First post on here: We have a situation where we see under replicated blocks ~200K and are slowly getting replicated. We wanted to see if there was a way to speed that process up.
dfs.namenode.replication.max-streams -- 50
dfs.namenode.replication.max-streams-hard-limit -- 100
dfs.namenode.replication.work.multiplier.per.iteration -- 20
Its a 4 datanode cluster that is sitting idle, network is about 15mb/seconds