Search results

Would node 1 (still on 5.4 but corosync 3) still be part of the HA cluster so i can quickly migrate to node2 and node3 (and also test the newer kernal works with my containers)

I'm following the guide PVE 5 to 6 but i wanted to check that after i have stopped all the HA services and performed the corosync upgrade and then upgraded all the nodes other packages. Will my containers all still be running fine on the node i left them on ? or will i need to reboot each of the...

I take it you've done a clean boot of the host, since the last try ? as the card might need to be reset gracefully. further than that i'm out of options, having not one this on my box for years now.

So without any PCI gpu pass through you can't complete the windows install with VNC ?

Going back a few years i had to use vnc to get the install completed, then pass through the gpu. I belive when the system collects info it will try to start the graphics card, which may then not graceful reset, causing the box to lock up, just my two cents.

Or test in the username field to make sure the characters are being typed correctly.

Yes, normally it's cause my keyboard layout has changed to something else and my password has a differently mapped character. I normally have an easy password and once everything is setup as i want i change the password to a more complex one. (just makes it quicker if having to type long...

Also have a look at iftop, if plex is pulling the file really quick over SMB you could be maxing out the network and it's unable to stream with the remaining bandwidth.

Amen

Windows RDP screen size is setup on connection, and you can't change the size via right clicking on the desktop. Proxmox console doesn't pass sound. It's pretty normal

Some cards need a dummy plug or a real monitor plugged in to allow full usage, tho which cards and whether it's just 3D acceleration i don't know for sure.

Wait so your setup is <Proxmox host> Bear metal <Ubuntu 18.04> KVM Virtual Machine <Docker inside KVM> Plex I mean the first thing i would try is running it in a contianer on the proxmox host, or only one level deep in a KVM VM (with the cpu set to host)

I would try a version of mint that is up to date, Mint 17 stopped getting support April 2019. I've got 19.2 and it works fine for me.

Yes, sorry i meant a KVM VM with containers inside it. I forgot ZFS wasn't an option. Hmmmm i feel a project coming on.......

So shower thought... If I had a 3 node cluster with ZFS and live migration of one virtual machine between nodes, and in the virtual machine i had containers running. I could (in theory) live migrate the virtual machine (with containers running on it) to another node and the containers would...

Any details on the motherboard you are using ? what BIOS version you were using and then what version you downgraded to ? it all helps others that have issues and stumble across a post that matches, but just says "Don't worry, fixed it"

If someone bad has had access to your box, I wouldn't trust it any more. Backup your containers and VMS and install a fresh proxmox host. Unless you can be 100% sure they have not left any nasty things on it.

From the logs it looks like someone has access to your Proxmox Host.

Could you not use the ZFS send in proxmox and send snapshots every 5 minutes?

I kept getting asked this, as we only have two machines, the short answer as above it no. You need to have odd voting, so in theory you could have a node with 2 votes and one with 1, but this will mean it will never really failover (unless you then edit the votes after the host goes down) It's...