Datastore sizes? Am I doing something totally wrong?

[rkirk]

After having some issues I never ironed out first go around I decided id dig into rebuilding my beelink NUC that im using to backup my main system as I was having some issues previously with the datastores but I got it working. well this time around appears the same issue. I started with a fresh proxmox install and then installed PBS using the tteck scripts available for the LXC. So I got PBS spun up and went to add my secondary hard drive I installed in the beelink nuc and I can't the drive sizes to align on any of the drives.

I can see the drive fine from proxmox but when I try to add it as a datastore in pbs it's like .01% of the 500gb drive.

Any tips?? Im lost...

Proxmox View


PBS View of Storage


PBS Datastore view

 

[dcsapak]

Hi,

how did you add it as a datastore? what filesystem is on there? can you post your datastore config? (/etc/proxmox-backup/datastore.cfg)

from the symptom I'd guess that the path you entered is wrong and you're not actually using the disks you wanted

 

[rkirk]

Hi,

how did you add it as a datastore? what filesystem is on there? can you post your datastore config? (/etc/proxmox-backup/datastore.cfg)

from the symptom I'd guess that the path you entered is wrong and you're not actually using the disks you wanted

Here you go.

root@proxmox-backup-server:~# cat /etc/proxmox-backup/datastore.cfg
datastore: ssd
comment
notification-mode notification-system
path /dev/sda

datastore: hdd
comment
notification-mode notification-system
path /dev/sdb

datastore: nvme
comment
notification-mode notification-system
path /dev/nvme0n1p3

datastore: test
comment
notification-mode notification-system
path /dev/sdb1

 

[dcsapak]

ok, yes as i assumed, you simply used the path of the block device for the datastore. This will not work. In linux the disks are represented as files in /dev/ (e.g. /dev/sda) but those are not usable as a filesystem, for that you have to format the disk etc.
to properly use it, remove the datastore from the config and use the Administration -> Storage/Disks Panel
Use either 'Directory' or 'ZFS' (depending on what you want/need) and use the create dialog there where you can select the disk

this will create a filesystem for you (and add the datastore automatically if you check the box)

 

[rkirk]

ok, yes as i assumed, you simply used the path of the block device for the datastore. This will not work. In linux the disks are represented as files in /dev/ (e.g. /dev/sda) but those are not usable as a filesystem, for that you have to format the disk etc.
to properly use it, remove the datastore from the config and use the Administration -> Storage/Disks Panel
Use either 'Directory' or 'ZFS' (depending on what you want/need) and use the create dialog there where you can select the disk

this will create a filesystem for you (and add the datastore automatically if you check the box)

So while trying to create the datastore as you instructed by going to stroage and disks / directory and I get this error.


Ran this command and got this from the logs.

