Minions Issue, Job failure is large number, monitors are getting mixed responses

Hi,
We have upgraded the minions version recently to 3.0.28, we are at docker version 17.12.1-ce. We are getting mixed responses from minions, also we can see a large number of job are in queue for each minions. Are there any specific number of monitors needs to lined up for each minions. I am asking this because I wanted know if there would be a load balancing issue on minions. Currently there are around 550 monitors running in 2 private minions.