You are not logged in.
Looking for some help with the graphical console because its frozen. This only seems to be affecting my debian VMs (i got 2 in total). Ive used my debian VMs for weeks with success. Today i decided to work with one and the top half of the graphical console is sort of semi-frozen (it reacts to my cli input but freezes again immediately). The bottom half seems to be completely frozen. Surprisingly this only affects the VMs with debian installed on them but NOT the ones with Windows 10, Windows server 2022 and FreeBSD.
My setup:
i3wm
qemu and virt-manager
The Virtual machines:
Debian 12
Both are very similar and have no DE or WM
Both use 2 vCPU and 4gib of memory
Hardware:
i7 8550U
Intel UHD Graphics 620
40 GB DDR4 ram
Ask me more questions if needed.
Offline
If the VMs were running kernel 6.1.64-1 with ext4 you may have encountered https://bugs.debian.org/cgi-bin/bugrepo … ug=1057843.
If so see if fsck can do anything then try upgrading to the 12.4 point release, which includes a fix.
If not check the journal for clues. This might be better pursued over at the Debian forums though. I would join you there but I've been banned...
Jin, Jîyan, Azadî
Offline
Filesystem is ext4 and the kernel version is 6.1.0-17-amd64
Offline
So what does the journal say? And what do you mean by "graphical console"?
Jin, Jîyan, Azadî
Offline
The graphical console is the actual screen of the VM. Im using this word because its what its called when hovering over the button to it on virt-manager
Last edited by bananahorde (2024-01-03 17:05:48)
Offline
Sorry but I've never used virt-manager. Can you boot the disk with the qemu-system-x86_64 command? I am familiar with that and it would help eliminate virt-manager as the cause of the problem.
Any news on the journal contents? It's very bad form to ignore questions when seeking help.
Jin, Jîyan, Azadî
Offline
Im not exactly sure what i am supposed to be looking for in the journal. Any keywords i could use with grep?
Im gonna try booting with the qemu-system-x86_64 command. Give me some time for this, ive never actually done it with a command. Still quite new to this QEMU/KVM stuff.
Forgive my newbie approach to this forum, im not very experienced. I should have answered both questions.
Offline
Im not exactly sure what i am supposed to be looking for in the journal. Any keywords i could use with grep?
If you upload it to a pastebin and share a link here we can look at it. Best not to grep stuff out because that can miss important context but any errors or warnings in general might be pertinent.
If the VM runs X then also check the log(s) for that.
Im gonna try booting with the qemu-system-x86_64 command. Give me some time for this, ive never actually done it with a command
The wiki has a page for that: https://wiki.archlinux.org/title/QEMU#R … zed_system
Jin, Jîyan, Azadî
Offline
Alright i am back. I ran the command
sudo qemu-system-x86_64 -hda debian12.qcow2 -m 2048The vm started without any freezing screen problems and works just as well as it did before all the freezing problems started. I still havent checked the Journal but im not sure if thats needed now?
Last edited by bananahorde (2024-01-03 17:31:43)
Offline
Probably not but we now need to find out what's wrong with virt-manager.
Does it have a log? Sorry but I really don't know anything about it.
Jin, Jîyan, Azadî
Offline
There seems to be some logs at ~/.cache/virt-manager
Theres 4 log files and i assume that its one log file for each VM. But i have 5 in total so ones missing. Not quite sure though...
Offline
My assumption about "one log file for each VM" seems to be false. It seems like its based on time. This one in the pastebin is from today when looking at the timestamps. The other 4 files seem to be from other days.
I am not sure if these are the ONLY log files for virt-manager but its the ones i found. There were some other file locations suggested on the internet but i wasnt able to find them.
Last edited by bananahorde (2024-01-03 18:03:30)
Offline
I'm experiencing the same poor performance using virt-manager virtual console with Linux VMs. I have a Debian 12 server (no desktop), Kali with XFCE, and Windows 10 VMs. The virtual console performance is fine with the Windows VM, but performance is very bad with both Linux VMs. I've confirmed it is only the virtual console by SSHing into Linux VM's and everything is OK that way.
I tried SSH'ing to the both Linux VM's and installing updates but that didn't help.
To remove virt-manager, I used the virsh utility for libvirt to start the Debian VM and then use a stand-alone SPICE viewer to connect to the VM's SPICE display. I tried three SPICE viewers and they all had the same bad performance (virt-viewer, remote-viewer, and spicy).
virt-manager appears to be using libvirt under the hood and the virtual machine configuration is saved in /etc/libvirt/qemu/<vm_name>.xml e.g. /etc/libvirt/qemu/debian12.xml.
The relevant parts seem to be
<graphics type='spice' autoport='yes'>
<listen type='address'/>
<image compression='off'/>
</graphics>
<video>
<model type='virtio' heads='1' primary='yes'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x0'/>
</video>
This creates a SPICE listener on localhost:5900 which is what I was connecting to with the stand-alone SPICE viewer apps.
I don't see any errors in journalctl or in the SPICE viewer apps, so not much to go on.
Offline
The issue was resolved for me with QEMU 8.2.0-2.
Offline