Cannot expand volume after adding new drive

[DSfuchs]

When you have copied your 11TB data

Note: Do not copy; either back up or move!

Moving a share to an unprotected storage pool is very safe in DSM, since the commitment takes place only upon complete transfer.

However, with a "backup only" approach, programs, among other things, will be lost. Provision should also be made for a system outage lasting several days. Given the uncertain outcome, even double that duration should be considered tolerable.

 

[ragnarok]

Note: Do not copy; either back up or move! Moving shares to an unprotected storage pool is very safe in DSM.
However, with a "backup only" approach, programs, among other things, will be lost.

DSFuchs!
I would really appreciate, if you now stop chopping bits and pieces out of my responses in this thread and disrupt the context with your multiple single-line comments.

It's pretty annoying tbh...

 

[DSfuchs]

@ragnarok
From your perspective, what speaks against this smart approach?

 

[DSfuchs]

DSFuchs!
I would really appreciate, if you now stop chopping bits and pieces out of my responses in this thread and disrupt the context with your multiple single-line comments.

It's pretty annoying tbh...

No problem, I had already thought of that myself. That way, your subsequent changes won't get lost in the context, either.

 

[ragnarok]

No problem, I had already thought of that myself. That way, your subsequent changes won't get lost in the context, either.

Thank you very much. Highly appreciated.

 

[DSfuchs]

Thank you very much. Highly appreciated.

Yes, the partial allowed "Reply-to" headers do indeed cause confusion. I wish they were disabled, especially for newcomers who don't give it much thought.

 

[Telos]

Googling it now confirms the limitation of the DS414.

As does the Synology datasheet for that model.

The more drives there are in a storage pool, the higher the probability of a complete failure, too.

So, you recommend single drive storage pools? How about putting some numbers to that statement in the context of RAID 5 in a 4 drive NAS, and the use of "complete" failure. Do you also assume no use of backups? Never have I heard that a 4 drive RAID 5 is inherently at great risk.

 

[Coop777]

As does the Synology datasheet for that model.

So, you recommend single drive storage pools? How about putting some numbers to that statement in the context of RAID 5 in a 4 drive NAS, and the use of "complete" failure. Do you also assume no use of backups? Never have I heard that a 4 drive RAID 5 is inherently at great risk.

Agree. Single drive storage poos defeat the entire purpose of NAS storage.

 

[MrClicky]

Yes. And you would probably pick a BTRFS volume nowadays, and not EXT4 as you had on your DS414.

I suppose an alternate method would just be to blow it all away, create a new Storage Pool and BTRFS volume from all 5 drives then copy all the data from my backup (yes I have a back up)

But your procedure below sounds excellent, thanks so much for taking the time to write it up for me, much appreciated @ragnarok !

What a great forum!

An additional note on the procedure above:

As your old disks are all smaller than 16TB, the order of copying, raid array creation and volume expansion becomes significant.

When you have copied your 11TB data safely aside (make sure you have a backup too!) to the intermediate volume (16TB disk), you would want to

• Delete your original volume and storage pool
• Recreate your storage pool using disk 1-4 (and pick BTRFS, if you decided for that)
• Recreate your volume from scratch using the newly created storage pool
• Copy data from the intermediate volume (16TB disk) back to the new volume
• Remove the intermediate volume
• Remove the 16TB disk from its storage pool
• Add the 16TB to the storage pool containing disk 1-4 and then,
• Expand the new volume with the 16TB disk

 

[MrClicky]

Agree. Single drive storage poos defeat the entire purpose of NAS storage.

I don't fancy creating a new storage poo but thanks anyway

 

[MrClicky]

Thanks to everyone for their help, seems like this thread really kicked off after I put myself into storage for the night

Appreciate all the answers and civility @ragnarok and @DSfuchs - a great way to join finding such energetic members willing to help. cheers!

 

[BangFR]

I thought that using SHR, you should use same size of disk in order to optimize the storage you get if all of them are attached to the same volume.

There with 2TB and 6TB disk, I wonder how your volume will react if you got a dead disk

Is it right or there is a slight detail (or a big one) I didn't catch ?

 

[DSfuchs]

I thought that using SHR, you should use same size of disk in order to optimize the storage you get if all of them are attached to the same volume.

There with 2TB and 6TB disk, I wonder how your volume will react if you got a dead disk

Is it right or there is a slight detail (or a big one) I didn't catch ?

The real problem is that the proprietary SHR scatters partitions across the hard drives haphazardly. We recently had a drive reported here as defective that was still hosting a partition lacking redundancy. It was absolute chaos and inconceivable to recover on non-Synology hardware.
Therefore, I wouldn't want to take on the additional risk of a presumably old, slowing 2TB hard drive in a 30TB installation.

Use the same size disk in order to optimize storage; this applies to all scenarios.

 

[MrClicky]

With SHR you can use mixed sizes and still achieve maximum space, it just depends on the combination. I'm aiming for a mix of 14 & 16 to achieve no unused space. If all were the same size I'd probably just use RAID 5.

Getting the same size is a different matter at the moment. HDs are both very expensive and scare with AI gobbling them all up for the foreseeable future. Two years ago I bought two 14TB for the same price as one now if you can get it! I was going to fill it with 14s but I recently had to buy a 16TB as 14s weren't available in WD or Seagate at retail. As an independent maker I need to use what I have.

Having said that I've now decommissioned the 2TB and stuck it in the old 414 which I'm using as a backup for production elements on the NAS. Less essential archive material goes on the older drives with each drive as its own volume as a third tier back up.

Thanks for your help gang.