[SOLVED] how to investigate a running VM which does not work

[eseep]

Greetings,
I did restarted a Debian VM with the command `reboot` from SSH terminal.
After this, the VM would not be accessible.
I then look at proxmox webui and manually restart the VM.
I can't connect nor ping the VM.
The webui list the VM as running without any problem but i see that noVNC does not work on this VM which is unusual.
I tried the CLI method of `qm stop/start` but does not have any error message.
I do not know what to do to investigate this issue since there is no error message anywhere and proxmox list the VM as running without any issue.

Could you recommand me some way to investigate this issue before I simply try to reboot the Proxmox server ?
thanks

 

[bbgeek17]

Try from command line as single command from a stopped state:
qm start [vmid];qm terminal [vmid]

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[eseep]

Try from command line as single command from a stopped state:
qm start [vmid];qm terminal [vmid]

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

thankes, unfortunately it seems I do not have a terminal :

# qm start 102; qm terminal 102
Use of uninitialized value in split at /usr/share/perl5/PVE/QemuServer/Cloudinit.pm line 100.
generating cloud-init ISO
unable to find a serial interface

 

[bbgeek17]

you could try to add it, but it may not output anything, depending on your OS setup
qm set $VMID --serial0 socket --vga serial0

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[eseep]

you could try to add it, but it may not output anything, depending on your OS setup
qm set $VMID --serial0 socket --vga serial0

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

Thank you @bbgeek17,
your suggestion gave me the hint I needed.
Indeed this VM have an iGPU pass through. I did not realised it is not compatible with the noVNC interface of the webui since I manage the server from SSH.
desactivating the pass through gave me back access to noVNC.
I could see a file system corruption at boot which required a fsck.
upon doing this, the system booted correctly and I could re activate the iGPU pass through and manage my VM with SSH again.
thanks again for your help !