I have a 3-node Hadoop cluster (master and 2 slaves) with Hbase installed. Zookeeper and HMaster were installed on the Hadoop master node. After I started HBase Cluster by calling start-hbase.sh, the HMaster and the 2 HRegionServer processes were started.
However, the HBase cluster showed:
"Waiting for region servers count to settle..."
When I checked the log file on slaves, it showed an error message as
"Opening socket connection to server localhost/127.0.0.1:2181. Will not attempt to authenticate using SASL (unknown error)"
Not sure what that meant. I guess that the HRegionServer was trying to connect to localhost instead of the master node, which had the Zookeeper and HMaster installed. I have confirmed that communication between Master and Slaves through SSH was working either way, and all the physical IPs were correctly defined in /etc/hosts files.
Can any of you point out what could went wrong to interrupt the communication from HRegionServers to HMaster?
