[ https://issues.apache.org/jira/browse/HDFS-6745?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14170203#comment-14170203 ]
Ming Ma commented on HDFS-6745: ------------------------------- At RPC layer, we can add a new method similar to ClientProtocol.listCorruptFileBlocks. Maybe the new method can take replication threshold as a parameter to retrieve all blocks below that threshold. Then ClientProtocol.listCorruptFileBlocks could become a special case of the new method. > Display the list of "very-under-replicated blocks" as well as the files on NN > webUI > ----------------------------------------------------------------------------------- > > Key: HDFS-6745 > URL: https://issues.apache.org/jira/browse/HDFS-6745 > Project: Hadoop HDFS > Issue Type: Sub-task > Reporter: Ming Ma > > Sometimes admins want to know the list of "very-under-replicated blocks" > before major actions such as decommission; as these blocks are more likely to > turn into missing blocks. "very-under-replicated blocks" are those blocks > with live replica count of 1 and replicator factor of >= 3. -- This message was sent by Atlassian JIRA (v6.3.4#6332)