[jira] [Commented] (SENTRY-699) Memory leak when running Sentry w/ HiveServer2

["Hadoop QA (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SENTRY-699?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14501031#comment-14501031 ] 

Hadoop QA commented on SENTRY-699:
----------------------------------

Here are the results of testing the latest attachment
https://issues.apache.org/jira/secure/attachment/12726323/SENTRY-699.1.patch against master.

{color:green}Overall:{color} +1 all checks pass

{color:green}SUCCESS:{color} all tests passed

Console output: https://builds.apache.org/job/PreCommit-SENTRY-Build/552/console

This message is automatically generated.

> Memory leak when running Sentry w/ HiveServer2
> ----------------------------------------------
>
>                 Key: SENTRY-699
>                 URL: https://issues.apache.org/jira/browse/SENTRY-699
>             Project: Sentry
>          Issue Type: Bug
>    Affects Versions: 1.5.0
>            Reporter: Lenni Kuff
>            Assignee: Prasad Mujumdar
>         Attachments: SENTRY-699.1.patch
>
>
> Each time a query executes we create a new instance of HiveAuthzConf which in turn creates a ResourceAuthorizationProvider. Within ResourceAuthorizationProvider there is a thread local variable: "ThreadLocal<List<String>> lastFailedPrivileges" which is not getting garbage collected since the parent thread is still alive. Over time, this leads to a memory leak in HS2. We need to be sure that any ThreadLocal state is cleaned up when we are done using it (or consider avoiding the use of ThreadLocal if possible).



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)