I had a Plex vm running on Debian 10 (starting with Proxmox 6) for quite a while with no notable crashing issues. However, recently I have noticed that the VM will not run for more than a day or 2 without becoming completely unusable. When the problem exists, the vm will not unmount the ZFS over iscsi storage inside the VM on shutdown, and I have to manually "stop" the VM from Proxmox.
In order to eliminate possible problems, yesterday I made a brand new Debian 12 VM, attached those ZFS over iscsi disks to the new VM, and migrated my Plex server config to the new VM. After letting it run for a few hours, I installed Radarr, and sure enough the problem came right back. It ran for a few hours before becoming unusable, and I had to manually kill it.
For more info, I am currently running the latest version of PVE as of today (8.2.4) as well as the latest version of Debian 12. I usually notice something has gone wrong when Plex is no longer working and the vm console shows several hung_task_timeout messages.
Here is my storage.cfg and my vm.conf:
Anyone have any recommendations? This is extremely frustrating, and it is very difficult to troubleshoot.
In order to eliminate possible problems, yesterday I made a brand new Debian 12 VM, attached those ZFS over iscsi disks to the new VM, and migrated my Plex server config to the new VM. After letting it run for a few hours, I installed Radarr, and sure enough the problem came right back. It ran for a few hours before becoming unusable, and I had to manually kill it.
For more info, I am currently running the latest version of PVE as of today (8.2.4) as well as the latest version of Debian 12. I usually notice something has gone wrong when Plex is no longer working and the vm console shows several hung_task_timeout messages.
Here is my storage.cfg and my vm.conf:
dir: local path /var/lib/vz content backup,iso,vztmplzfspool: local-zfs pool rpool/data content images,rootdir sparse 1[I][B]zfs: datastor1[/B][/I] blocksize 128k iscsiprovider comstar pool goliath portal 10.0.0.5 target iqn.2010-08.org.illumos:02:8a4ca3a6-a226-6980-a6a1-ae0e913b46a5 content images nowritecache 0[B][I] sparse 1[/I][/B][I][B]zfs: kylefiber[/B][/I] blocksize 128k iscsiprovider comstar pool ringo portal 10.0.0.6 target iqn.2010-08.org.illumos:02:1352e462-fabe-cc35-986f-876c8d241428 content images nowritecache 0[B][I] sparse 1[/I][/B]agent: 1boot: order=scsi0;ide2;net0cores: 14cpu: hostide2: none,media=cdrommemory: 32768meta: creation-qemu=9.0.0,ctime=1721053033name: plex-v3net0: virtio=BC:24:11:EB:D2:41,bridge=vmbr0,firewall=1numa: 0onboot: 1ostype: l26scsi0: vmstor:vm-100-disk-0,discard=on,iothread=1,size=32G,ssd=1[I][B]scsi1: datastor1:vm-102-disk-0,backup=0,discard=on,size=80T[/B][/I][B][I]scsi2: kylefiber:vm-102-disk-0,backup=0,discard=on,size=55T[/I][/B]scsi3: vmstor:vm-102-disk-0,cache=writeback,discard=on,size=750G,ssd=1scsihw: virtio-scsi-singlesmbios1: uuid=f7558573-a3db-4a81-9c96-bd1503afd2d1sockets: 2vmgenid: 1860f258-6c3b-4208-92c4-e2a59ba8b747Anyone have any recommendations? This is extremely frustrating, and it is very difficult to troubleshoot.