NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Solved
Messed up trying to upgrade drives, help!
Hi, I could use some help upgrading the drives in my ReadyNAS RN212. I had 2 Seagate 6TB drives in RAID1, and wanted to upgrade to 2 12TB drives. I don't think I followed the correct procedure in doi...
Your procedure was the correct one.
I'd start by testing the two new drives in a Windows PC using Seatools. I always start with the non-destructive full test, and follow up with the full write-zeros test. That will take awhile (perhaps one day per drive) - but it will give some assurance that both disks are ok. I have purchased some drives that passed one of these tests out of the box, but failed the other. You can connect the drives with directly with SATA (assuming a desktop PC) or get a USB-SATA dock.
Assuming the drives pass, then boot up the NAS with only the 6 TB drive installed in slot 2. Then hot-insert one of the (now blank) 12 TB drives in slot 1. That should resync.
Once you are sure the resync has completed, you can then try hot-swapping drive 2 again.
However, when powered back up, and accessed the web dashboard, I saw that the drives were now re-syncing, which I did not want since 12TB #2 had already synced.
Likely disk 1 was being resynced to disk 2. Probably not an error, and interrupting that sync explains why disk 1 is now red.
One thing you should do for extra data safety is to connect the 6 TB drive from slot 1 to a PC and format it there. Then copy your data to that drive over the network. You have no redundancy at the moment, so any problems with drive 2 will result in data loss.
Hi Stephen,
Thank you so much for your quick reply! I am following your suggestion of fully testing the 12TB drives, and will also backup the working 6TB drive.
Had a few followup questions if you can spare the time:
Is there a concept of Primary and Secondary with the drive bays, or any functional difference between them? You mentioned booting the NAS with 6TB drive in slot 2, blank drive in slot 1. Does the placement of drives when swapping & syncing matter?
And related: how does the NAS know which direction to sync? For example, as you explained that the unexpected re-sync I observed was likely disk 1 syncing to disk 2, when I had expected the opposite? Is there any way to control this behavior?
Lastly a general RAID question: once a drive has been setup in RAID1, is it possible to remove that drive and mount it elsewhere, ie to copy the data off? Or does this require disabling/removing the RAID array first in ReadyNAS before removing the drive?
Again, thanks again for all your help, I have seen your posts elsewhere in the forum and am very grateful for all your contributions!
Scott
Is there a concept of Primary and Secondary with the drive bays, or any functional difference between them? You mentioned booting the NAS with 6TB drive in slot 2, blank drive in slot 1. Does the placement of drives when swapping & syncing matter?
With RAID-1 the two drives are mirrored, so there is no difference in their contents. The placement normally doesn't matter, as the RAID software can determine the slot order even when you are using a more complicated RAID mode.
I generally suggest preserving the order, as this can make some troubleshooting a bit easier to do.
And related: how does the NAS know which direction to sync? For example, as you explained that the unexpected re-sync I observed was likely disk 1 syncing to disk 2, when I had expected the opposite? Is there any way to control this behavior?
There are couple of ways.
- If one drive is blank/unformatted, then of course that is the destination of the sync.
- If a sync is interrupted, it is supposed to resume when the system is rebooted. Clearly that didn't work in your case. That could be due to the amount of disk shuffling you did while troubleshooting. But a disk failure of one of the new drives could also explain it.
- There is a transaction counter saved on each drive partition in the RAID group. When they aren't the same, the drive with the fewest transactions is synced. The idea here is that some writes to the disk were lost.
In your case, something was likely written to the array during the first sync, That would have updated the counter on the second disk, so it would be ahead of the first one. That can happen even if you didn't update any files, as the BTRFS file system can do some background maintenance.
There's not much you can do to control this behavior. If you are just doing off-line testing, or migrating disks to another NAS, then powering down the NAS before removal, and leaving it powered down until the disks are reinserted will prevent resyncs. But that won't help when upgrading.
Lastly a general RAID question: once a drive has been setup in RAID1, is it possible to remove that drive and mount it elsewhere, ie to copy the data off?
Yes. In your specific case, the working 6 TB drive (disk 2) will still be mountable after the full upgrade to 2x12 TB. You could put it back in the NAS by itself, and boot it up. Or mount it in another linux system. (Windows and MacOS don't support the file system used by the NAS).
Normally that would also be the case with the other 6 TB drive. But the interrupted sync can get in the way of that.
FWIW, once done, the system would also boot with just one of the 12 TB drives. The volume status will be degraded (since there is no mirror).
Some people have tried to use this behavior as a backup mechanism. But that is a really bad idea - the sync process requires a lot of disk I/O, so you don't want resync routinely. That said, it would be a good idea to keep disk 2 around (and intact) for a while after you upgrade, in case something goes wrong with the expanded volume. Whether you do that or not, you should have a backup plan in place for your critical data. Although RAID is useful, it does not guarantee data safety.
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