root@proxmox-backup-server:/var/log/proxmox-backup# journalctl --since "2025-02-19 10:25:00" --until "2025-02-19 10:26:00"
Feb 19 10:25:05 proxmox-backup-server zed[11585]: Failed to initialize libzfs
Feb 19 10:25:05 proxmox-backup-server systemd[1]: zfs-zed.service: Main process exited, code=exited, status=1/FAILURE
Feb 19 10:25:05 proxmox-backup-server systemd[1]: zfs-zed.service: Failed with result 'exit-code'.
Feb 19 10:25:05 proxmox-backup-server systemd[1]: zfs-zed.service: Scheduled restart job, restart counter is at 4016.
Feb 19 10:25:05 proxmox-backup-server systemd[1]: Stopped zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:05 proxmox-backup-server systemd[1]: Started zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:05 proxmox-backup-server zed[11586]: ZFS Event Daemon 2.2.7-pve1 (PID 11586)
Feb 19 10:25:15 proxmox-backup-server zed[11586]: Failed to initialize libzfs
Feb 19 10:25:15 proxmox-backup-server systemd[1]: zfs-zed.service: Main process exited, code=exited, status=1/FAILURE
Feb 19 10:25:15 proxmox-backup-server systemd[1]: zfs-zed.service: Failed with result 'exit-code'.
Feb 19 10:25:15 proxmox-backup-server systemd[1]: zfs-zed.service: Scheduled restart job, restart counter is at 4017.
Feb 19 10:25:15 proxmox-backup-server systemd[1]: Stopped zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:15 proxmox-backup-server systemd[1]: Started zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:15 proxmox-backup-server zed[11587]: ZFS Event Daemon 2.2.7-pve1 (PID 11587)
Feb 19 10:25:25 proxmox-backup-server zed[11587]: Failed to initialize libzfs
Feb 19 10:25:25 proxmox-backup-server systemd[1]: zfs-zed.service: Main process exited, code=exited, status=1/FAILURE
Feb 19 10:25:25 proxmox-backup-server systemd[1]: zfs-zed.service: Failed with result 'exit-code'.
Feb 19 10:25:26 proxmox-backup-server systemd[1]: zfs-zed.service: Scheduled restart job, restart counter is at 4018.
Feb 19 10:25:26 proxmox-backup-server systemd[1]: Stopped zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:26 proxmox-backup-server systemd[1]: Started zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:26 proxmox-backup-server zed[11590]: ZFS Event Daemon 2.2.7-pve1 (PID 11590)
Feb 19 10:25:36 proxmox-backup-server zed[11590]: Failed to initialize libzfs
Feb 19 10:25:36 proxmox-backup-server systemd[1]: zfs-zed.service: Main process exited, code=exited, status=1/FAILURE
Feb 19 10:25:36 proxmox-backup-server systemd[1]: zfs-zed.service: Failed with result 'exit-code'.
Feb 19 10:25:36 proxmox-backup-server systemd[1]: zfs-zed.service: Scheduled restart job, restart counter is at 4019.
Feb 19 10:25:36 proxmox-backup-server systemd[1]: Stopped zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:36 proxmox-backup-server systemd[1]: Started zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:36 proxmox-backup-server zed[11591]: ZFS Event Daemon 2.2.7-pve1 (PID 11591)
Feb 19 10:25:46 proxmox-backup-server zed[11591]: Failed to initialize libzfs
Feb 19 10:25:46 proxmox-backup-server systemd[1]: zfs-zed.service: Main process exited, code=exited, status=1/FAILURE
Feb 19 10:25:46 proxmox-backup-server systemd[1]: zfs-zed.service: Failed with result 'exit-code'.
Feb 19 10:25:46 proxmox-backup-server systemd[1]: zfs-zed.service: Scheduled restart job, restart counter is at 4020.
Feb 19 10:25:46 proxmox-backup-server systemd[1]: Stopped zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:46 proxmox-backup-server systemd[1]: Started zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:46 proxmox-backup-server zed[11592]: ZFS Event Daemon 2.2.7-pve1 (PID 11592)
Feb 19 10:25:56 proxmox-backup-server zed[11592]: Failed to initialize libzfs
Feb 19 10:25:56 proxmox-backup-server systemd[1]: zfs-zed.service: Main process exited, code=exited, status=1/FAILURE
Feb 19 10:25:56 proxmox-backup-server systemd[1]: zfs-zed.service: Failed with result 'exit-code'.
Feb 19 10:25:56 proxmox-backup-server systemd[1]: zfs-zed.service: Scheduled restart job, restart counter is at 4021.
Feb 19 10:25:56 proxmox-backup-server systemd[1]: Stopped zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:56 proxmox-backup-server systemd[1]: Started zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:25:56 proxmox-backup-server zed[11593]: ZFS Event Daemon 2.2.7-pve1 (PID 11593)


If I go to storage and disks / ZFS I get this error and this output from the logs.



