We have Cloudera 5.6
Our YARN and HDFS services stays in bad health after recovering. It has been 7 days since the issues resolved but in bad condition.
Cluster users are getting annoyed with it and keep asking me what the problem is. They don't feel comfortable when I say "there's actually no problem, i don't know why it's in bad condition"
Is this normal?
We do have kerberized cluster, could this be affecting health test update communication?