[HADOOP-743] org.apache.hadoop.ipc.Server join method doees not wait for its subthreads to die

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 0.6.1
• Component/s: ipc
• Assignee: Owen O'Malley
• Reporter: Thomas Friol
• Created: 22/Nov/06 10:11
• Updated: 15/Dec/06 21:55

Description

Is there an explanation about this ?

In fact, my problem is that I have to launch consecutively two MiniMRCluster in order to perform unit tests as following :

• Start first MiniMrCLuster
• Do some tests
• Stop first MiniMRCluster
• Start second MiniMRCluster
• Do some tests
• Stop second MiniMRCluster

When executing this, the second MiniMRcluster give me continually this log :
06/11/22 11:09:22 INFO ipc.RPC: Problem connecting to server: localhost/127.0.0.1:60030
Waiting for task tracker to start.

The only solution I have found now is to insert a Thread.sleep(2000) between first MiniMRCluster stop and the second one start to ensure that the first one is completely stopped.

Any idea, suggestions, thoughts ?