spark

spark

95M Downloads

Spark dies after trying to stop a profiler.

Cy4Shot opened this issue ยท 4 comments

commented

When running /spark profiler --stop, 3 errors are printed in the console, 5 seconds apart:

[25Jul2022 12:39:36.645] [spark-worker-pool-1-thread-4/WARN] [spark/]: A command execution has not completed after 5 seconds, it might be stuck. Trace:
[email protected]/jdk.internal.misc.Unsafe.park(Native Method)
[email protected]/java.util.concurrent.locks.LockSupport.park(LockSupport.java:211)
[email protected]/java.util.concurrent.locks.AbstractQueuedSynchronizer.acquire(AbstractQueuedSynchronizer.java:715)
[email protected]/java.util.concurrent.locks.AbstractQueuedSynchronizer.acquire(AbstractQueuedSynchronizer.java:938)
[email protected]/java.util.concurrent.locks.ReentrantLock$Sync.lock(ReentrantLock.java:153)
[email protected]/java.util.concurrent.locks.ReentrantLock.lock(ReentrantLock.java:322)
TRANSFORMER/[email protected]/me.lucko.spark.common.SparkPlatform.lambda$executeCommand$2(SparkPlatform.java:300)
TRANSFORMER/[email protected]/me.lucko.spark.common.SparkPlatform$$Lambda$23947/0x00000008033853d8.run(Unknown Source)
[email protected]/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:539)
[email protected]/java.util.concurrent.FutureTask.run(FutureTask.java:264)
[email protected]/java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:304)
[email protected]/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)
[email protected]/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
[email protected]/java.lang.Thread.run(Thread.java:833)

After this, spark commands do not work.

System:

  • Running on a 1.18.2 Forge Server. Forge version: 40.1.54.
  • Running Linux (amd64), 8 CPU threads available.
  • CPU is AMD Ryzen 7 5800X 8-Core Processor.
  • Java 17.0.3.1.

Modlist:

commented

Do you have the other error messages to hand?

commented

Do you have the other error messages to hand?

Nope. Nothing while starting the server, or when running the command.

commented

I can't help without seeing the other error messages :(

commented

Seems like Spark freezes when an existing command is not completed yet, which is what my case was, as Spark timed out trying to upload profiler results. Took a while for Spark to spit out connection timed out error and everything is ok after that.