hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Eli Collins (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HADOOP-8689) Make trash a server side configuration option
Date Fri, 17 Aug 2012 19:19:38 GMT

     [ https://issues.apache.org/jira/browse/HADOOP-8689?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Eli Collins updated HADOOP-8689:
--------------------------------

    Attachment: hadoop-8689.txt

Thanks for the review ATM. Updated patch attached.

#1 I moved the initialization of the checkpoint interval to trash policy creation (where the
deletion interval is set) rather than when an emptier object is created so the emptier uses
the interval configured by the trash policy that creates the emptier. I think this is more
clear, and in practice doesn't change things since the just NN creates an emptier once. As
an aside, the emptier should probably only ever be run from the server side (that's mostly
true currently) therefore the checkpoint interval is really server side as well (unlike the
trash interval shell clients don't run emptiers). I filed HADOOP-8710 to remove the one place
where a user might be able to run an emptier (w/o explicitly creating objects themselves).

#2 Fixed, the content of that description was actually incorrect so I update it.

#3 Good suggestion done.

I noticed the log for setting the checkpoint interval if its not set is a warn, made that
an info, also it prints values in ms but the message says "seconds" so I fixed that bug as
well.
                
> Make trash a server side configuration option
> ---------------------------------------------
>
>                 Key: HADOOP-8689
>                 URL: https://issues.apache.org/jira/browse/HADOOP-8689
>             Project: Hadoop Common
>          Issue Type: Improvement
>          Components: fs
>    Affects Versions: 2.0.0-alpha
>            Reporter: Eli Collins
>            Assignee: Eli Collins
>         Attachments: hadoop-8689.txt, hadoop-8689.txt
>
>
> Per ATM's suggestion in HADOOP-8598 for v2 let's make {{fs.trash.interval}} configured
server side. If it is not configured server side then the client side configuration is used.
The {{fs.trash.checkpoint.interval}} option is already server side as the emptier runs in
the NameNode. Clients may manually run an emptier via hadoop org.apache.hadoop.fs.Trash but
it's OK if it uses a separate interval. 

--
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

        

Mime
View raw message