[
https://issues.apache.org/jira/browse/HBASE-11449?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14050884#comment-14050884
]
Andrew Purtell commented on HBASE-11449:
----------------------------------------
Patch looks good.
Can we leave this part out?
{code}
@@ -107,7 +108,9 @@ public class MultiThreadedReaderWithACL extends
MultiThreadedReader {
result = localTable.get(get);
}
boolean isNullExpected = ((((int) keyToRead %
specialPermCellInsertionFactor)) == 0);
- LOG.info("Read happening from ACL " + isNullExpected);
+ if (isNullExpected) {
+ LOG.info("Read happening from ACL " + isNullExpected);
+ }
long end = System.nanoTime();
verifyResultsAndUpdateMetrics(verify, get, end - start, result,
localTable, isNullExpected);
} catch (IOException e) {
{code}
That INFO level logging will be messy. Let's make it TRACE level or remove it.
> IntegrationTestIngestWithACL fails to use different users after HBASE-10810
> ---------------------------------------------------------------------------
>
> Key: HBASE-11449
> URL: https://issues.apache.org/jira/browse/HBASE-11449
> Project: HBase
> Issue Type: Bug
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.99.0
>
> Attachments: HBASE-11449.patch, HBASE-11449_1.patch
>
>
> Im MultiThreadedReaderWithACL we needed the Table instances per user and the
> same was getting cached. Now after HBASE-10810 the table instance is
> obtained from the connection maintained which is the connection created by
> the 'root' user and not by the user intended. So IT does not work in trunk.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
