HDFS Balancer (2): Configurations & CLI Options:  https://community.cloudera.com/t5/Community-Articles/HDFS-Balancer-2-Configurations-CLI-Options/ta-p/246687 ... View more

Thanks a lot! Ming and I have tried it. It works well. ... View more

For recoverLease using CLI, see https://community.hortonworks.com/questions/146012/force-closing-a-hdfs-file-still-open-because-uncor.html?childToView=146021#answer-146021 ... View more

clear /etc/resolv.conf I think the problem is resolved dns。 @Silvio del Val , clear /etc/resolv.conf I think the problem is resolved dns。 @Silvio del Val ... View more

Hello Team, I already have balanced cluster where all datanodes have same space utilization. 6*4TB storage across all datanodes and all are 80% utilized. Since we do not have extra node to add for now we are adding extra strorage of 6*4TB across all datanodes. My question is: 1. Do i need to re-balance cluster? 2. If yes Do i still need to run balancer with default threshold of 10% or any else? - Vijay Mishra ... View more

Thanks for the the very useful article. Will there be any follow ups coming? I am particularly interested in the changes that came about from https://issues.apache.org/jira/browse/HDFS-8818 and how settings like dfs.balancer.moverThreads needs to be increased from default when balancing a large number of unbalanced nodes. (e.g the setting used in the comment in HDFS-8188 herehttps://issues.apache.org/jira/browse/HDFS-8818?focusedCommentId=15997429&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15997429) ... View more