[HBASE-3894] Thread contention over row locks set monitor

• Type: Bug
• Status: Closed
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 0.90.2
• Fix Version/s: 0.90.4
• Hadoop Flags: Reviewed
• Assignee: Dave Latham
• Reporter: Dave Latham
• Created: 17/May/11 18:55
• Updated: 20/Nov/15 12:42
• Resolved: 13/Jun/11 23:51

Description

HRegion maintains a set of row locks. Whenever any thread attempts to lock or release a row it needs to acquire the monitor on that set. We've been encountering cases with 30 handler threads all contending for that monitor, blocked progress on the region server. Clients timeout, and retry making it worse, and the region server stops responding to new clients almost entirely.

Issue Links

• is depended upon by: HBASE-3893 HRegion.internalObtainRowLock shouldn't wait forever
• is related to: HBASE-2295 Row locks may deadlock with themselves
• is related to: HBASE-3893 HRegion.internalObtainRowLock shouldn't wait forever