Ravuri Sushma sree created HDFS-14765:
-----------------------------------------
Summary: fsck and dfsadmin -report gives different results at
UnderReplicated Blocks after decomissioning
Key: HDFS-14765
URL: https://issues.apache.org/jira/browse/HDFS-14765
Project: Hadoop HDFS
Issue Type: Bug
Affects Versions: 3.1.2
Environment: Working on a cluster with 2 Namenodes and 6 Datanodes
with 6,00,000 files in it
One of the datanode is decommissioned
After the node is decommissioned, fsck , dfsadmin/WebUi shows the following
result :
*FSCK* :
Status: HEALTHY
Number of data-nodes: 6
Number of racks: 1
Total dirs: 83
Total symlinks: 0
Replicated Blocks:
Total size: 7431010 B
Total files: 600000
Total blocks (validated): 581839 (avg. block size 12 B)
Minimally replicated blocks: 581839 (100.0 %)
Over-replicated blocks: 0 (0.0 %)
{color:#de350b}Under-replicated blocks: 0 (0.0 %){color}
Mis-replicated blocks: 0 (0.0 %)
Default replication factor: 2
Average block replication: 3.0021775
Missing blocks: 0
Corrupt blocks: 0
Missing replicas: 0 (0.0 %)
DecommissionedReplicas: 255396
*Dfsadmin -report* :
Configured Capacity: 792665681920 (738.23 GB)
Present Capacity: 447796567710 (417.04 GB)
DFS Remaining: 435391111168 (405.49 GB)
DFS Used: 12405456542 (11.55 GB)
DFS Used%: 2.77%
Replicated Blocks:
{color:#de350b}Under replicated blocks: 252067{color}
Blocks with corrupt replicas: 0
Missing blocks: 0
Missing blocks (with replication factor 1): 0
Pending deletion blocks: 0
Reporter: Ravuri Sushma sree
Assignee: Ravuri Sushma sree
Fix For: 3.3.0
Fsck and dfsadmin show different results of underreplicated blocks after one of
the datanodes is decommmissioned
--
This message was sent by Atlassian Jira
(v8.3.2#803003)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-dev-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-dev-h...@hadoop.apache.org
