Skip to content

Instantly share code, notes, and snippets.

@recoder

recoder/Synology.md

Created June 6, 2019 19:07
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save recoder/9410587f4caf9895c5a41537d28b8eec to your computer and use it in GitHub Desktop.
Save recoder/9410587f4caf9895c5a41537d28b8eec to your computer and use it in GitHub Desktop.
Download ZIP
Raw
Synology.md

Replacing a failed disk

If your NAS supports hot swapping (check Synology's website for that!), you could just pull the drive cage with the falling drive out, replace the drive in the cage and put the cage with the new drive back it.

However unless you actually need continuous availability I would just shut the NAS down, swap out the drive and power it back up again.

Either way in DSM your volume/storage pool will be shown as degraded, the unit might keep beeping (somewhere in the control panel there's a button to shut that off) and you will need to add the new drive to your storage pool (in the Storage Manager app).

Then the system will resync your RAID array with the new drive, which could take several days.

Please allow some additional remarks:

  • You didn't mention whether you run SHR-1 (one-disk redundancy) or SHR-2 (two-disk redundancy). Resync is actually a process that is very stressful on the drives. There is a chance that the resync of the array might actually cause one of the other three disks to fail. With one-disk redundancy that would cause the whole array to crash and all data to be lost, with two-disk redundancy that would mean that the array would stay degraded, even after the resync and you'd have to replace the newly failed disk and go through another resync.

A way to reduce this chance a bit, would be to put the "brakes" on the resync process. You can do this by going to Storage Manager -> Configuration and in the section "RAID Resync Speed Limits" select the option "Lower impact on overall system performance". This will slow down the resync, causing it to run even longer, but will hopefully ease the wear and tear on the drives.

Please note that a lot of times, folks are actually eager to complete their resync as fast as possible and will select the option "Run RAID resync faster" here. But let's hope for them that they have SHR-2/RAID 6 with two-disk redundancy.

BTW I don't know, if you need to set this option BEFORE you start the resync or if changing this setting will also affect an ongoing resync, so I suggest that you check (and if necessary adjust) this setting beforehand.

  • And to be really on the safe side, I would actually conduct a burn-in test for the new drive BEFORE pulling the old drive from the NAS. Use a powered SATA to USB adapter, connect the new drive to your PC and use a software that will write and read the entire disk multiple times, say for 48 hours or better 96 hours. If you got a bad disk, chances are that it will fail this test and you can return it to the vendor for a refund or replacement. If it survives, it has a reasonable chance of lasting you a while.

  • During the resync your drives will probably get warmer than usual, so crank up your system fan in Control Panel -> Hardware and Power. I would use at least the "Cool mode" setting, "Full-speed mode" is probably not necessary. But you should monitor the temperature of your CPU and your drives ever so often (Control Panel -> Info Center).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment