jianghuazhu commented on pull request #3544: URL: https://github.com/apache/hadoop/pull/3544#issuecomment-945826731
Thanks @jojochuang for the comment. Thank you for your reminder. Sorry, I overlooked this point. I also found this phenomenon when I tested the online cluster. Before the test, there are 3 DN nodes in the online cluster, for example: ' hadoop1.xxxx.org:50010 hadoop2..xxxxorg:50010 hadoop3..xxxxorg:50010 ' The Xferport of these 3 DN nodes are the same. When I try to connect to the online namenode, 3 new DataNode services will be created locally, for example: ’ xxxx.xxxx.xxxx.xxxx:1 xxxx.xxxx.xxxx.xxxx:2 xxxx.xxxx.xxxx.xxxx:3 ‘ When creating a file, you need to get the location of the datanode from the remote namenode through addBlock(). At this time, you may get hadoop1.xxxx.org:50010, hadoop2..xxxxorg:50010, hadoop3..xxxxorg:50010. For the safety of the entire link, I think these DNs should be filtered out when executing addBlock(). -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
