[
https://issues.apache.org/jira/browse/HBASE-20375?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16431504#comment-16431504
]
Wei-Chiu Chuang commented on HBASE-20375:
-----------------------------------------
The change seems good to me. I'm curious about the next debug message:
{code:java}
logDebug("appliedCredentials:" + appliedCredentials + ",credentials:" +
credentials)
{code}
IIUC, it would call Credentials.toString() to convert the object into string,
regardless of log level. That doesn't sound like a great approach from a
performance perspective.
> Remove use of getCurrentUserCredentials in hbase-spark module
> -------------------------------------------------------------
>
> Key: HBASE-20375
> URL: https://issues.apache.org/jira/browse/HBASE-20375
> Project: HBase
> Issue Type: Bug
> Components: spark
> Reporter: Ted Yu
> Assignee: Ted Yu
> Priority: Major
> Attachments: 20375.v1.txt
>
>
> When compiling hbase-spark module against Spark 2.3.0 release, we would get:
> {code}
> [ERROR]
> /a/hbase/hbase-spark/src/main/scala/org/apache/hadoop/hbase/spark/HBaseContext.scala:68:
> error: value getCurrentUserCredentials is not a member of
> org.apache.spark.deploy. SparkHadoopUtil
> [ERROR] @transient var credentials =
> SparkHadoopUtil.get.getCurrentUserCredentials()
> [ERROR] ^
> [ERROR]
> /a/hbase/hbase-spark/src/main/scala/org/apache/hadoop/hbase/spark/HBaseContext.scala:236:
> error: value getCurrentUserCredentials is not a member of
> org.apache.spark.deploy. SparkHadoopUtil
> [ERROR] credentials = SparkHadoopUtil.get.getCurrentUserCredentials()
> [ERROR] ^
> [ERROR] two errors found
> {code}
> {{getCurrentUserCredentials}} was removed by SPARK-22372.
> This issue is to replace the call to {{getCurrentUserCredentials}} with call
> to {{UserGroupInformation.getCurrentUser().getCredentials()}}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
