[
https://issues.apache.org/jira/browse/HBASE-8402?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13639411#comment-13639411
]
Himanshu Vashishtha commented on HBASE-8402:
--------------------------------------------
ScanMetrics is published only once in the scan's lifetime; so it can be called
in the close method. The current code actually calls the but in case the cache
value is large than the actual result set, the callable is set to null before
calling the close method. This results that writeScanMetrics is never called.
The testScanMetrics also assumed this to be a valid case. I think it should be
corrected. The above patch also changes the testScanMetrics method.
TestClientSide passes with the patch.
> ScanMetrics depends on number of rpc calls to the server.
> ---------------------------------------------------------
>
> Key: HBASE-8402
> URL: https://issues.apache.org/jira/browse/HBASE-8402
> Project: HBase
> Issue Type: Bug
> Components: Client, metrics
> Affects Versions: 0.95.0
> Reporter: Himanshu Vashishtha
> Priority: Minor
> Attachments: HBASE-8402-v1.patch
>
>
> Currently, scan metrics is not published in case there is one trip to server.
> I was testing it on a small row range (200 rows) with a large cache value
> (1000). It doesn't look right as metrics should not depend on number of rpc
> calls (number of rpc call is just one metrics fwiw).
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
