Docker shutdown signal not passed on about docker-palworld-dedicated-server

[jammsen]

Fix is applied in my local version:

Console

palworld-dedicated-server  | [S_API FAIL] Tried to access Steam interface SteamUser021 before SteamAPI_Init succeeded.
palworld-dedicated-server  | [S_API FAIL] Tried to access Steam interface SteamFriends017 before SteamAPI_Init succeeded.
palworld-dedicated-server  | [S_API FAIL] Tried to access Steam interface STEAMAPPS_INTERFACE_VERSION008 before SteamAPI_Init succeeded.
palworld-dedicated-server  | [S_API FAIL] Tried to access Steam interface SteamNetworkingUtils004 before SteamAPI_Init succeeded.
^CGracefully stopping... (press Ctrl+C again to force)
[+] Stopping 1/1
 ✔ Container palworld-dedicated-server  Stopped                                                                                                     3.3s
palworld-dedicated-server exited with code 0
canceled

3.3 instead of 10 🚀

docker-compose logs -f

palworld-dedicated-server  | [S_API FAIL] Tried to access Steam interface SteamUser021 before SteamAPI_Init succeeded.
palworld-dedicated-server  | [S_API FAIL] Tried to access Steam interface SteamFriends017 before SteamAPI_Init succeeded.
palworld-dedicated-server  | [S_API FAIL] Tried to access Steam interface STEAMAPPS_INTERFACE_VERSION008 before SteamAPI_Init succeeded.
palworld-dedicated-server  | [S_API FAIL] Tried to access Steam interface SteamNetworkingUtils004 before SteamAPI_Init succeeded.
palworld-dedicated-server  | FUnixPlatformMisc::RequestExitWithStatus
palworld-dedicated-server  | FUnixPlatformMisc::RequestExit
palworld-dedicated-server  | dlmopen steamservice.so failed: steamservice.so: cannot open shared object file: No such file or directory
palworld-dedicated-server  | Exiting abnormally (error code: 143)
palworld-dedicated-server  | Shutdown handler: cleanup.

This should do the trick i think. @uncaught

[0Downtime]

This would be helpful since the server needs rebooted every 2 hours due to memory leak.

[jammsen]

Hey @uncaught
Looks to me like this is the problem? https://docs.docker.com/compose/faq/#why-do-my-services-take-10-seconds-to-recreate-or-stop

[uncaught]

Yes, your entrypoint (servermanager.sh) needs to delegate the signals to the services it starts.

I don't know where PalServer.sh comes from, but if that handles a shutdown signal (the docker stop signal can be adjusted), then maybe this could already do the trick:

-    ./PalServer.sh "$START_OPTIONS"
+    exec ./PalServer.sh "$START_OPTIONS"

exec will replace the top level process, so the signals go there now.

Not sure if you need your own process running still though.

[jammsen]

Hey @uncaught
the PalServer.sh comes with the server, it sets some unreal-engine things.
PalServer.sh:

#!/bin/sh
UE_TRUE_SCRIPT_NAME=$(echo \"$0\" | xargs readlink -f)
UE_PROJECT_ROOT=$(dirname "$UE_TRUE_SCRIPT_NAME")
chmod +x "$UE_PROJECT_ROOT/Pal/Binaries/Linux/PalServer-Linux-Test"
"$UE_PROJECT_ROOT/Pal/Binaries/Linux/PalServer-Linux-Test" Pal "$@" 

But im working on a fix for that, a collegue of mine has something the might work and he was okay with me using it.

[uncaught]

3.3 seconds sounds good :)

[jammsen]

Please go ahead and test things.
New Docker-Compose File is needed.
You can find now examples in the README.md.
The docker-compose.yml was removed because of dublicate code.

[jammsen]

If this issue is solved, please give me feedback on the patch and please consider giving me a Star here and on Docker-Hub.

[jammsen]

@uncaught Ping please read the last 2 comments and tell me things

[uncaught]

I've seen it, but can't test, yet. I've been shutting down the server via rcon lately and it's not running right now. (Damn memory leak ;) )

I've been using the updated script though already since yesterday.

I'm going to assume this is solved (until they change the name of PalServer-Linux-Test I guess ;) ).

[jammsen]

Well its their binary and their shell-script, up to them to manage that beast, hehe 👍