You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ambari.apache.org by "Sumit Mohanty (JIRA)" <ji...@apache.org> on 2016/03/10 03:19:40 UTC

[jira] [Updated] (AMBARI-15364) Ability to trigger NameNode checkpoint from Ambari Web

     [ https://issues.apache.org/jira/browse/AMBARI-15364?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sumit Mohanty updated AMBARI-15364:
-----------------------------------
    Attachment: Screen Shot 2016-03-09 at 6.18.28 PM.png

> Ability to trigger NameNode checkpoint from Ambari Web
> ------------------------------------------------------
>
>                 Key: AMBARI-15364
>                 URL: https://issues.apache.org/jira/browse/AMBARI-15364
>             Project: Ambari
>          Issue Type: Bug
>    Affects Versions: 2.2.0
>            Reporter: Sumit Mohanty
>             Fix For: 2.4.0
>
>         Attachments: Screen Shot 2016-03-09 at 6.18.28 PM.png
>
>
> Expose the checkpoint action in the UI and make it work E2E (and ability to check on progress):
> Services > HDFS > Service Actions > Save Namespace
> Ample documentation is available that describe checkpoint process for HDFS. Roughly, this translate to
> {code}
> >> hdfs dfsadmin -safemode enter
> Safe mode is ON
> >> hdfs dfsadmin -saveNamespace
> Save namespace successful
> >> hdfs dfsadmin -safemode leave
> Safe mode is OFF
> {code}
> One approach is to have custom commands for each operation. In general, having support to get into and out of SAFEMODE is useful.
> saveNamespace may be a long running command. Therefore custom command for this operation may be long running and thus there is a chance of timeout or may even block execution of other commands on the NN host. So the better approach would be to design a custom command that can start "saveNamespace", check the status, and if needed kill the command.
>  



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