One common additional disk for many vm

[esc]

I would like to connect an additional shared disk to each vm. Is it possible ?

possibly how?

In the example, I have a vm with jdownloader and another vm with plex and how to transfer files here?

 

[hepo]

external storage with NFS mount to both VMs will be the solution that come's to mind.
or "storage VM" that acts as NFS server so the others can connect to it... one stores, other reads

 

[esc]

I tried ntfs, I tried ext4 added as a second disk but the system creates a separate disk for each one. I added via datacenter -> Storage -> Add -> Directory or NFS but then what?
Then I can add this disk to each vm but it will be a separate eg. for
VM1 -> Hardware -> ADD -> Harddisk adds and I have my disk:
100 / vm-100-disk-1
and for the next instance my disk:
101 / vm-101-disk-1

so these are separate disks and how to mount an existing one ?. And how to make a common and non-USB?

 

[Dunuin]

VMs can't use shared virtual disks. If two VMs would access the same disk at the same time the data would get corrupted. So the way to go is to use network shares like hepo already said.
Create a third VM and use a NAS OS like TrueNAS/OpenMediaVault, give it a virtual disk to store stuff and let it share folders using NFS/SMB. Then you can mount that SMB/NFS share to your plex and jdownloader VMs.

 

[peterka]

VMs can't use shared virtual disks. If two VMs would access the same disk at the same time the data would get corrupted. So the way to go is to use network shares like hepo already said.
Create a third VM and use a NAS OS like TrueNAS/OpenMediaVault, give it a virtual disk to store stuff and let it share folders using NFS/SMB. Then you can mount that SMB/NFS share to your plex and jdownloader VMs.

This is what I tried, but from (client-)VMs I can not see the smb-shares, only from external clients I see the smb-shares.

 

[Dunuin]

What does your hosts /etc/network/interfaces look like?

 

[peterka]

VM/UCS runs on vmbr=0, https://192.168.188.12/univention
nextcloud runs as docker under UCS on https://192.168.188.12/nextcloud

=======================================
root@s1:/etc/network# more interfaces
# network interface settings; autogenerated
# Please do NOT modify this file directly, unless you know what
# you're doing.
#
# If you want to manage parts of the network configuration manually,
# please utilize the 'source' or 'source-directory' directives to do
# so.
# PVE will preserve these directives, but will NOT read its network
# configuration from sourced files, so do not attempt to move any of
# the PVE managed interfaces into external files!

auto lo
iface lo inet loopback

iface enp1s0 inet manual
#card

iface enp0s31f6 inet manual
#mainboard

auto vmbr0
iface vmbr0 inet static
address 192.168.188.11/24
gateway 192.168.188.1
bridge-ports enp1s0
bridge-stp off
bridge-fd 0
#card

auto vmbr1
iface vmbr1 inet static
address 192.168.188.13/24
bridge-ports enp0s31f6
bridge-stp off
bridge-fd 0
#mainboard

auto vmbr2
iface vmbr2 inet manual
bridge-ports none
bridge-stp off
bridge-fd 0
#virtual

auto vmbr3
iface vmbr3 inet manual
bridge-ports none
bridge-stp off
bridge-fd 0
#virtual

 

[Dunuin]

auto vmbr0
iface vmbr0 inet static
address 192.168.188.11/24
bridge-ports enp1s0

auto vmbr1
iface vmbr1 inet static
address 192.168.188.13/24
bridge-ports enp0s31f6

You should try to avoid giving Proxmox two IPs on the same subnet or stuff can get mixed up.

 

[peterka]

I moved

auto vmbr1
iface vmbr1 inet static
address 192.168.188.13/24
bridge-ports enp0s31f6

to
address 192.168.189.13/24
then rebootet PVE

this didn not improve the issue
(and I can not access PVE on 192.168.189.13/24 from 192.168.188.xx/24, just to test it)

 

[Dunuin]

(and I can not access PVE on 192.168.189.13/24 from 192.168.188.xx/24, just to test it)

Yeah that is normal. You can't access another subnet without a router doing the routing between them.

 

[peterka]

Yeah that is normal. You can't access another subnet without a router doing the routing between them.

shure, but though I now avoid giging PVE two IPs on the same subnet, I stil can not see the smb-share from an inside VM/client

 

[hepo]

Which vmbrige you have assigned on the storage and vm you want to access the storage?

 

[peterka]

Which vmbrige you have assigned on the storage and vm you want to access the storage?

vmbr0

 

[hepo]

What storage solution are you using?
Let me recap:
- both the VM and storage are on the same bridge (vmbr0)
- you can access the storage from the network e.g. your laptop/desktop
- you cannot access the storage from the VM?!?

Can you access the VM form the network?
Can you reach internet from the VM? e.g. ping yahoo.com
Can you ping the storage from the VM?

 

[peterka]

What storage solution are you using?

https://www.univention.com/products/ucs/
under this, I installed https://nextcloud.com/

Let me recap:
- both the VM and storage are on the same bridge (vmbr0)

UCS/nextcloud run in a VM within PVE.
I created another VM with LM20.1m https://linuxmint.com/edition.php?id=285 and the same on my PC.

- you can access the storage from the network e.g. your laptop/desktop

yes, smb

- you cannot access the storage from the VM?!?

yes, smb

Can you access the VM form the network?

yes

Can you reach internet from the VM? e.g. ping yahoo.com

yes

Can you ping the storage from the VM?

yes

Each network service, which I tested, runs well. Only smb within vmbr0 seems to not working.

 

[peterka]

I switched 192.168.188.11 to the onboard-NIC and removed the card-NIC. The smb-issue stays.
Then I moved the VM/UCS/nextcloud and the VM/client to a virtual vmbr (without connection to the NIC). The smb-issue stays.

 

[hepo]

I really have no further suggestions for you.
I would ditch the UCS and try openmediavault or TrueNAS Core

 

[peterka]

I testes OMV5.5 (https://www.openmediavault.org/) smb-shares. They work from inside/outside to inside/outside PVE6.4 without any problem.