root@proxmox-backup-server:/var/log/proxmox-backup# journalctl --since "2025-02-19 10:46:00" --until "2025-02-19 10:47:00"
Feb 19 10:46:01 proxmox-backup-server proxmox-backup-api[180]: error getting zfs devices: command "zpool" "list" "-H" "-p" "-P" "-v" failed - status code: 1 - /dev/zfs and /proc/self/mounts are required.
Feb 19 10:46:01 proxmox-backup-server proxmox-backup-api[180]: Try running 'udevadm trigger' and 'mount -t proc proc /proc' as root.
Feb 19 10:46:01 proxmox-backup-server proxmox-backup-api[180]: failed to gather smart data for /dev/sda – command "smartctl" "-H" "-A" "-j" "/dev/sda" failed - status code: 2 - no error message
Feb 19 10:46:01 proxmox-backup-server proxmox-backup-api[180]: failed to gather smart data for /dev/nvme0n1 – command "smartctl" "-H" "-A" "-j" "/dev/nvme0n1" failed - status code: 2 - no error message
Feb 19 10:46:01 proxmox-backup-server proxmox-backup-api[180]: failed to gather smart data for /dev/sdb – command "smartctl" "-H" "-A" "-j" "/dev/sdb" failed - status code: 1 - no error message
Feb 19 10:46:06 proxmox-backup-server zed[11973]: Failed to initialize libzfs
Feb 19 10:46:06 proxmox-backup-server systemd[1]: zfs-zed.service: Main process exited, code=exited, status=1/FAILURE
Feb 19 10:46:06 proxmox-backup-server systemd[1]: zfs-zed.service: Failed with result 'exit-code'.
Feb 19 10:46:06 proxmox-backup-server systemd[1]: zfs-zed.service: Scheduled restart job, restart counter is at 4139.
Feb 19 10:46:06 proxmox-backup-server systemd[1]: Stopped zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:46:06 proxmox-backup-server systemd[1]: Started zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:46:06 proxmox-backup-server zed[11984]: ZFS Event Daemon 2.2.7-pve1 (PID 11984)
Feb 19 10:46:11 proxmox-backup-server proxmox-backup-api[180]: command "zpool" "create" "-o" "ashift=12" "-m" "/mnt/datastore/storagessd" "storagessd" "sda" failed - status code: 1 - /dev/zfs and /proc/se>
Try running 'udevadm trigger' and 'mount -t proc proc /proc' as root.
Feb 19 10:46:16 proxmox-backup-server zed[11984]: Failed to initialize libzfs
Feb 19 10:46:16 proxmox-backup-server systemd[1]: zfs-zed.service: Main process exited, code=exited, status=1/FAILURE
Feb 19 10:46:16 proxmox-backup-server systemd[1]: zfs-zed.service: Failed with result 'exit-code'.
Feb 19 10:46:16 proxmox-backup-server systemd[1]: zfs-zed.service: Scheduled restart job, restart counter is at 4140.
Feb 19 10:46:16 proxmox-backup-server systemd[1]: Stopped zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:46:16 proxmox-backup-server systemd[1]: Started zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:46:16 proxmox-backup-server zed[11986]: ZFS Event Daemon 2.2.7-pve1 (PID 11986)
Feb 19 10:46:22 proxmox-backup-server proxmox-backup-api[180]: GET /api2/json/nodes/localhost/disks/zfs: 400 Bad Request: [client [::ffff:192.168.1.51]:64932] command "zpool" "list" "-H" "-p" "-P" failed >
Try running 'udevadm trigger' and 'mount -t proc proc /proc' as root.
Feb 19 10:46:26 proxmox-backup-server zed[11986]: Failed to initialize libzfs
Feb 19 10:46:26 proxmox-backup-server systemd[1]: zfs-zed.service: Main process exited, code=exited, status=1/FAILURE
Feb 19 10:46:26 proxmox-backup-server systemd[1]: zfs-zed.service: Failed with result 'exit-code'.
Feb 19 10:46:26 proxmox-backup-server systemd[1]: zfs-zed.service: Scheduled restart job, restart counter is at 4141.
Feb 19 10:46:26 proxmox-backup-server systemd[1]: Stopped zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:46:26 proxmox-backup-server systemd[1]: Started zfs-zed.service - ZFS Event Daemon (zed).
Feb 19 10:46:26 proxmox-backup-server zed[11987]: ZFS Event Daemon 2.2.7-pve1 (PID 11987)


Looks like it's pissed about ZFS-ZED.Service??

This is a new install and I have no qualms starting from scratch on the host or the pbs lxc ive spun up. and def appreciate the help.

 

[rkirk]

ok, yes as i assumed, you simply used the path of the block device for the datastore. This will not work. In linux the disks are represented as files in /dev/ (e.g. /dev/sda) but those are not usable as a filesystem, for that you have to format the disk etc.
to properly use it, remove the datastore from the config and use the Administration -> Storage/Disks Panel
Use either 'Directory' or 'ZFS' (depending on what you want/need) and use the create dialog there where you can select the disk

this will create a filesystem for you (and add the datastore automatically if you check the box)

Been looking at this again this evening and im still coming up short.

Should I be creating the datastore on the PVE host thats running PBS in an LXC or should I be doing it from PBS interface? I can create and manipulate the discs from the PVE host without an issue, but when I to do it from PBS I get errors.

 

[dcsapak]

that's weird, pve and pbs use very similar methods to extract e.g. the smart data

can you maybe post the versions of both products?

pveversion -v
proxmox-backup-manager versions --verbose

EDIT:
ah i just now saw that you have pbs in a container... then it's no wonder not everything works
in that case i'd say you have to either (depending on what you want exactly):
* create a filesystem on the host and bind mount it into the container (so both can access it)
* create a vm/ct storage on the host and give the container a new disk (this filesystem is then only visible in the container, not on the host)

in either case you have to choose a directory with the mount and use that for the datastore on pbs