How to Fix servers not displayed in System Management screen

[intapiuser]

In some cases, customers will navigate to the System Management screen and see the following:

As we can see, the nodes all seem that they are down as well as showing an undefined state. If this is not an issue with connecting to the servers (attempting to telnet within the cluster fails), this is a Management pod issue that can be resolved.

How to resolve it?

# Access the mongo pod (make sure this is the
[ec2-user:node1@~] kubectl -n sisense exec sisense-mongo-replicaset-0 -it -- mongo
 
# Switch to work with the 'be' DB
rs0:PRIMARY> use be
swtichted to db be
rs0:PRIMARY> db.getCollection("nodeDocument").remove()
 
# Exit mongo pod
rs0:PRIMARY> quit
bye
 
# Restart management pod
[ec2-user:node1@~] kubectl -n sisense delete pod -l="app=management"

Then go to the Data Groups page and delete the query/build values.