Going from (n) amount of smaller disks to less disks but larger size


Recommended Posts

Hi

So i'm planing to migrate my current unraid server to new hardware, and in the same run, i would like to go from my current array 6 (inc. parity) disks that are from 1-2tb in size each, to maybe just 4 disks but then do 3tb disks instead.

Is there a good way to do this? Size upgrade seems easy, but the i could see an issue in going to a lower number of disks, since data is split between all of them.

Link to comment

The way I would do it depends on a few factors.

 

Which license do you have?

 

What format are the current disks?

 

Do you have full backups at this moment?

 

Why are you considering such small disks? The minimum I would spend money on for new disks would be 8TB. If you have some 3TB verified good disks hanging around, fine, but I wouldn't buy them new.

Link to comment

I've got the Plus license

They are formatted as xfs.

I do have a full backup to cloud, though i would not wanna do a full re-down if possible, that would be rough :)

Really its a price/need thing, currently i got 6tb array, which i really good enough for a while, so just going 4x3tb would be 9tb (1 parity) which would be way more then enough for a long while for me :)
I could even just go 3x3tb just to save a bit initally since i dont really need the storrage increase just yet.

I really wanna stay on WD Reds, and they are about 121 usd pr 3tb drive (converted from DKK) here i DK.
So i could get away with an initial spend of 363usd contra 605 usd (2x8tb wd reds).

But i do get your point :)

Also maybe an important note, my current server only has support for 2.5" drives, where the new can do both (also one of the reasons for migrating to a new server). though atm the new one only got bays for 4 disks, but the plan is to upgrade that to 8 or 9, if/when needed.

Link to comment

Given what you've said, I would proceed somewhat like this.

1. Move all current data drives + new parity drive to new server.

2. Build parity with current data drives, do a correcting parity check. Any errors = stop and evaluate.

3. Replace current disk1 with new drive, rebuild, do a non-correcting parity check. Any errors = stop and evaluate.

4. Replace current disk2 with new drive, rebuild, do a non-correcting parity check. Any errors = stop and evaluate.

5. Use rsync or whatever copy method you like to COPY, not move, data from remaining data drives until disk1 and disk2 is comfortably full. Keep track of data copied, hopefully full drive at a time starting at highest number drive. So copy all of disk5, and as much of disk4, etc. If you evaluate your folder sizes and available disk space properly, you should be able to fit everything on the two new drives. You can verify the copies using whatever tool you feel comfortable with.

6. After all data is on the 2 new drives, set a new config and rebuild parity with only the new drives. All current drives should still contain your original data as backup, should something go sideways during this procedure.

 

I realize you said the new server only has 4 bays, but I assume you can temporarily rig something up to attach 6 drives for the migration.

 

Another option that would probably be slower would be to set up the new server with a trial key, and keep the old server intact and copy across the network.

  • Like 1
Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.