[SOLVED] Debian 12.1 cloud image template network problem?

[compiz]

Hi, I saw that Debian 12 was updated to 12.1 and I thought of making new cloud template.
I run 2 Proxmox 8 servers, 1 in Hetzner and 1 at home, the problem i am facing can be reproduced in both of them.
The steps I did to make the template are:

1) wget https://cloud.debian.org/images/clo...0/debian-12-generic-amd64-20230723-1450.qcow2
2) qm create 8000 --memory 2048 --core 2 --name debian12-cloud --net0 virtio,bridge=vmbr0 --ostype l26
3) qm importdisk 8000 debian-12-generic-amd64-20230723-1450.qcow2 local
4) qm set 8000 --scsihw virtio-scsi-pci --scsi0 local:vm-8000-disk-0
5) qm set 8000 --ide2 local:cloudinit
6) qm set 8000 --boot c --bootdisk scsi0
7) qm set 8000 --serial0 socket --vga serial0
8) Config the vm and cloudinit
9) qm template 8000

Basically I use the same commands to make Ubuntu and Debian cloud image templates.

Now when i make the template like that, I make a full clone of it and boot it up, it always stays long time at networks, it continues after a while but when get it, the vm doesn't have network access at all.
As I login, I type: ip a
And I get: 2: ensl8: BROADCAST,MULTICAST mtu 1500 qdisc noop state DOWN group default qlen 1000 link/ether mac:address brd altname enp0s!8

I have tried to change the nic from virtIO to intel or realtek but it is always the same.
Thankfully I had taken a backup of the 12.0 template and restored that and it works fine.
I assume this is a Debian 12.1 cloud image problem or is there any setting I have done wrong?

 

[bbgeek17]

The image you specified works just fine here without any special tricks.
Try to only do this in step 8:
qm set 8000 -ipconfig0 ip=dhcp
qm set 8000 -sshkeys ~/.ssh/id_rsa.pub

---

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

 

[compiz]

Thanks for the reply,"qm set 8000 -sshkeys ~/.ssh/id_rsa.pub" I am setting ssh keys in the cloud-init but i will try the "qm set 8000 -ipconfig0 ip=dhcp" even though I am pretty sure this is the default and I have been using this form for templates for months without issues. It is just debian12.1 that gives me troubles

 

[bbgeek17]

booting an unmodified image without "-ipconfig":
[ *** ] Job systemd-networkd-wait-online.se…tart running (1min 10s / no limit)

shut it down and add "-ipconfig"
[ 9.093766] cloud-init[321]: ci-info: | eth0 | True | 172.16.150.162 | 255.255.255.0 | global | 4e:ba:2c:94:64:11 |

I am not going to attempt to bisect the image to find out why there is a difference, but if you have free time and want to report back - I'd be curios.

If you feel there is a problem with image and it shouldnt behave as it does, you can report it to maintainers.

---

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

 

[compiz]

This seems to have done the trick qm set 8000 -ipconfig0 ip=dhcp and thank you for your answer
I just find it odd that if I do the procedure from scratch again but with debian 12.0 or 11.7 or ubuntu 22.04.2 it works without it!
Anyway! thanks again, problem solved!

 

[bbgeek17]

you can mark it solved, to make it easier for others to find, by editing the first post and changing the prefix to the title.

---

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

 

[compiz]

I am sorry for digging this one up, I have a question about a configuration for the cloud images.
I see in terminal with the command "parted -l" that the default FS for the VMs is EXT4, is there any command I can run to the cloud-init to make it use btrfs or zfs instead of EXT4?