Checks to be done if  contrail_analytics_database_nodemgr process is down on one or all the servers.

contrail_analytics_database_nodemgr process went down

In the problematic controller , First check if the service is active as per contrail-status

== Contrail database ==
nodemgr: active
query-engine: active
cassandra: active

== Contrail analytics ==
nodemgr: active
api: active
collector: active

It should be active in status, and should not be inactive , timeout or initializing state.

2. Check the status of analytics_database_nodemgr_1 container

[root@aio-controller ~]# docker ps | grep analytics_database_nodemgr_1
07fa777b6d9e    hub.juniper.net/contrail/contrail-nodemgr:21.4.L2.329            "/entrypoint.sh /bin…"  9 months ago    Up 9 months                    analytics_database_nodemgr_1

Check and if container is Up and running
 

3. Check the logs for any error

# docker logs --tail 10 -f analytics_database_nodemgr_1

4. You may attempt a restart of container and see if stabilizes the situation and check for hints on fresh logs.
Command to restart service is 

# docker restart analytics_database_nodemgr_1