@Andrey, the issue is not about contrail-status reflecting the correct state of the docker. The issue is that nodemgr doesn't send the correct status of the docker in the NodeStatus UVE and hence alarms are not being raised.
The url below shows the kafka status as RUNNING. This is incorrect.
Nodemgr should have detected that the docker was stopped and updated the same in the NodeStatus UVE.
@Andrey, the issue is not about contrail-status reflecting the correct state of the docker. The issue is that nodemgr doesn't send the correct status of the docker in the NodeStatus UVE and hence alarms are not being raised.
The url below shows the kafka status as RUNNING. This is incorrect.
Nodemgr should have detected that the docker was stopped and updated the same in the NodeStatus UVE.
https:/ /10.204. 217.46: 8143/proxy? proxyURL= http:// 5.5.5.234: 8081/analytics/ uves/database- node/nodeg6? cfilt=NodeStatu s:process_ info
{ STATE_RUNNING" , <<<<<<<<<<<<<<<<<<< Incorrect STATE_RUNNING" , database- nodemgr" , STATE_RUNNING" , STATE_RUNNING" ,
NodeStatus: {
process_info: [
{
process_name: "kafka",
start_count: 1,
process_state: "PROCESS_
last_stop_time: null,
core_file_list: [ ],
last_start_time: "1523364755000000",
stop_count: 0,
last_exit_time: null,
exit_count: 0
},
{
process_name: "zookeeper",
start_count: 1,
process_state: "PROCESS_
last_stop_time: null,
core_file_list: [ ],
last_start_time: "1523364755000000",
stop_count: 0,
last_exit_time: null,
exit_count: 0
},
{
process_name: "contrail-
start_count: 1,
process_state: "PROCESS_
last_stop_time: null,
core_file_list: [ ],
last_start_time: "1523364755000000",
stop_count: 0,
last_exit_time: null,
exit_count: 0
},
{
process_name: "cassandra",
start_count: 1,
process_state: "PROCESS_
last_stop_time: null,
core_file_list: [ ],
last_start_time: "1523364755000000",
stop_count: 0,
last_exit_time: null,
exit_count: 0
}
]
}
}