In ZK service I had 5 different config versions and the service was all in-sync. Then I made an older version current. After like 5s the 'Restart Required' message showed up. This problem becomes worse on larger clusters - it took 20s on a 20 node cluster etc.
We need to see if UI is updating the status with such delay, or if there is a specific API that is slow to provide the updated status. If it is an API issue, please specify the exact API and turn it into a BE issue.