Kyle Weaver created BEAM-7412:
---------------------------------
Summary: portable spark: thread/memory leak in local mode
Key: BEAM-7412
URL: https://issues.apache.org/jira/browse/BEAM-7412
Project: Beam
Issue Type: Bug
Components: runner-spark
Reporter: Kyle Weaver
Assignee: Kyle Weaver
When running validatesPortableRunnerBatch test on local mode, the portable
Spark runner creates ~18k threads before becoming unable to create any more
threads, at which point it crashes. This does not happen on standalone cluster
mode, where ephemeral executors (independent JVMs) are used and then shut down,
preventing whatever leakage is occurring from becoming too much of a problem.
Sample stack trace:
{{"pool-3965-thread-3" #16059 daemon prio=5 os_prio=0 tid=0x00007f9f85a81000
nid=0x32a0 waiting on condition [0x00007f9f8b9f1000]}}
{{ java.lang.Thread.State: TIMED_WAITING (parking)}}
{{ at (C/C++) 0x00007fa2f32c2dae (Unknown Source)}}
{{ at (C/C++) 0x00007fa2f2851351 (Unknown Source)}}
{{ at sun.misc.Unsafe.park(Native Method)}}
{{ - parking to wait for <0x0000000727bda848> (a
java.util.concurrent.SynchronousQueue$TransferStack)}}
{{ at java.util.concurrent.locks.LockSupport.parkNanos(LockSupport.java:215)}}
{{ at
java.util.concurrent.SynchronousQueue$TransferStack.awaitFulfill(SynchronousQueue.java:460)}}
{{ at
java.util.concurrent.SynchronousQueue$TransferStack.transfer(SynchronousQueue.java:362)}}
{{ at java.util.concurrent.SynchronousQueue.poll(SynchronousQueue.java:941)}}
{{ at
java.util.concurrent.ThreadPoolExecutor.getTask(ThreadPoolExecutor.java:1073)}}
{{ at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1134)}}
{{ at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)}}
{{ at java.lang.Thread.run(Thread.java:748)}}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
