Hello.
I’m running a node on my PC with 1 Tb disk dedicate to Storj storage. As I’m running Debian, I want to increase space disk by adding a 500 Gb disk and I want to use LVM.
So I have add my 500 Gb disk to a Physical Volume then create Volume Group and then a Logical Volume.
I stop and delete my node, copy all files from 1 Tb disk to disk LV, change node start command and start the node. But when I check the node with “docker ps -a”, it’s always restarting.
Is it because I’m using a 500 Gb with STORAGE=“820GB” in the command ?
As my 1 Tb disk contains only 73 Gb of data, I think I can use the 500 Gb disk as a buffer while I add the 1 Tb disk to the LV.
Or do i have to add 1 Tb to LV before restarting the node (as I need to format the 1 Tb disk, I will loose everything on the disk) ?
Thanks.
LVM is not recommended for the purposes of spanning disks. If one disk fails, all the data are lost.
I would recommend one of the following:
- Wait until the first drive is nearly full and start a second node with the second hard drive.
- Add a third hard drive and use ZFS … raidz
1 Like
Since the second HDD seems to be 500GB that won’t work as it won’t be sufficient for the minimum requirement.
To answer the question, you can’t run a node only on the 500GB HDD as it’s not large enough for the minimum space requirement. But if you remove the data, your node will be disqualified.
So you either need to figure out where to temporarily store your data while you set up the LVM or just stick with the 1TB only. I agree with @anon27637763 though that LVM is not a great idea as you’re increasing the risk of node failure. If either drive is lost, your node will be disqualified. It’s up to you whether you want to take on that risk for the extra 500GB.
Thanks for your replies.
500Gb was planned to be use temporarily just during the time to format the 1 Tb and add it to the Logical Volum.
But I will follow your recommandations and I will run with my 1 Tb disk and wait for it’s full capacity before adding a new disk and a new node.