[
https://issues.apache.org/jira/browse/HIVE-5519?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13807420#comment-13807420
]
Thejas M Nair commented on HIVE-5519:
-------------------------------------
I think it would be more useful to use the jobid instead of the start time in
the API. One problem with the time is that it is start time of jobtracker
server, which won't change until restart. Also, webhcat will be isolated from
changes to the jobid string format if we use the entire string.
> Use paging mechanism for templeton get requests.
> ------------------------------------------------
>
> Key: HIVE-5519
> URL: https://issues.apache.org/jira/browse/HIVE-5519
> Project: Hive
> Issue Type: Bug
> Components: WebHCat
> Reporter: Hari Sankar Sivarama Subramaniyan
> Assignee: Hari Sankar Sivarama Subramaniyan
> Attachments: HIVE-5519.1.patch.txt, HIVE-5519.2.patch.txt,
> HIVE-5519.3.patch.txt
>
>
> Issuing a command to retrieve the jobs field using
> "https://mwinkledemo.azurehdinsight.net:563/templeton/v1/queue/<job_id>?user.name=admin&fields=*"
> --user u:p
> will result in timeout in windows machine. The issue happens because of the
> amount of data that needs to be fetched. The proposal is to use paging based
> encoding scheme so that we flush the contents regularly and the client does
> not time out.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
