[ 
https://issues.apache.org/jira/browse/HDFS-2430?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13160409#comment-13160409
 ] 

Aaron T. Myers commented on HDFS-2430:
--------------------------------------

Thanks a lot for the review, Eli. Comments below.

{quote}
The api for NNResourcePolicy#areResourcesAvailable is a little strange, if a 
required resource is not available we return false, but if there are not enough 
redundant resources we throw a RTE. Do callers need to distinguish between 
these two cases? If not, we should return false in both cases.
If so, there should be two separate apis. I'd make the method name better 
reflect the purpose in the comment (whether we can log edits), the method is 
generic but the particular policy is edits-specific.
{quote}

It only throws an RTE if the *configured* number of redundant resources is less 
than the configured minimum number of redundant resources. i.e. it throws only 
in the case of misconfiguration. If indeed the number of configured redundant 
resources is greater than the minimum, but the number of available redundant 
resources is below the minimum, the method will return false. Does that seem 
reasonable?

bq. Does the dir checking use canonical paths so we're sure the checks pass in 
case of symlinks?

Yes indeed. {{FSNS#getStorageDirs}} ends up calling 
{{o.a.h.hdfs.server.common.Util#fileAsURI}} which calls 
{{File#getCanonicalFile}}.

I'll upload a patch in a moment which addresses the nits.
                
> The number of failed or low-resource volumes the NN can tolerate should be 
> configurable
> ---------------------------------------------------------------------------------------
>
>                 Key: HDFS-2430
>                 URL: https://issues.apache.org/jira/browse/HDFS-2430
>             Project: Hadoop HDFS
>          Issue Type: New Feature
>          Components: name-node
>    Affects Versions: 0.24.0
>            Reporter: Aaron T. Myers
>            Assignee: Aaron T. Myers
>         Attachments: HDFS-2430.patch, HDFS-2430.patch, HDFS-2430.patch
>
>
> Currently the number of failed or low-resource volumes the NN can tolerate is 
> effectively hard-coded at 1. It would be nice if this were configurable.

--
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

        

Reply via email to