[HBASE-17587] Do not Rethrow DoNotRetryIOException as UnknownScannerException

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.3.0, 2.0.0
• Fix Version/s: 1.4.0, 1.3.1, 1.2.5, 1.1.9, 2.0.0
• Component/s: Coprocessors, regionserver, rpc
• Hadoop Flags: Reviewed
• Assignee: Zach York
• Reporter: Zach York
• Created: 02/Feb/17 20:50
• Updated: 09/Nov/17 19:03
• Resolved: 03/Feb/17 01:12

Description

HBase commit https://github.com/apache/hbase/commit/94ade6a514e9935a8c283befb31f29cd8d3a2045 broke co-processors (such as Phoenix) throwing DoNotRetryIOExceptions when scanning a table. This change rethrows them as UnknownScannerExceptions which the HBase client will retry on. This is unintended behavior since co-processors should be able to throw DoNotRetryIOExceptions back to the client. This came up through a phoenix IT when trying to upgrade to HBase 1.3.0 https://github.com/apache/phoenix/pull/230

Issue Links

• relates to: HBASE-17187 DoNotRetryExceptions from coprocessors should bubble up to the application
• links to: Review Board (branch-1.3)