DSM 7.1 Replacing drives to gain unavailable capacity

[Puddleglum]

Sorry to ask for advice on a fairly basic question...

I have a DS918+ running DSM 7.1.1, which had 2 x 8 TB Seagate Ironwolf Pro and 2 x 10 TB Seagate Ironwolf HDDs in one SHR storage pool (using btrfs) with 2-drive fault tolerance.

Both 10TB drives unfortunately failed a short time apart and I replaced them both with 16 TB Pros so I could keep the system running with fault tolerance. As expected, the storage pool stayed at the original size of 14.5 TB.

I'm now running out of space (12.5 TB in use) and wondered what is the best means of gaining the 'lost' capacity of the 16 TB drive(s).

Based on the info on the Synology website about that situation, I'll need to replace one or both 8 TBs with >=16 TBs.

If I do so, will I gain the full capacity of both the old as well as the new 16TB drives?

As an alternative, I've thought of getting a DS725+, putting the new 16TB drives in it and then using Hyper Backup to do a full system backup of the DS918+ (after upgrading it to DSM 7.2 or 7.3). I'd then nuke the DS918+, reinstall DSM, set up the 4 disk array and then restore everything from the backup.

However, would this just recreate the storage pool the way that it is at present with 14.5 TB of space?

I could do a files only backup and restore, but this would mean having to recreate all the settings on the system manually, including those of a dozen docker containers.

I'd appreciate any guidance on what the best way forward would be - thanks!

This post includes affiliate links. As an Amazon Associate, SynoForum.com may earn a commission if you make a purchase — at no extra cost to you.
It helps support our community! Learn more...

 

[Telos]

running DSM 7.1.1

Update DSM (v7.3.x)

I have a DS918+ ... with 2-drive fault tolerance.

Unnecessary for pools < 8 drives. Revert to SHR.

which had 2 x 8 TB Seagate Ironwolf Pro and 2 x 10 TB Seagate Ironwolf HDDs

In your case, the 10 TB drives are acting as 8 TB drives. More wasted space.

10TB drives unfortunately failed a short time apart and I replaced them both with 16 TB Pros

Now you have 16 TB drives operating as 8 TB drives. Just why please?

I'm now running out of space (12.5 TB in use)

You've made quite suboptimal choices in your drive deployment and RAID.

I'll need to replace one or both 8 TBs with >=16 TBs.

A sure improvement. But SHR is the correct answer here (not SHR2).

 

[Puddleglum]

Update DSM (v7.3.x)

Unnecessary for pools < 8 drives. Revert to SHR.

In your case, the 10 TB drives are acting as 8 TB drives. More wasted space.

Now you have 16 TB drives operating as 8 TB drives. Just why please?

You've made quite suboptimal choices in your drive deployment and RAID.

A sure improvement. But SHR is the correct answer here (not SHR2).

Sorry about the delayed reply.

Yes, I made the mistake of switching from SHR -> SHR2 when I added the two 10TB drives at a later date.

I was thinking in RAID terms, i.e. RAID1 with 2 drives, then when 2 more are added -> RAID 10.

As it was, I did nearly lose data. After the first 10TB drive died, the 2nd 10TB drive crashed shortly after the rebuild.

I used a 16TB drive for the rebuild because that was what I had, not thinking that I'd be limiting the capacity. I mistakenly thought that the system would adjust the storage space as it would have with SHR.

So, if I replace the 8TBs with 16TBs in the SHR2 setup, given that drives 3 & 4 are also 16TB, would this give me approximately 30TB in total or around 24TB?

It sounds like the best option would be to hyperbackup everything to a hyperbackup vault on a second NAS, remove the storage pool on NAS 1, recreate the storage pool setting to SHR and then do a restore.

Would I need both NAS units to be on the same version of DSM to do this?

 

[Telos]

Would I need both NAS units to be on the same version of DSM to do this?

No. It's more dependent that the HB/HB Vault versions are compatible.