[ https://issues.apache.org/jira/browse/HDFS-12886?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16283101#comment-16283101 ]
Daryn Sharp commented on HDFS-12886: ------------------------------------ Thanks for the ping, will make time to review in the next few days. Just based on cursory read, agree recovery should be able to succeed regardless of min replication. If the number of recovered/available replicas is less than desired then it’s just under replicated. You can’t wait for min replicas when there physically aren’t that many - think rack loss. Waiting only increases odds of data loss. That said, rarely is anything easy to fix as it seems with block management... > Ignore minReplication for block recovery > ---------------------------------------- > > Key: HDFS-12886 > URL: https://issues.apache.org/jira/browse/HDFS-12886 > Project: Hadoop HDFS > Issue Type: Bug > Components: hdfs, namenode > Reporter: Lukas Majercak > Assignee: Lukas Majercak > Attachments: HDFS-12886.001.patch, HDFS-12886.002.patch > > > Ignore minReplication for blocks that went through recovery, and allow NN to > complete them and replicate. -- This message was sent by Atlassian JIRA (v6.4.14#64029) --------------------------------------------------------------------- To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org