I don't have access to network provider now but I wonder if there's a way to bring back a slave node that is lost? in 8088 it shows the node is lost but in 50070 it says the node is healthy. I can ssh all the slave VMs and hadoop dfsadmin report shows all the nodes are healthy so I don't know why it tells node is lost in 8088.
P.S.: Should I restart yarn resource manager for this purpose and how?
P.P.S.: Why the container shows a -1 and how can I fix it in 8088?
@Mona Jalal - It looks like your datanode service is running fine but there seems problem with nodemanager, login to lost not and try to look at nodemanagers logs ( /var/log/hadoop-yarn/yarn/yarn-yarn-nodemanager-<hostname>.log )
You can restart nodemanager service on lost node to get this back in action. No need to restart Resource manager.
As per screenshot given, it looks like node is decommissioned.
Goto ambari UI --> Click on Yarn service --> Click on Nodemanagers --> Select lost node --> Goto Actions (top left) --> Filtered host --> Nodemanagers --> Recommission
I am not the network provider and I don't know what Ambari is. Our TA gave us these VMs from CloudLab.us so can I still do this and if yes, which port of public IP should I browse to?