I lost all heartbeats on some datanodes of my cluster after a restart of machines. The problem is located just after the connection of ambari-agent of the host with ambari-server The last log I received in /var/log/ambari-agent/ambari-agent.log file of the defected DataNode : INFO 2016-04-20 17:59:48,925 PingPortListener.py:50 - Ping port listener started on port: 8670 INFO 2016-04-20 17:59:48,927 main.py:283 - Connecting to Ambari server at https://hmaster1.xxx.local:8440 (10.10.238.111) NetUtil.py:60 - Connecting to https://hmaster1.xxx.local:8440/ca With the working Datanodes the process continue with this line of log : INFO 2016-04-20 17:51:22,147 threadpool.py:52 - Started thread pool with 3 core threads and 20 maximum threads In the Log of the ambari-server located in /var/log/ambari-server/ambari-server.log file. I see anything between the defected DataNode and Ambari Master. I notice that I use the last version of ambari 2.2.1.1 and centos 7 with the last updates. I disabled all firewall rules and I have the same configuration for the working dataNode and the defected one. Any idea about this strange problem ? ... View more