Alexander Belyak created IGNITE-23569:
-----------------------------------------
Summary: Jdnc connection close exception left running threads
Key: IGNITE-23569
URL: https://issues.apache.org/jira/browse/IGNITE-23569
Project: Ignite
Issue Type: Improvement
Components: jdbc
Affects Versions: 3.0
Reporter: Alexander Belyak
Attachments: 24498.zip, bb.zip
In case of errors on Connection.close() method there is a running threads that
prevents JVM from stopping:
{code:java}
"%172.24.1.210800%TcpClientChannel-timeout-worker" #43 prio=10 os_prio=0
cpu=427.72ms elapsed=4795.29s tid=0x00007f92240011e0 nid=0x6aa3 waiting on
condition [0x00007f92cc9f7000]
java.lang.Thread.State: TIMED_WAITING (sleeping)
at java.lang.Thread.sleep([email protected]/Native Method)
at
org.apache.ignite.internal.future.timeout.TimeoutWorker.body(TimeoutWorker.java:106)
at
org.apache.ignite.internal.util.worker.IgniteWorker.run(IgniteWorker.java:108)
at java.lang.Thread.run([email protected]/Thread.java:840) {code}
Steps on how I get it:
# Start single node cluster
# Run TPC-H bench-base benchmark
Pls, find the stack trace of the benchbase client after getting the exception
in 24498.jstack attachment.
It prevents utilities like benchmark or ETL script from stopping.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
