CM users,
I have a very small cluster with dedicated master for namenode and job
tracker. recently i noticed that my master and 1 slave in the cluster have
their disk full. when i checked on the master, the following dir has taken
up pretty much all the space
/dfs/dn/current/BP-1602286141-10.81.81.208-1361056237771/current/finalized
There are many subdirs[1-53] under the above dir.
Please let me know how can i safely clean up these dirs
regards
#sudhir
