Search results

Yeah they wont start whilst you have so many PGs down. You could try restarting all the OSD servers as to be honest I cant see anything sticking out for why they arent peering.

What is your pool setup?

Can all your servers talk to all your other servers? No network issues or changes?

Nowhere does it state in the errors that an OSD is full, you just have a large amount of PG peering. Does this progress at all? Or does it show the same amount of PG's in the same state?

I don't think you understand. When you just install an OS on the server that OS has access to all 24 threads. When you run a VM and only give it 3 or 6 CPU then your only giving it access to a max of 12.5% or 25% of the max CPU power. So your always going to have less performance. I don't think...

Because that one OS would have access to all 24 threads at once..

From my experience game servers are very CPU intensive, I think you are just pushing the CPU harder than it has capacity wise. I don't see this is linked to Proxmox in anyway.

I understand that, however I mean you can't expect more performance than just 3 threads of a CPU that is cut into 24 threads can offer.

So you have a 24 thread CPU, but your only passing through 3 of these threads to that VM.

Are you seeing the load hit high % on the VM? How many CPU core are you passing through to the VM? Can you provide a screenshot of the VM Hardware config page?

That is impossible to say, every CPU is different depends on the environment / temps / motherboard how hard your pushing the CPU. Only testing will show you if it is stable or not, if it becomes unstable / crashes then lower it a bit and try again.

This looks to me like you have created a LVM thin pool, then created a LVM of nearly the full size and mounted this as a filesystem on the VM. If you have wrote up to around the 500GB at some point unless the FS is passing through trim/discard commands the LVM layer wont remove old data untill...

If you altered the CPU on the server to run at a higher frequency than designed then yes you would also benefit of this extra CPU within the VM, however may make the hardware more unstable.

If you dont need to do any live migrations to another server with a different CPU pick HOST (bottom of the list)

Are you applying it in the network config, any nic reset will cause it to be lost hence you need to enter it into your network config as an if up so its applied every time.

I would say this is very much nothing to do with Proxmox then. And more to do with the script that you are running within the VM on PHP. Maybe a different PHP version is being installed on the Proxmox VM's compared to the older/original VMWare VMs you have.

Without any more information its possible to understand what you are talking about issue wise and if its related to Proxmox in anyway. Saying in the logs php, sql could be 1 in a million things. We need more information, but if your running a pure KVM/QEMU then Proxmox really has no/minimal...

Where are you getting / seeing these errors?

This isn't really a Q for Proxmox either, however id check: - Make sure you dont have the Proxmox FW enabled? - Have you connected plex to MyPlex account? - From another PC if you go to IP:32400/web does it load?

What does the route table look like within windows?