I have 3 node instances -
master, slave1 and slave2
SSHing between these nodes works fine.
Here is the processes that starts on each node when I say ./sbin/start-dfs.sh
master:
SecondaryNameNode
Jps
NameNode
DataNode
slave1:
Jps
DataNode
slave2:
Jps
DataNode
But when I try to access hdfs from slave nodes, I get a connection timed out
.
Also when I check the hdfs dfsadmin -report
, I only see one datanode(on master's localhost) as part of hdfs.
Datanodes available: 1 (1 total, 0 dead)
Live datanodes:
Name: 127.0.0.1:50010 (localhost)
Hostname: master
Decommission Status : Normal
Configured Capacity: 8309932032 (7.74 GB)
DFS Used: 24576 (24 KB)
Non DFS Used: 2309738496 (2.15 GB)
DFS Remaining: 6000168960 (5.59 GB)
DFS Used%: 0.00%
DFS Remaining%: 72.20%
Here is my /etc/hosts file mapping on all the three nodes.
127.0.0.1 localhost
<ip of master> master
<ip of slave1> slave1
<ip of slave2> slave2
Here is my $HADOOP_HOME/etc/hadoop/slaves file on the master
localhost
slave1
slave2
In short, datanodes on slaves are unable to connect with hdfs.
What am I doing wrong?