-
Notifications
You must be signed in to change notification settings - Fork 8
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
High server load messages / low CPU utilization #113
Comments
+1 Got exactly the same issue. With only 2 players got rubber banding and the high server load message in game. Server has more then enough resources with a i5 13500, 20 threads and 64gb, 1 GBit/s network. Meanwhile the enshrouded server used max 2gb ram with a max load of 300% aswell. |
Hey there, well I'm experience the same on my setup. I mostly get messages but the game is still playable with 4. In Discord I see also some complains about the performance even on windows servers. So unfortunately I can't do anything but wait for fixes from the core enshrouded server. A native linux server would also help, so I can get rid of the wine virtualization. For now you can only try out the |
Can this use the already existing data folder and savegame for the wine version, or do I need to migrate the savegame to a new data folder? Edit: I read you can define CPU affinity in proto through WINE_CPU_TOPOLOGY variable. I'm no expert for proton so not sure where to look for this setting. Do you maybe know? This is from the official proton repository https://github.com/search?q=repo%3AValveSoftware%2FProton%20WINE_CPU_TOPOLOGY&type=code
Wonder what the "default_cpu_limit[appid]" is Edit2: Found this in the Proton sources.
Seems like some games don't like high core count so the limit what the game sees by the appid. Enshrouded is not on the list. |
You can use the same setup, just switch to the other docker tag and start it back up. If you run into problems, you would migrate the savegames or use one of the backup zip files. I never played around with the If you want to play around with the proton or wine env vars, just slap them on your docker compose or docker command and start/restart you docker container. The env variable is passed to the start script and gets picked up by proton/enshrouded. services:
enshrouded:
...
environment:
...
- PROTON_CPU_TOPOLOGY=4 PROTON_CPU_TOPOLOGY=6
PROTON_CPU_TOPOLOGY: 4
And don't be surprised, the debug log is not available in the current builds. |
Thanks, yes the CPU_Topology env var is being passed on, I can see it from the logs, but it doesn't seem to have any (positive) effect on CPU usage. It's still capped to 3 cores/threads. I've tried the wine-dev container and that did actually help to reduce the number of ingame messages. I also noticed that the wine image consumes far more RAM (I've seen >5,5 GB with 2 Players) while the Proton image is using barely more than 1 GB, so I guess that is one reason why there is less rubberbanding. But the CPU usage is also capped with the wine image. |
I'm encountering performance issue's as well. I'm hosting a server for 3 players. It's using about ~300% CPU and ~3Gb of ram with lots more resources avalible. |
We have the same messages in the server, when more than 2 people are on the server. When we have 4, the problems become worse and someone losts connection sometimes. We have an I5 4590 in the system. What is really different to the other is our used ram memory. Normally the server uses 12 GB and with 4 poleople on the Server up to 15 GB. |
Is there an existing issue for this?
Used image version
latest - 1d77460
Used docker compose file or docker command
What you expected to see, versus what you actually saw
I'm running this container on a Synology NAS with AMD Ryzen CPU and 6 Cores/12 Threads. As soon as more than one player is connected, the game starts complaining about high server load or even excessive load. While looking into the CPU utilization of the container I see only a utilization of up to 300 %, i.e 3 cores or threads.
The Synology Resource Manager (kind of a taskmanager like in Windows) shows a Cpu utilization of only around 20 %. I've tried many things like redeploying the container, turning on resource restriction, turning it off. Setting the priority of the container to high and even started messing with the docker -cpushares parameter through ssh. Nothing has ever increase CPU utilization.
At this point I suspect somethings off with the container, rather than the host machine or configuration.
Any ideas?
Logs from the container
No response
The text was updated successfully, but these errors were encountered: