[HBASE-5974] Scanner retry behavior with RPC timeout on next() seems incorrect

• Type: Bug
• Status: Closed
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: 0.90.7, 0.92.1, 0.94.0, 0.95.2
• Fix Version/s: 0.95.0
• Component/s: Client, regionserver
• Hadoop Flags: Reviewed
• Release Note: Adds a sequence id that is incremented on each Scanner#next invocation so we can distingush calls that arrive out of order. Throws OutOfOrderScannerNextException when we encounter the latter case.
• Assignee: Anoop Sam John
• Reporter: Todd Lipcon
• Created: 09/May/12 20:25
• Updated: 19/Sep/14 20:12
• Resolved: 25/Oct/12 16:45

Description

I'm seeing the following behavior:

• set RPC timeout to a short value
• call next() for some batch of rows, big enough so the client times out before the result is returned
• the HConnectionManager stuff will retry the next() call to the same server. At this point, one of two things can happen: 1) the previous next() call will still be processing, in which case you get a LeaseException, because it was removed from the map during the processing, or 2) the next() call will succeed but skip the prior batch of rows.