James Clampffer created HDFS-11027:

             Summary: libhdfs++: Make sure HA retry policy is aware of 
authentication failures
                 Key: HDFS-11027
                 URL: https://issues.apache.org/jira/browse/HDFS-11027
             Project: Hadoop HDFS
          Issue Type: Sub-task
            Reporter: James Clampffer
            Assignee: James Clampffer

"Authentication failed" status falls into the general !status.ok() block in the 
HA retry policy so it will keep attempting to failover.  If the client isn't 
kerberized, or doesn't have the right ticket it should give up and return a 
meaningful error message (right now it returns a generic bad connection failure 

Wouldn't hurt to check the FixedDelayRetryPolicy to make sure that doesn't also 
keep attempting to retry in the same way.  I suspect it does.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: hdfs-dev-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-dev-h...@hadoop.apache.org

Reply via email to