[
https://issues.apache.org/jira/browse/HDFS-8095?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15816849#comment-15816849
]
Kai Zheng commented on HDFS-8095:
---------------------------------
bq. When you say data corruption, do you mean data loss if the cluster has
insufficient # datanodes or racks?
What I meant is, changing of the system default policy may impact big. But if
we can ensure the policy ID is always associated with the file and persisted in
meta, it should be ok even after the system default policy has changed.
Your understanding reminded me that the user specified policy may be not
validated for the use, as you said, there may be no enough datanodes or racks.
So I thought it's a good idea we should add some necessary check in NN side
when user specifying a policy for a file/folder, or whenever a policy is newly
referenced by client requests. We need a separate issue to handle this guess?
> Allow to configure the system default EC schema
> -----------------------------------------------
>
> Key: HDFS-8095
> URL: https://issues.apache.org/jira/browse/HDFS-8095
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Kai Zheng
> Assignee: Kai Zheng
> Labels: hdfs-ec-3.0-nice-to-have
>
> As suggested by [~umamaheswararao] and [~vinayrpet] in HDFS-8074, we may
> desire allowing to configure the system default EC schema, so in any
> deployment a cluster admin may be able to define their own system default
> one. In the discussion, we have two approaches to configure the system
> default schema: 1) predefine it in the {{ecschema-def.xml}} file, making sure
> it's not changed; 2) configure the key parameter values as properties in
> {{core-site.xml}}. Open this for future consideration in case it's forgotten.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
