[
https://issues.apache.org/jira/browse/MAPREDUCE-3940?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Vinod Kumar Vavilapalli updated MAPREDUCE-3940:
-----------------------------------------------
Attachment: MAPREDUCE-3940-20120425.txt
bq. The MR AM will end up seing some harmless exceptions when invoking
stopContainer
Ths issue of using ContainerTokens with stopContainer() and
getContainerStatus() is known, but I did miss the MR AM getting exceptions. We
will need to fix circumventing the verification logic of ContainerTokens for
the stopContainer() and getContainerStatus() APIs, will open a separate ticket.
Short term, we can get this into trunk and merge it into 23 branches after the
followup ticket is fixed.
Updating the patch to be applicable against later trunk.
> ContainerTokens should have an expiry interval
> ----------------------------------------------
>
> Key: MAPREDUCE-3940
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-3940
> Project: Hadoop Map/Reduce
> Issue Type: Sub-task
> Components: mrv2, security
> Affects Versions: 0.23.0
> Reporter: Vinod Kumar Vavilapalli
> Assignee: Vinod Kumar Vavilapalli
> Fix For: 0.23.2
>
> Attachments: MAPREDUCE-3940-20120308.txt,
> MAPREDUCE-3940-20120416.txt, MAPREDUCE-3940-20120425.txt
>
>
> - RM should generate the expiry time for a container
> - A ContainerToken should have its expire time encoded
> - NMs should reject containers with expired tokens.
> - Expiry interval for a ContainerToken is same as the expiry interval for a
> container.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
