boreal marsh
https://spark.lucko.me/K5rBM8yOKN
Server has quite a bit of memory to play with but it seems to be only allowing it to use 2GB when spark creates a profile. I'm not sure if it's just an error on spark's end but the GC is undeniable and the server is constantly hiccup'ing and running around 2000ms behind.
pallid pagoda
What are your startup arguments?
With hiccups make sure you don’t allocate unlimited CPU, I’ve had issues with allocating unlimited before.
boreal marsh
What are your startup arguments?
With hiccups make sure you don’t allocate unlim...
java -Xms128M -Xmx$((14305-1024))M -XX:+UseG1GC -XX:+ParallelRefProcEnabled -XX:MaxGCPauseMillis=200 -XX:+UnlockExperimentalVMOptions -XX:+DisableExplicitGC -XX:G1NewSizePercent=30 -XX:G1MaxNewSizePercent=40 -XX:G1HeapRegionSize=8M -XX:G1ReservePercent=20 -XX:G1HeapWastePercent=5 -XX:G1MixedGCCountTarget=4 -XX:InitiatingHeapOccupancyPercent=15 -XX:G1MixedGCLiveThresholdPercent=90 -XX:G1RSetUpdatingPauseTimePercent=5 -XX:SurvivorRatio=32 -XX:+PerfDisableSharedMem -XX:MaxTenuringThreshold=1 -Dusing.aikars.flags=https://mcflags.emc.gs -Daikars.new.flags=true -Dterminal.jline=false -Dterminal.ansi=true -jar server.jar
I'll try also setting the CPU usage lower, should I just set it to 100%?
pallid pagoda
ram avaliable will vary depending on what those startup args have allocated. I can see you set a range for xmx which will create that behaviour. I do the following for my max ram to ensure I cannot overuse memory but also to ensure that the container will get the memory it needs. -XX:MaxRAMPercentage=95.0
Thats worked for years. Weather or not you wanna do it like that is up to you.
But it does me that the full amount is allocated.
https://spark.lucko.me/86uPlyMuje
Here you can see that the allocated memory to the process is only 1GB, but the container can easilly ask for more and will see the full amount that is allocated to the server.
