Background
- The DS216j was about 7 years old and decent specs had started to appear, so I bought a DS223 and migrated to it
- Neither the manual nor the website got me to the answer quickly, so I’m leaving a log of the trial and error
Outcome
- The migration itself seems to be just a matter of inserting one of the DS216j’s HDDs into slot 1 of the DS223.
- Once the DS223 boots and the migration completes, the storage pool ends up in a degraded state (if you had redundancy), so you need to power off once, insert the other HDD into the remaining slot, and repair the storage pool
Work Log
First, a quick description of the setup. The DS216j had a 3TB and a 4TB HDD in it. Each had been in use for 3 years.
Trying the DS216j’s Hard Disk in the DS223
At first I inserted the DS216j’s 4TB HDD into the DS223. But all I got was the initial setup screen, with no obvious way to go further.
The 3TB and 4TB drives are redundant with each other (SHR, similar to RAID1), so losing one would be acceptable, but they hold important things like photos, so that felt a bit risky.
Reluctantly, I decided to buy a new 4TB HDD and do a fresh initial setup of the DS223 (the plan being to expand to 4TB & 4TB and keep the 3TB as a physical backup).
As it turned out, at that point I had inserted one of the DS216j’s HDDs into slot 2 only, which is apparently why it was not recognized.
Setting Up the DS223
I inserted the new HDD and ran setup. For a new drive, btrfs (b-tree fs) was recommended. My reaction was: huh, not ext4. Setup completed as usual. Of course, being a new HDD, the data was empty.
I then thought: if I insert the DS216j’s 4TB HDD into the DS223 after setup is complete, maybe it can be copied physically? So I put the DS216j HDD into slot 1. At that point, the DS223 asked at boot whether I wanted it to migrate based on the DS216j’s information.
This is where I realized I had put the new HDD in slot 2.
I set up the DS223 again based on the information from the 4TB HDD that had been used in the DS216j, now in slot 1. The migration finished and it booted successfully with the DS216j’s information intact.
However, the storage pool ended up in a degraded state. The reason is obvious: the new HDD in the DS223’s slot 2 had already been set up separately, so there were two storage pools (and on top of that the DS216j one was ext4 while the other was btrfs).
From Storage Manager, I first deleted the storage pool of the new HDD (the one initially set up on the DS223 — safe, since the data was empty), and after that repaired the degraded state of the storage pool on the 4TB HDD that had been used in the DS216j.