Hey all long time user first time poster.
Got cant say i have ever had an issue running Plex ever but i seem to have a memory leak within the application.
Im running a Dell R710 with Proxmox
CPU(s) 24 x Intel(R) Xeon(R) CPU X5660 @ 2.80GHz (2 Sockets)
Kernel Version Linux 5.4.60-1-pve #1 SMP PVE 5.4.60-2 (Fri, 04 Sep 2020 10:24:50 +0200)
98Gig of ram
GPU is a Quadro P2000
I guess i will start with the full story as it my or not be connected.
Recently i had Plex use all of its os drive 64gig or there abouts, i found it odd but did very minor investigation and put it down to maybe because i run 13tb of media on there, so i expanded the boot volume by 10 gig and moved on.
Which brings us to two days later, i went to watch a movie and the server was un-responsive, i checked and sure enough full cpu mem and swap.
I stoped and restarted and i watched it climb till max, so then i decided what the hell i only assigned 8gig ram to it maybe ill give it another 8, sure enough same thing happened.
I have just cleared all my logs and done a clean reboot and let the memory slowly max and downloaded the log right before crash.
How are you running Plex on Proxmox? Are you installing it right on the base install of Proxmox or are you using a virtual machine and then installing Plex in the virtual machine?
I’ll give you my advice… I don’t think you should be running Plex inside an LXC container. Containers are dumbed down versions of Linux without all the bells and whistles. Since Proxmox is qemu/KVM I highly recommend installing a full CentOS 7 system ontop of Proxmox and then using podman/docker to run Plex. Or if you choose CentOS Stream then just podman…
I know this isn’t what you want to hear but I would scrap Proxmox alltogether and use XCP-NG. I run CentOS 8 ontop of XCP-NG and use the same version of Plex as you via podman and have no memory leaks.
I seriously think it’s either a container issue or a problem with Proxmox itself.
Scrap container, install full VM using qemu/KVM on Proxmox. If no memory leak then it was a container issue. If memory leak still present then Proxmox issue.
kinda disagree with that, anyways did some digging looks like something went astray in the db, could have been from a power outage that occurred during the floods.
Removed plex wiped its files, and reinstalled, directory went from 60+gb to 8 which it was before these problems, all running fine now.
I used Centos7 because i pass through my GPU from the PVE, never had an issue till now, been running for 5+ years without issue.
