[ https://issues.apache.org/jira/browse/HDFS-11643?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15965269#comment-15965269 ]
SammiChen commented on HDFS-11643: ---------------------------------- Hi [~andrew.wang], yea, we need face this a "replicated" EC policy now. Let me first clear understand the requirements. So to your knowledge, can a file create API which forces use default replication solve all the issues we met so far and all foreseeable issues? Or should we also provide a way to set a "replicated" EC policy on directory? If only a create API is required, I would suggest not introduce the "replicated" EC policy concept, we can add a Boolean parameter to create function to enforce the file to be replication file. If we also need to provide set an directory back to replication, other than inherit its parent's EC policy, then "replicated" EC policy is a must. > Balancer fencing fails when writing erasure coded lock file > ----------------------------------------------------------- > > Key: HDFS-11643 > URL: https://issues.apache.org/jira/browse/HDFS-11643 > Project: Hadoop HDFS > Issue Type: Bug > Components: balancer & mover, erasure-coding > Affects Versions: 3.0.0-alpha1 > Reporter: Andrew Wang > Priority: Critical > Labels: hdfs-ec-3.0-must-do > > At startup, the balancer writes its hostname to the lock file and calls > hflush(). hflush is not supported for EC files, so this fails when the entire > filesystem is erasure coded. -- This message was sent by Atlassian JIRA (v6.3.15#6346) --------------------------------------------------------------------- To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org