[
https://issues.apache.org/jira/browse/HDFS-4947?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13725819#comment-13725819
]
Colin Patrick McCabe commented on HDFS-4947:
--------------------------------------------
bq. Even a client has multiple IPs, my understanding is that only one IP is
used by the client for the NFS connection. Please let us know if it answers
your question.
Another question I have is whether there may be multiple hostnames associated
with the client's IP address. I'm pretty sure this is allowable, so we may
have to deal with that case here.
bq. Since reverse lookup is expensive to do for each request, it can be helpful
to have a lookup cache and update individual cache entry periodically(e.g.,
every 15 minutes).
Yeah, I agree that a lookup cache is a good idea.
> Add NFS server export table to control export by hostname or IP range
> ---------------------------------------------------------------------
>
> Key: HDFS-4947
> URL: https://issues.apache.org/jira/browse/HDFS-4947
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: nfs
> Affects Versions: 3.0.0
> Reporter: Brandon Li
> Assignee: Jing Zhao
> Attachments: HDFS-4947.000.patch, HDFS-4947.001.patch,
> HDFS-4947.002.patch, HDFS-4947.003.patch
>
>
> This jira adds NFS server export table (hostname or IP range) to control
> export. Specifically, it
> 1. adds two new classes AccessPrivilege and ClientHostsMatcher for loading
> and checking the mapping between client hosts and their access privileges.
> These two classes are directly copied from Brock Noland's NFS4 implementation.
> 2. adds the access check (based on the export) to NFS3 procedures.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
