[
https://issues.apache.org/jira/browse/RANGER-2639?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Pradeep Agrawal resolved RANGER-2639.
-------------------------------------
Resolution: Fixed
Patch is already committed :
[https://github.com/apache/ranger/commit/446fb478ee49779271557cca1076db2a777bfaa8]
> XUserService has duplicate db query in when populateViewBean for XXUser
> ------------------------------------------------------------------------
>
> Key: RANGER-2639
> URL: https://issues.apache.org/jira/browse/RANGER-2639
> Project: Ranger
> Issue Type: Improvement
> Components: admin
> Affects Versions: 0.6.0, 1.0.0, 0.7.1, 1.1.0, 1.2.0
> Environment: hdp 2.2
> Reporter: zhangxiong
> Assignee: zhangxiong
> Priority: Minor
> Labels: optimization, performance, ranger, security
> Attachments: XUserService.patch
>
> Original Estimate: 1h
> Remaining Estimate: 1h
>
> when request http api /service/xusers/users with GET method.ranger admin will
> query db and get XXUser,then transform the XXUser to VXUser. in the
> tranformation process,
> XUserService.populateUserAttributes() method is called two times.and make
> extra 2 times db query.when the result users are many,this causes significant
> performance issue,rougly 40% performance down。so indeed we need to call
> populateUserAttributes() once .
> we found this issue in our production environment with 1000+ users in ranger
> admin,and start the kafka or other component which enabled ranger
> authority.in the start process, setup_ranger_xxxx() is called and make http
> /service/xusers/users request to ranger.as it query all users in
> ranger,ranger admin will reponse in several ten seconds.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
