Duo Zhang created HBASE-22316:
---------------------------------
Summary: Consider always creating a new exception in
FutureUtils.get
Key: HBASE-22316
URL: https://issues.apache.org/jira/browse/HBASE-22316
Project: HBase
Issue Type: Sub-task
Reporter: Duo Zhang
This is for debugging. As in async client, the retry will be done in the retry
timer thread, so the exception we get from the CompletableFuture will have a
stack trace starting from the root of the retry timer. If we just throw this
exception out when calling future.get(by unwrapping the ExecutionException),
the upper layer even can not know where is the exception thrown...
This happens for me many times, so I propose that we always create a new
exception in FutureUtils.get, so at least we can record the stack trace for the
method calling FutureUtils.get...
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
