Hi Momoxixi, Sorry for the late response. So you mean, the invocation of "table.checkAndPut" may throw a "RetriesExhaustedException" but actually, the change has been taken effectively in the HBase server side, right?
If so, two comments here: 1) This change adds a "try-catch" embedded in the original "try-catch", the embedded try-catch should be avoided. 2) Besides the "RetriesExhaustedException", are there other similar cases? Can it be generalized to "IOException"? Thanks for your contribution to Kylin! [ Full content available at: https://github.com/apache/kylin/pull/198 ] This message was relayed via gitbox.apache.org for [email protected]
