I have 4 nodes cluster with NN-HA ..when i restart two datanodes i have 100+ corrupted blocks(shown in ambari>hdfs>servieces ) and i have replication factor 3 so i want to deleted these corrupted blocks..before i deleted this i want to check which are files are corrupted so i run below command and got below ouput ,here did not get any corrupted blocks or files? please help me ..i have this issue since long ..
P1-230-8:/root> hdfs fsck / | egrep -v '^\.+
Connecting to namenode via http://stlpr8711:50070 FSCK started by hdfs (auth:SIMPLE) from /126.96.36.199 for path / at Fri Aug 05 06:27:52 CDT 2016
...........................Status: HEALTHY T
otal size: 39189208876540 B (Total open files size: 83162 B)
Total dirs: 710383 Total files: 2184527
Total symlinks: 0 (Files currently being written: 7) Total blocks (validated): 2309173 (avg. block size 16971101 B) (Total open file blocks (not validated): 7)
Corrupt blocks: 0
Number of data-nodes: 4
Number of racks: 1
FSCK ended at Fri Aug 05 06:28:40 CDT 2016 in 47634 milliseconds The filesystem under path '/' is HEALTHY
I would highly suggest you to upgrade your ambari server with latest release. As a workaround restart the Ambari (server and agents) as well as Ambari Metrics and see if that resolve the issue.