[ https://issues.apache.org/jira/browse/HDFS-7923?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14588763#comment-14588763 ]
Andrew Wang commented on HDFS-7923: ----------------------------------- If we're really worried about starvation, we could add a failsafe to the DN side that sets a lease ID of 0 after say, three FBR intervals. This will skip the rate limiting. Overall though I agree with Colin, we get starvation when a NN remains continually fully loaded with FBR work from other nodes. Such a cluster would be basically unusable for real work. The default # concurrent FBRs is also >1, so we should be gated on NN CPU rather than the much slower heartbeat interval. > The DataNodes should rate-limit their full block reports by asking the NN on > heartbeat messages > ----------------------------------------------------------------------------------------------- > > Key: HDFS-7923 > URL: https://issues.apache.org/jira/browse/HDFS-7923 > Project: Hadoop HDFS > Issue Type: Sub-task > Affects Versions: 2.8.0 > Reporter: Colin Patrick McCabe > Assignee: Colin Patrick McCabe > Fix For: 2.8.0 > > Attachments: HDFS-7923.000.patch, HDFS-7923.001.patch, > HDFS-7923.002.patch, HDFS-7923.003.patch, HDFS-7923.004.patch, > HDFS-7923.006.patch, HDFS-7923.007.patch > > > The DataNodes should rate-limit their full block reports. They can do this > by first sending a heartbeat message to the NN with an optional boolean set > which requests permission to send a full block report. If the NN responds > with another optional boolean set, the DN will send an FBR... if not, it will > wait until later. This can be done compatibly with optional fields. -- This message was sent by Atlassian JIRA (v6.3.4#6332)