Hi everyone.
Glad to post on Lemmy for the first time.
I have an ubuntu that runs a whole jellyfin/arr/torrent docker stack and used to use it as my main work and backup server.
I decided it would be best practice to host my work data on a separate machine in case anything would ever go south virus wise.
I only download and host movies, shows and music there and its all being played through the jellyfin docker.
Am I being overly cautious? Can I even get a virus like that? Has that ever happened?
Or should I continue to separate work and entertainment?
More details on my setup: i3 12100 NVMe 500 GB hosting OS and docker files (including jellyfin cache for snappy access) 5x4TB HDD mergerfs and snapraid
Ubuntu 22 LTS Tailscale Mullvad
You are not being overly cautious. You should absolutely practice isolation. The LastPass hack happened because one of their engineers had a vulnerable Plex server hosted from his work machine. Honestly, next iteration of my home network is going to probably have 4 segments. Home/Users, IOT, Lab, and Work.
How can anyone run a Plex server on their work machine? And why doesn’t their IT dept monitor their devices?
interesting, even if they got access to the plex service, how they could have escaped the plex docker container?
i run pretty much the same stack as OP, but also run immich and paperless. i very much care if someone else have a way to access those…
They weren’t using docker and the Plex software was multiple years out of data:
https://thehackernews.com/2023/03/lastpass-hack-engineers-failure-to.html
The shortcoming, which was discovered and reported to Plex by Tenable in March 2020, was addressed by Plex in version 1.19.3.2764 released on May 7, 2020. The current version of Plex Media Server is 1.31.1.6733.
“Unfortunately, the LastPass employee never upgraded their software to activate the patch,” Plex said in a statement. “For reference, the version that addressed this exploit was roughly 75 versions ago.”
In the LastPass case, I believe it was a native Plex install with a remote code execution vulnerability. But still, even in a Linux container environment, I would not trust them for security isolation. Ultimately, they all share the same kernel. One misconfiguration on the container or an errant privilege escalation exploit and you’re in.
It wasnt containerized sadly but remember in a container you still share (albeit split by cgroups) kernel space and the kernel. Only userland is isolated.
So kernel level sploits are still a concern. Wasn’t the case here but still.
Thanks. I mean I probably don’t have that threat level, but you are right - it also feels good to have isolation.
The downside was the cost of the 2nd machine (~400$) and running it (~5$/month) and the time involved.
But I tend toward thinking it is the right choice
Have you considered replacing the OS with proxmox and running everything in virtual machines?
I haven’t. I always liked to have the possibility to use GNOME on my machine. That wouldn’t work then if I were to have a Ubuntu VM, right?
And do you know if I would I get quicksync access to the i3 for transcoding ?
Thanks
Proxmox has a virtual monitor in its web interface, so you can access the desktop of a virtual machine that way. It’s a little clunky but works ok for quick configuration. Alternately you could remote desktop into the virtual machine.
Quicksync is a little more tricky. GPU pass through is a pain, and I’m not sure off the top of my head about that. You can Google “proxmox quicksync passthrough” and see if any solutions will work for you. There’s a chance that all you would need to do is set the processor type correctly in the virtual machine settings, but I’m not sure.
As the other reply mentioned, you can use the webui for a quick GUI in a pinch, and as for QSV, I’ve had more luck passing through GPUs to LXC containers, as that’s what I’m doing for my Jellyfin setup with an Intel GPU in Proxmox. Link below can hopefully get you started