Hi,
What i can see from the logs so is it something very strange that is going on with the cluster under heavy load (for a long amount of time, it could take about 3-4 days after the load was ended to the problem start to exist in the cluster.
I can not found any drops on the interface that...