[ 
https://issues.apache.org/jira/browse/IMPALA-9086?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17818987#comment-17818987
 ] 

ASF subversion and git services commented on IMPALA-9086:
---------------------------------------------------------

Commit 9dc146de7752bb4f684203d34233deeb215f2bc6 in impala's branch 
refs/heads/master from Saurabh Katiyal
[ https://gitbox.apache.org/repos/asf?p=impala.git;h=9dc146de7 ]

IMPALA-9086: Show Hive configurations in /hadoop-varz page

It's a two part improvement:
1. Updated /hadoop-varz static class implementation for coordinator webUI
   from: org.apache.hadoop.conf.Configuration
   to  : org.apache.hadoop.hive.conf.HiveConf (Inherits Configuration)

2. Added similar implementation of /hadoop-varz for catalog webUI

Now /hadoop-varz page has lot more information about HMS client
and configuration including warehouse.dir

Change-Id: I5af0eb68e71afeed64660d4d40584208ea503217
Reviewed-on: http://gerrit.cloudera.org:8080/20682
Reviewed-by: Impala Public Jenkins <impala-public-jenk...@cloudera.com>
Tested-by: Impala Public Jenkins <impala-public-jenk...@cloudera.com>


> Show Hive configurations in /hadoop-varz page
> ---------------------------------------------
>
>                 Key: IMPALA-9086
>                 URL: https://issues.apache.org/jira/browse/IMPALA-9086
>             Project: IMPALA
>          Issue Type: Improvement
>          Components: Frontend
>            Reporter: Quanlong Huang
>            Assignee: Saurabh Katiyal
>            Priority: Major
>              Labels: ramp-up, supportability
>
> Hive configurations are not shown in /hadoop-varz page. It helps to check 
> hive.metastore.warehouse.dir and hive.metastore.warehouse.external.dir used 
> in FE. We should also add the /hadoop-varz page in catalogd WebUI.
> The URL callback is registered in be/src/service/impala-http-handler.cc:
> {code:cpp}
>   webserver->RegisterUrlCallback("/hadoop-varz", "hadoop-varz.tmpl",
>       MakeCallback(this, &ImpalaHttpHandler::HadoopVarzHandler), true);
> {code}
> https://github.com/apache/impala/blob/379038f7639731605bca4356337616fa69f35f9d/be/src/service/impala-http-handler.cc#L133-L134
> The handler, ImpalaHttpHandler::HadoopVarzHandler(), is implemented by 
> calling a JNI method to invoke FE method JniFrontend.getAllHadoopConfigs():
> https://github.com/apache/impala/blob/379038f7639731605bca4356337616fa69f35f9d/fe/src/main/java/org/apache/impala/service/JniFrontend.java#L626-L631
> We simply create a static object of org.apache.hadoop.conf.Configuration and 
> use it for each /hadoop-varz http request. We can consier changing its type 
> to org.apache.hadoop.hive.conf.HiveConf (inherrits Configuration) which might 
> contain all the hive configurations.
> To add the same page for catalogd, we need to register a similar handler in 
> CatalogServer::RegisterWebpages()
> https://github.com/apache/impala/blob/379038f7639731605bca4356337616fa69f35f9d/be/src/catalog/catalog-server.cc#L434
> Also add a new method in JniCatalog.java to return the configurations 
> throught JNI, add JNI stuffs in catalog.cc and catalog.h (similar to existing 
> JNI methods, e.g. get_catalog_object_id_).



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-all-unsubscr...@impala.apache.org
For additional commands, e-mail: issues-all-h...@impala.apache.org

Reply via email to