[jira] [Updated] (RANGER-3248) In Ranger Audit collection number of shards should be depending on the number of infra-solr nodes

["Dhaval Shah (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/RANGER-3248?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Dhaval Shah updated RANGER-3248:
--------------------------------
    Attachment: RANGER-3248.patch

> In Ranger Audit collection number of shards should be depending on the number of infra-solr nodes
> -------------------------------------------------------------------------------------------------
>
>                 Key: RANGER-3248
>                 URL: https://issues.apache.org/jira/browse/RANGER-3248
>             Project: Ranger
>          Issue Type: Improvement
>          Components: Ranger
>            Reporter: Dhaval Shah
>            Assignee: Dhaval Shah
>            Priority: Major
>         Attachments: RANGER-3248.patch
>
>
> In Ranger while creating collection ranger_audits in solr the [default no. of shards|https://github.com/apache/ranger/blob/master/embeddedwebserver/src/main/java/org/apache/ranger/server/tomcat/SolrCollectionBootstrapper.java#L134] we configured is 1. It is configurable under ranger-admin-site.xml having porperty name "ranger.audit.solr.no.shards" .
> The number of shards initially should be depending on the number of solr live nodes. This can be achieve using SolrCloudClient object in SolrCollectionBootstrapper.java during the collection creation.
> {code:java}
> SolrCloudClient.getClusterStateProvider().getLiveNodes().size() {code}
> Shards value will be visible as INFO in file catalina.out under ranger admin log dir for verification.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)