Michael Stack created HBASE-24123: ------------------------------------- Summary: [Flakey Tests] TestThrift?ServerCmdLine Key: HBASE-24123 URL: https://issues.apache.org/jira/browse/HBASE-24123 Project: HBase Issue Type: Bug Reporter: Michael Stack
TestThrift2ServerCmdLine and TestThriftServerCmdLine are flakey. They fail frequently in local runs for myriad reasons: protocol failures, wrong answers on query, connection resets, etc. Each test runs 40 different combinations of config. This is after work done to try and stabilize by fixing port clash issues. Let me list in here the ways in which it fails. I will disable after I get a good list. Will also try and do some more stabilization too in sub-issues. Lets see which prevails; stabilization or disabling. For starters, this morning I got this: {code} ------------------------------------------------------------------------------- Test set: org.apache.hadoop.hbase.thrift2.TestThrift2ServerCmdLine ------------------------------------------------------------------------------- Tests run: 32, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 126.301 s <<< FAILURE! - in org.apache.hadoop.hbase.thrift2.TestThrift2ServerCmdLine org.apache.hadoop.hbase.thrift2.TestThrift2ServerCmdLine.testRunThriftServer[29] Time elapsed: 0.964 s <<< ERROR! java.lang.Exception: org.apache.thrift.transport.TTransportException: java.net.SocketException: Connection reset at org.apache.hadoop.hbase.thrift2.TestThrift2ServerCmdLine.talkToThriftServer(TestThrift2ServerCmdLine.java:92) Caused by: java.net.SocketException: Connection reset at org.apache.hadoop.hbase.thrift2.TestThrift2ServerCmdLine.talkToThriftServer(TestThrift2ServerCmdLine.java:92) {code} -- This message was sent by Atlassian Jira (v8.3.4#803005)