[jira] [Commented] (AMBARI-17280) RU to write out client configs that are dependencies of Hive, ATS, and Oozie during upgrades that change configs

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

    [ https://issues.apache.org/jira/browse/AMBARI-17280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15335029#comment-15335029 ] 

Hadoop QA commented on AMBARI-17280:
------------------------------------

{color:red}-1 overall{color}.  Here are the results of testing the latest attachment 
  http://issues.apache.org/jira/secure/attachment/12811157/AMBARI-17280.trunk.patch
  against trunk revision .

    {color:green}+1 @author{color}.  The patch does not contain any @author tags.

    {color:green}+1 tests included{color}.  The patch appears to include 1 new or modified test files.

    {color:green}+1 javac{color}.  The applied patch does not increase the total number of javac compiler warnings.

    {color:green}+1 release audit{color}.  The applied patch does not increase the total number of release audit warnings.

    {color:red}-1 core tests{color}.  The patch failed these unit tests in ambari-server:

                  org.apache.ambari.server.state.ServiceComponentTest

Test results: https://builds.apache.org/job/Ambari-trunk-test-patch/7389//testReport/
Console output: https://builds.apache.org/job/Ambari-trunk-test-patch/7389//console

This message is automatically generated.

> RU to write out client configs that are dependencies of Hive, ATS, and Oozie during upgrades that change configs
> ----------------------------------------------------------------------------------------------------------------
>
>                 Key: AMBARI-17280
>                 URL: https://issues.apache.org/jira/browse/AMBARI-17280
>             Project: Ambari
>          Issue Type: Bug
>          Components: stacks
>    Affects Versions: 2.2.0
>            Reporter: Ana Gillan
>            Assignee: Alejandro Fernandez
>            Priority: Critical
>             Fix For: 2.4.0
>
>         Attachments: AMBARI-17280.branch-2.4.patch, AMBARI-17280.trunk.patch
>
>
> During RU, HiveServer2 is restarted but the newer tez configs have not yet been saved, which is incorrect because Hive has a dependency on Tez.
> This is important when configs change during a major stack upgrade, e.g., HDP 2.4 -> 2.5. What happens today is,
> * Install packages generates /etc/tez/2.5.0.0-1/0 and copies the configs from /etc/tez/2.4.0.0-1/0/ to the new folder
> * If configs change during RU, then Hive is restarted and the classpath means that it will pick up the older tez configs from the new /etc/tez/2.5.0.0-1/0 folder
> This problem exists for all of these components:
> HiveServer: depends on Tez and MapReduce clients
> ATS: depends on Tez and Spark clients
> Oozie: depends on Tez, Spark, and MapReduce clients
> This problem only exists when configs change (so crossing major stack version) and during RU (because it is allowed to change configs during the middle of restarting services).



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