You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@oozie.apache.org by "Mohammad Kamrul Islam (JIRA)" <ji...@apache.org> on 2011/08/26 10:11:29 UTC
[jira] [Updated] (OOZIE-29) Support hadoop job log retrieval
through Oozie WS API
[ https://issues.apache.org/jira/browse/OOZIE-29?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Mohammad Kamrul Islam updated OOZIE-29:
---------------------------------------
Description:
Users want to utilize Oozie as only entry/gateway to the Grid system. Currently they could get the Oozie log related to their jobs. They could also access to the corresponding hadoop job details through Oozie UI that redirect to the JT UI of that job.
However, users strongly prefer to get the job log as Oozie WS API response or through Oozie CLI for the debugging purpose.
This JIRA is to discuss the different options (if any) that Hadoop currently supports. And subsequent Oozie's action plan.
Questions are:
1. Does hadoop provide any API to retrieve the job log?
2. Does hadoop provide any API to get the path(s) of hadoop job logs?
3. If the answer is no for the above questions:
a. Is there any alternatives?
b. Should we create a Hadoop JIRA for asking to support either of the options?
At the same time, we have to make sure the followings:
1. Oozie will work as a pass-thru to hadoop.
2. Oozie will not store any hadoop log.
3. If hadoop restart loss the job log, oozie will return the same message that hadoop replies back
4. Oozie could retrieve the hadoop job log started by pig.
Please comment.
was:
Users want to use Oozie as only entry/gateway to the Grid system. Currently they could get the Oozie log related to their jobs. They could also access to the corresponding hadoop job details through Oozie UI that redirect to the JT UI of that job.
However, users strongly prefer to get the job log as Oozie WS API response or through Oozie CLI.
This JIRA is to discuss the different options (if any) that Hadoop currently supports. And subsequent Oozie's action plan.
Questions are:
1. Does hadoop provide any API to retrieve the job log?
2. Does hadoop provide any API to get the path(s) of hadoop job logs.
3. If the answer is no for the above questions:
a. Is there any alternatives?
b. Should we create a Hadoop JIRA for asking to support either of the options?
Please comment.
Summary: Support hadoop job log retrieval through Oozie WS API (was: Support hadoop job log log retrieval through Oozie WS API)
> Support hadoop job log retrieval through Oozie WS API
> -----------------------------------------------------
>
> Key: OOZIE-29
> URL: https://issues.apache.org/jira/browse/OOZIE-29
> Project: Apache Oozie (Incubating)
> Issue Type: New Feature
> Reporter: Mohammad Kamrul Islam
>
> Users want to utilize Oozie as only entry/gateway to the Grid system. Currently they could get the Oozie log related to their jobs. They could also access to the corresponding hadoop job details through Oozie UI that redirect to the JT UI of that job.
> However, users strongly prefer to get the job log as Oozie WS API response or through Oozie CLI for the debugging purpose.
> This JIRA is to discuss the different options (if any) that Hadoop currently supports. And subsequent Oozie's action plan.
>
> Questions are:
> 1. Does hadoop provide any API to retrieve the job log?
> 2. Does hadoop provide any API to get the path(s) of hadoop job logs?
> 3. If the answer is no for the above questions:
> a. Is there any alternatives?
> b. Should we create a Hadoop JIRA for asking to support either of the options?
> At the same time, we have to make sure the followings:
> 1. Oozie will work as a pass-thru to hadoop.
> 2. Oozie will not store any hadoop log.
> 3. If hadoop restart loss the job log, oozie will return the same message that hadoop replies back
> 4. Oozie could retrieve the hadoop job log started by pig.
> Please comment.
>
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira