| 27 Sep 2022 |
 Elnur M. | VBoxManage modifyvm talos-1 --cpuidset 00000001 000106e5 00100800 0098e3fd bfebfbff
seems like VBOX commands above have no effect at all, could be a bug, as I've recently upgraded it | 13:01:13 |
 Gerard de Leeuw | Hmm weird. I don't know much about VirtualBox unfortunately, so I hope maybe someone else knows.. | 13:19:43 |
 Fabian B | I geit "Error starting kvm device plugin" when I try to create a VM with KubeVirt in Talos. Is there anything I have to configure in order to be able to start VMs on Talos (on bare-metal)? | 15:07:55 |
 usrbinkat | Right now Kubevirt only works on up to kubernetes version 1.24.x | 15:11:04 |
| usrbinkat | They are working on patches to support 1.25 | 15:11:19 |
| usrbinkat | That might be the problem? | 15:11:29 |
| Fabian B | Ah! Interesting, yes that might be the problem indeed. Im running 1.25.0. | 15:11:59 |
| Fabian B | Can I downgrade to 1.24 via talosctl? | 15:12:17 |
| usrbinkat | I do not know the answer to that myself | 15:12:37 |
| Fabian B | Let's see I just edited the image version in the controlplane and worker configs from 1.25 to 1.24 and applied it, its a test cluster anyway. | 15:16:47 |
|  @dallas_bruno:matrix.org left the room. | 17:37:48 |
 Katie Patterson | Hi all! can I get a show of hands for who will be attending KubeCon North America in a few weeks? | 18:10:50 |
| 28 Sep 2022 |
|  Daniel Mawhirter joined the room. | 00:20:37 |
| Daniel Mawhirter changed their display name from _slack_taloscommunity_U0448EKBW4C to Daniel Mawhirter. | 00:20:38 |
| Daniel Mawhirter set a profile picture. | 00:20:38 |
| Daniel Mawhirter changed their profile picture. | 00:34:38 |
|  Jason Sievert joined the room. | 01:30:51 |
| Jason Sievert changed their display name from _slack_taloscommunity_U044VT96ZGQ to Jason Sievert. | 01:30:51 |
| Jason Sievert set a profile picture. | 01:30:52 |
| Fabian B | I can confirm kubevirt works with Talos and k8s 1.24 thx usrbinkat | 07:32:11 |
 andrey | > Can I downgrade to 1.24 via talosctl?
talosctl upgrade-k8s doesn't support downgrades, as it might not always be safe to do so, but yep, manual config patching works | 09:15:44 |
|  Shahaf Benita changed their profile picture. | 13:47:56 |
| Shahaf Benita | Hi all,
we are running several clusters using kubeadm.
we would like to use only the worker part for nodes to join to our existing cluster initiated by kubeadm (is it even possible?)
I am trying to init a worker using PXE. and it does not seem to get the configuration from my http service using the following configuration:
label talos
menu default
kernel /boot/talos-vmlinuz-amd64
initrd /boot/talos-initramfs-amd64.xz
append initrd=/boot/talos-initramfs-amd64.xz init_on_alloc=1 slab_nomerge console=tty0 console=ttyS0 printk.devkmsg=on talos.platform=metal talos.config=http://10.110.74.49/worker.yaml
I can see the kernel loading but it does not seem to make a request to the http server to get the worker.yaml file that I generated with talosctl. | 13:47:56 |
| andrey | Talos workers can't join kubeadm-based control plane nodes, as they don't have trustd Talos component. | 13:50:01 |
| andrey | (vice versa should work though) | 13:50:10 |
| Shahaf Benita | oh got it thanks! | 13:51:15 |
| Shahaf Benita | and about the configuration, do you see any reason for it not to contact the http service at boot ? | 13:51:32 |
| andrey | as for the config not being loaded - the easiest is to look at the console logs to see why it failed.
if using regular console you can add panic=10 arg to see the message before it reboots | 13:51:48 |
| andrey | https://www.talos.dev/v1.2/reference/kernel/#panic | 13:52:14 |
| Shahaf Benita | alright, great thanks ! | 13:52:20 |
