[HBASE-18035] Meta replica does not give any primaryOperationTimeout to primary meta region

• Type: Bug
• Status: Resolved
• Priority: Critical
• Resolution: Fixed
• Fix Version/s: 1.4.0, 2.0.0
• Hadoop Flags: Reviewed

Release Note

When a client is configured to use meta replica, it sends scan request to all meta replicas almost at the same time. Since meta replica contains stale data, if result from one of replica comes back first, the client may get wrong region locations. To fix this, "hbase.client.meta.replica.scan.timeout" is introduced, a client will always send to primary meta region first, wait the configured timeout for reply. If no result is received, it will send request to replica meta regions. The unit for "hbase.client.meta.replica.scan.timeout" is microsecond, the default value is 1000000 (1 second).

• Assignee: Hua Xiang
• Reporter: Hua Xiang
• Created: 11/May/17 21:24
• Updated: 01/Aug/18 06:20
• Resolved: 21/May/17 13:44

Description

I was working on my unittest and it failed with TableNotFoundException. I debugged a bit and found out that for meta scan, it does not give any primaryOperationTimeout to primary meta region. This will be an issue as the meta replica will contain stale data and it is possible that the meta replica will return back first than primary.

https://github.com/apache/hbase/blob/master/hbase-client/src/main/java/org/apache/hadoop/hbase/client/ConnectionImplementation.java#L823

Issue Links

• Is contained by: HBASE-18223 Track the effort to improve/bug fix read replica feature