How to replace 2HDD Volume (HBR duplicated) in a 4 bay nas(920)
I got suggested (Synology AI) to make a full hyper backup, but this requires a C@ subscription or a second NAS…
Starting the docker containers causes a file checksum mismatch on a docker container file, and Volume1 becoming readonly.
Preventing to start of the docker containers, prevents triggering the file mismatch, and so keeping a rw RW Volume1
During backup I have some multiple other file checksum mismatches on file which did not trigger a Volume1 becoming readonly (and I have some older backups)
Volume1 is hybrid raid with 1 drive fault tolerance btrfs 5.2 TB 3.3 TB free. And is scrubbed.
So 2 slots in my 4 bay system
3rd slot is empty, and 4th slot is for video surveillance recording ext4-fs.
Surveillance package stopped for now.
Both Volumes and all 3 drives reports Healthy.
Volume1 has 2 WD60EFRX-68L0BN1 6TB
Having 2 new HDDs:
Seagate IronWolf ST6000VN006 6TB
How to replace the Volume1 HDDs?
What is the best way?
1)Take them both out and replace by new? But what about user shares etc….
Can I put back the old volume! as disc3&4 and copy the data to the new Volume?
2)Turn off, Replace one, turn on, rebuild, turn off, replace the other, turn on, rebuild
3)Add a hot spare in the free slot 3, …
replace first and rebuild?
make second (old) hot spare?
4)
…???
All Replies
Viewing 1 replies (of 1 total)
Viewing 1 replies (of 1 total)
- You must be logged in to reply to this topic.

The safest way to replace the two drives in your Synology DS920+ is to change them one at a time, not both together. Stop Docker first so the checksum issue doesn’t force the volume into read-only mode. Deactivate one of the old WD drives in Storage Manager, replace it with a new Seagate IronWolf, and run a repair on the storage pool. Once the rebuild finishes and the pool shows healthy, repeat the same steps for the second drive. This keeps your SHR-1 redundancy in place the whole time and preserves your existing shares, users, and settings. Do not remove both drives at once, since SHR with single-drive fault tolerance cannot handle two disks being removed simultaneously.