[HBASE-15815] Region mover script sometimes reports stuck region where only one server was involved

• Type: Bug
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 1.1.2
• Fix Version/s: 1.4.0, 0.98.20
• Hadoop Flags: Reviewed
• Assignee: Ted Yu
• Reporter: Ted Yu
• Created: 11/May/16 20:51
• Updated: 27/May/16 17:44
• Resolved: 16/May/16 16:08

Description

Sometimes we saw the following in output from region mover script:

2016-05-11 01:38:21,187||INFO|3969|140086696048384|MainThread|2016-05-11 01:38:21,186 INFO  [RubyThread-7: /.../current/hbase-client/bin/thread-pool.rb:28-EventThread] zookeeper.ClientCnxn: EventThread shut down
2016-05-11 01:38:21,299||INFO|3969|140086696048384|MainThread|RuntimeError: Region stuck on hbase-5-2.osl,16020,1462930100540,, newserver=hbase-5-2.osl,16020,1462930100540

There was only one server involved.
Since the name of region was not printed, it makes debugging hard to do.