Satisfactory Server
This is a Dockerized version of the Satisfactory dedicated server.
Setup
According to the official wiki, expect to need 12GB - 16GB of RAM.
You'll need to bind a local directory to the Docker container's /config
directory. This directory will hold the following directories:
/backups
- the server will automatically backup your saves when the container first starts/gamefiles
- this is for the game's files. They're stored outside of the container to avoid needing to redownload 8GB+ every time you want to rebuild the container/saved
- this contains the game's blueprints, saves, and server configuration
Before running the server image, you should find your user ID that will be running the container. This isn't necessary in most cases, but it's good to find out regardless. If you're seeing permission denied
errors, then this is probably why. Find your ID in Linux
by running the id
command. Then grab the user ID (usually something like 1000
) and pass it into the -e PGID=1000
and -e PUID=1000
environment variables.
Run the Satisfactory server image like this:
Note: This is one command make sure to copy all of it!
docker run \
--detached \
--name=satisfactory-server \
--hostname satisfactory-server \
--restart unless-stopped \
--volume /path/to/config:/config \
--env MAXPLAYERS=4 \
--env PGID=1000 \
--env PUID=1000 \
--env STEAMBETA=false \
--memory-reservation=12G \
--memory 16G \
--publish 7777:7777/udp \
--publish 15000:15000/udp \
--publish 15777:15777/udp \
wolveix/satisfactory-server:latest
Explanation of the command:
--detached
-> Starts the container detached from your terminal.
If you want to see the logs replace it with--sig-proxy=false
.--name
-> Gives the container a unqiue name.--hostname
-> Changes the hostname of the container.--restart unless-stopped
-> Enables the restart policy that restarts the container unless it was stopped by the user.--volume
-> Binds the satisfactory config folder to the folder you specified. Allows you to easily access your savegames.- For the environment (
--env
) variables please see here. --memory-reservation
-> Is a memory soft limit.--memory 16G
-> Limits the RAM that the container uses to 16 Gigabytes.--publish
-> Specifies the ports that the container exposes.
Docker Compose
If you're using Docker Compose:
version: '3'
services:
satisfactory-server:
container_name: 'satisfactory-server'
hostname: 'satisfactory-server'
image: 'wolveix/satisfactory-server:latest'
ports:
- '7777:7777/udp'
- '15000:15000/udp'
- '15777:15777/udp'
volumes:
- '/path/to/config:/config'
environment:
- MAXPLAYERS=4
- PGID=1000
- PUID=1000
- STEAMBETA=false
restart: unless-stopped
deploy:
resources:
limits:
memory: 16G
reservations:
memory: 12G
Kubernetes
If you are running a Kubernetes cluster, we do have a service.yaml and statefulset.yaml available under the cluster directory of this repo.
If you are using Helm, you can find charts for this repo on ArtifactHUB. The k8s-at-home helm chart for Satisfactory can be installed with the below.
Some suggested default values.yaml
for the k8s-at-home chart - check out the vaules.yaml for full defaults, and the common chart for more values options.
values.yaml
env:
# Environmental variables as below can be passed in this yaml block
AUTOPAUSE: "true"
MAXPLAYERS: 3
service:
main: # Example setup for a LoadBalancer with an external IP
# MetalLB for example could be used if a Loadbalancer is not provided by your provider
type: LoadBalancer # Setting an external IP for simple port forwarding
externalTrafficPolicy: Cluster
loadBalancerIP: "192.168.2.200" # IP of the satisfactory server
persistence:
config: # Config/save data stored here
enabled: true
server-cache: # Game files stored here
# This is seperated to allow for backing up only game/config data
enabled: true
The values.yaml
could then be installed to your Kubernetes cluster with the below
helm repo add k8s-at-home https://k8s-at-home.com/charts/
helm repo update
helm install satisfactory k8s-at-home/satisfactory -f values.yaml
Environment Variables
Parameter | Default | Function |
---|---|---|
AUTOPAUSE |
true |
pause game when no player is connected |
AUTOSAVEINTERVAL |
300 |
autosave interval in seconds |
AUTOSAVENUM |
5 |
number of rotating autosave files |
AUTOSAVEONDISCONNECT |
true |
autosave when last player disconnects |
CRASHREPORT |
true |
automatic crash reporting |
DEBUG |
false |
for debugging the server |
DISABLESEASONALEVENTS |
false |
disable the FICSMAS event (you miserable bastard) |
MAXOBJECTS |
2162688 |
set the object limit for your server |
MAXPLAYERS |
4 |
set the player limit for your server |
MAXTICKRATE |
30 |
set the maximum sim tick rate for your server |
NETWORKQUALITY |
3 |
set the network quality/bandwidth for your server |
PGID |
1000 |
set the group ID of the user the server will run as |
PUID |
1000 |
set the user ID of the user the server will run as |
SERVERBEACONPORT |
15000 |
set the game's beacon port |
SERVERGAMEPORT |
7777 |
set the game's port |
SERVERIP |
0.0.0.0 |
set the game's ip (usually not needed) |
SERVERQUERYPORT |
15777 |
set the game's query port |
SKIPUPDATE |
false |
avoid updating the game on container start/restart |
STEAMBETA |
false |
set experimental game version |
TIMEOUT |
30 |
set client timeout (in seconds) |
Loading Your Save
To upload your save, connect to the server using the in-game Server Manager. From here, navigate to the save manager. Upload the save, then load it.
Experimental Branch
If you want to run a server for the Experimental version of the game, set the STEAMBETA
environment variable to true
.
How to Improve the Multiplayer Experience
The Satisfactory Wiki recommends a few config tweaks to really get the best out of multiplayer. These changes are already applied to the server, but they need to be applied to your local config too:
- Press
WIN + R
- Enter
%localappdata%/FactoryGame/Saved/Config/WindowsNoEditor
- Copy the config data from the wiki into the respective files
- Right-click each of the 3 config files (Engine.ini, Game.ini, Scalability.ini)
- Go to Properties > tick Read-only under the attributes
Known Issues
- The container is run as
root
. This is pretty common for Docker images, but is bad practice for security reasons. This change was made to address permissions issues - The server log will show various errors; most of which can be safely ignored. As long as the container continues to run and your log looks similar to the example log, the server should be functioning just fine: example log