Search results

The bug reports says that message is only a warning and does not prevent the IP from being set. It also includes a patch showing what to change to get rid of the warning. ETA: You could also install a newer version of Ubuntu. 24.04 is the current LTS release.

You need to set appropriate IP address and gateway in the VM's. You have to do that individually for each one. If you use DHCP, then you need to have a DHCP server on your network (maybe your router?) because Virtualbox bridged network does not provide one. That or you have to set static IP's.

What about in the VM? The IP on vmbr0 has exactly nothing to do with the VM, it is only the IP for the PVE server itself.

You have a different problem. Names appear to be resolving but none of the time servers are reachable (Reach column is zero for all). Something is not correct with your network setup. Perhaps the default gateway is wrong? Perhaps there is a firewall? It is hard to tell. What is the output of...

Ok, but are those the _current_ backups? Or some old ones? I dunno, maybe log into PVE and type "df -h" to see what's mounted and how much is used. This is basic Linux admin stuff, nothing really to do with PVE at all. There are many, many guides and videos you could look up.

Those Guest Additions are specific to Virtualbox. They won't work on PVE (or VMware or Hyper-V). You need instead qemu-guest-agent, which you should be able to install with "apt". apt install qemu-guest-agent If it isn't already installed (and it probably is). In general, if you want to...

Sounds like a problem with the disk. ETA: Maybe run a full SMART test. And check the journal for error messages related to I/O.

Why not just put all of the disks on LVM-thin? Who cares if it isn't in qcow2 format when the storage itself has thin provisioning and snapshots? https://pve.proxmox.com/pve-docs/pve-admin-guide.html#_storage_types

Why? And which one is vmbr0 connected to? The Bridge network won't have a DHCP server unless you have one external somewhere. This seems to be a warning only: https://bugs.launchpad.net/ubuntu/+source/netplan.io/+bug/2041727

You don't seem to have assigned an IP to the Ubuntu VM. Your PVE appears to be running in Virtualbox so there's a whole other layer of networking that you have not described. How it is Virtualbox networking configured (default NAT? Host? Bridged?). How that is set up will determine how you must...

You mean like maybe he's been hacked and a fake "ps" is mining bitcoin or something?

Why do you think you can update the host BIOS from within a VM? If that were possible it would be the biggest security hole ever.

And then set up a serial console or some other way to get in when you pass through your only GPU.

Disable virtualization in BIOS so your pass-thru VM won't start and you have a screen to log in and troubleshot. ETA: Or boot from a Live CD of some sort and fix it from there. How you do this depends on what your setup is (ZFS? Single Disk? SW Raid?).

What is your use-case? I ask because you have no Linux experience and PVE does require a certain amount of that, especially when things go wrong. Maybe you would be better off running VM's under Windows with Hyper-V or Virtualbox or VMWare workstation.

What is the guest OS? What version of PVE? What CPU do you have? It could be a bug. Some older kernels with PVE 7 triggered this bug when run on older hardware. https://forum.proxmox.com/threads/vm-shutdown-kvm-entry-failed-hardware-error-0x80000021.109410/page-1 And this wiki entry...

I would think that is plenty.

Why not? It is just a copy in memory. Try iperf between two VM's if you don't believe it.

Maybe reinstall from the ISO then, and check the journal to make sure there aren't any disk errors. Then redo the upgrade. If you get any errors, resolve them before rebooting.

One way to start would be to tell everyone exactly what didn't work, any error messages, and any other information about your system that might be relevant. ETA: The thread title doesn't really make any sense to me. Makes it sound like the initial install failed but the text says otherwise. Is...