[
https://issues.apache.org/jira/browse/AMBARI-12837?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Vijay Srinivasaraghavan updated AMBARI-12837:
---------------------------------------------
Attachment: AMBARI-12837.patch
> Server component should read HCFS type info from stack and propagate it in
> dictionary for agent and UI to consume
> -----------------------------------------------------------------------------------------------------------------
>
> Key: AMBARI-12837
> URL: https://issues.apache.org/jira/browse/AMBARI-12837
> Project: Ambari
> Issue Type: Task
> Affects Versions: trunk
> Reporter: Vijay Srinivasaraghavan
> Assignee: Vijay Srinivasaraghavan
> Attachments: AMBARI-12837.patch
>
>
> UI and agent code is using namenode configuration as a validation to check if
> HDFS is selected or not and perform some HDFS specific operations like
> creating /tmp directory, reading hadoop-env configurations like
> hdfs_log_dir_prefix, dropping fast-hdfs-resource.jar etc., because of which
> the configurations defined for HCFS types are not getting populated. The
> ambari server component should tarverse thhrough the stack configuration and
> populate (hadoop-env or cluster-env) dictionary if HCFS file system is
> asscoated as part of the stack. The dictionary configuration should be used
> in UI and agent code to allow FS operations require to bootstrap the cluster
> deployment.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
