My humble #homelab is complete (for tonight)
200mm Noctua exhaust fans are plugged into USB ports. Each cabinet has some open sophets at the toe area to let cooler air in.
Left cabinet houses a #geekpi #minirack. Three lenovo sff PCs run my proxmox cluster.
Right cabinet houses my networking, ups, NAS, and the TV stuff (Wii, #steamlink, Amazon tv).
Really happy how it all came out.
Thanks @geerlingguy for all your videos on the mini rack!
3 kleine Helfer: Proxmox-Spezial
Dieses Mal zeige ich Euch, wie Ihr das Paßwort eines Proxmox Backup Servers ändert und welche Änderungen Ihr auf dem Proxmox-Virtualisierungsserver machen müßt, damit Backups weiterhin möglich sind. Dazu kommen die Abfrage von Statusinformationen und wie Ihr eine physikalische SSD/HDD exklusiv einer VM zuweist.
#proxmox #vm #virtualisierung #linux
https://tube.tchncs.de/w/atg4Cn2TeV3ZPLaysEAYKh
https://youtu.be/ZwraoEdhEQM
I think I figured out the GPU device being named /dev/dri/card0 or card1 (flipping at boot). Some proxmox github discussion stated it was a bug in kernel 6.8, but I'm on 6.14 here. Then someone noted that whether a display is connected or not can cause the change. And that's exactly it - without my monitor connected, the GPU is /dev/dri/card1. It flipped to card0 earlier because I was working directly on the box to fix the network interface issue. Doh.
Here's a link to the comment
https://github.com/tteck/Proxmox/discussions/3235#discussioncomment-10553437
Hello, I'm using the LXC Plex script installation, it works great, but I'm confronted to a problem with my NUC, a Geekom Mini IGT13, with Intel Core i9-13900H (Raptor Lake), and iGPU Intel Iris Xe ...
Fun with #proxmox 9 and AppArmor continues.
https://blog.ktz.me/proxmox-9-made-unprivileged-lxcs-pointless-for-quicksync-users/
Any #proxmox admins here?
Just had a weird issue on my home server, running 9.0.10. After a shutdown to add a new drive, all the network interfaces were renamed. According to /etc/network/interfaces, my devices were supposed to be enp170s0, enp171s0, and wlp172s0.
After shutdown, ip link show listed them as enp171s0, enp172s0, wlp173s0 (all incremented by 1). I had to edit interfaces and restart networking.
Once I was back in the remote web interface, two of my LXCs wouldn't start up because they were expecting the GPU (integrated Intel) at /dev/dri/card1 but it turns out it's now at /dev/dri/card0. As soon as I edited the LXCs to card0, they both started up without issue.
I honestly don't recall this last time I simply rebooted the box, is this some weird NIC vs GPU PCIe thing, where the numbering might change depending on something something?
DefectiveWings ✈️
Linux Solutions
Child of darkness
Klaus Aschenbrenner
LinuxLebenAkaNdWlS
Chris_272
Geekland
Stu
Alex Kretzschmar