously used bank of memory. You can also intervene manually at boot time to choose which image to use for that boot, so you can back out of an update if necessary.

Community
  • 1
ewwhite
  • 194,921
  • 91
  • 434
  • 799
  • Well if the upgrade goes poorly I would go back to the drive we removed from the mirror. I've actually done that quite a few times in the past but budget and spare hard drives weren't a concern. – hax0r_n_code Dec 04 '14 at 13:52
  • What other risk does it carry? I completely agree with you in it not scaling to other raid levels or larger groups of disks. And thank you so much for the HP article! – hax0r_n_code Dec 04 '14 at 13:58
  • Thanks for the ESXi link too, didn't know about this either! – hax0r_n_code Dec 04 '14 at 13:59
  • 1
    @inquisitor Pulling disks for fun isn't reasonable. Also, you're using VMware. You're better off just [saving the ESXi host settings](http://serverfault.com/a/550282/13325) or relying on the built-in ESXi failsafe. – ewwhite Dec 04 '14 at 14:00
  • I wouldn't say it's for fun, sir ;). But in a Windows environment it saves you from having to rebuild or repair an OS. I understand it's probably not the best method, but it's quick and "dirty." Thanks for all your help, you've educated me on a lot alternatives today. – hax0r_n_code Dec 04 '14 at 14:03