Storage upgrade strategy - external replication

[john.doe79]

I got following HDD config (in my TrueNAS Core 13u3).

• 2 x WD Red 3 TB as a mirror pool for "important" data & time machine called "mirror"
• 1 x WD (something) 8 TB as a single disk pool for media
• 1 x WD (something) 12 TB as a single disk pool for media

TrueNAS OS is running on a SATA SSD which is not affected

looking now into buying 3 x 18 TB Seagate Exos X X18 or Toshiba Enterprise Capacity MG09ACA 512e. (any recommendations?)

As I only have 1 SATA Port available as a spare, I would build in one of the 18 TBs, create a pool (which config?) and copy the volumes (is their an easy way to migrate them?) from the "mirror" pool to the new pool, then removing both WD Red 3 TB and adding the other two 18 TBs to the server, adding them to the pool with redundancy, that would give me a size of about 33 TB? Next would be migrating the volumes from the 8 and 12 TB to the new pool. right in that assumption?

how long will it take to extend the 18TB pool with the 2 x 18 TB disks (approximately) ?

Am I also right, that I need to fix the drive mapping in the jails?

At the end of the day I would leave in the 8 TB as a download/temp drive and take the 12 TB to a Synology single bay NAS in a remote location as a backup target for the critical data (not the media) from my TrueNAS (doing a local seed first, move it off site and setup a replication).

 

[NugentS]

Not sure how you are going to do that safely
I assume you intend to end up with a Z1 of 3*18 which means you need a minimum of 2 spare ports to run a degraded Z2
So
Insert Single 18TB, Create stripe pool. Copy Mirror pool to Stripe. Copy 8TB to stripe
Remove Mirror, Remove 8TB.
You know have an 18TB stripe and a 12Tb Stripe and 3 spare ports along with 2 spare 18TB disks

How do I make a degraded 2 drive RAIDZ?

Hello everyone. I have a workstation machine running Debian stretch with 24 GB of RAM + 8TB WD Red with an exact setup in another city (at my parents home) and they mirror each other with a nightly rsync script. I'm running out of storage (83% occupied). I've decided to give FreeNAS a shot and...

www.truenas.com

You can now create a Z1 array using the two spare 18TB disks and a fake disk
Remove the fake disk, thus degrading the array. Do NOT copy data to the Z1 when it includes the fake disk
Copy the 18TB stripe to the degraded array
Remove the 18TB stripe and add the disk to the degraded array and resilver. Wait

Copy 12TB Stripe to Z1 array
Remove 12TB

Result Z1 Array - healthy.

[Caveat] This process is hazardous to your data. If it goes wrong - you are screwed, completely, and not in a good way. Make sure you have thoroughly stress-tested any new disk before you attempt this. Any failure will be very very bad. This is an unsafe process. Do not do this if you value your data. If you can back up anything important first and treat everything else (data) as disposable.

Can you post hardware details, any spare PCIe slots as there may be a better suggestion depending on hardware

 

[john.doe79]

It is this mainboard. Just noticed, it has 8 SATA Ports, not 6. But only 6 on the Intel Chipset based controller, the other 2 on a additional.

Anyway. If I would copy the data from the 3 TB mirror drive (only 600 GB used) to the 12 and 8 TB disk, remove it from the system. Then I have 3 ports on the intel available for the 3x 18 TB to directly build it as a healthy Z1. Copy the data from the previous Mirror Volume to the 3x18z1, and then the volumes from the 8 and 12 TB to the new volume - so at the end all the more ore less valuable date is on the new z1 with the 3x 18 TB. I now can remove the 12 TB and make the 8 TB the Temp Volume.