[
https://issues.apache.org/jira/browse/HIVE-11724?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14905395#comment-14905395
]
Lefty Leverenz commented on HIVE-11724:
---------------------------------------
{quote}
3. I believe templeton.jobs.listorder needs to be documented once the change
goes in. We might need to add a TODOC label.
{quote}
Yes, once you know the Fix Version/s an appropriate TODOC label can be added,
and then it gets removed after templeton.jobs.listorder has been documented in
the wiki. Here's where it belongs:
* [WebHCat Configuration -- Configuration Variables |
https://cwiki.apache.org/confluence/display/Hive/WebHCat+Configure#WebHCatConfigure-ConfigurationVariables]
By the way, I noticed that the config appearing before templeton.jobs.listorder
in the patch has not been documented yet, so I made a list of all the
undocumented WebHCat configs:
* templeton.enable.job.reconnect.default, templeton.hadoop.conf.dir,
templeton.python, templeton.mr.am.memory.mb, templeton.hive.home,
templeton.hcat.home, templeton.sqoop.archive, templeton.sqoop.path,
templeton.sqoop.home, templeton.controller.mr.am.java.opts,
templeton.unit.test.mode, templeton.hive.extra.files
Perhaps some of those don't need documentation (this question has come up
before) but I wanted to put the list here for review. Should I open a separate
jira?
> WebHcat get jobs to order jobs on time order with latest at top
> ---------------------------------------------------------------
>
> Key: HIVE-11724
> URL: https://issues.apache.org/jira/browse/HIVE-11724
> Project: Hive
> Issue Type: Improvement
> Components: WebHCat
> Affects Versions: 0.14.0
> Reporter: Kiran Kumar Kolli
> Assignee: Kiran Kumar Kolli
> Attachments: HIVE-11724.1.patch, HIVE-11724.2.patch,
> HIVE-11724.3.patch, HIVE-11724.4.patch, HIVE-11724.5.patch
>
>
> HIVE-5519 added pagination feature support to WebHcat. This implementation
> returns the jobs lexicographically resulting in older jobs showing at the
> top.
> Improvement is to order them on time with latest at top. Typically latest
> jobs (or running) ones are more relevant to the user. Time based ordering
> with pagination makes more sense.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
