Things to do ->
Improve /health path. Add a check there if elasticsearch enabled then check that it's still accessible and working, Otheruse return non 200 error. It will be very useful for kubernetes deployment
The gateway reports these errors after is has been running idle for sometime. Is there a way to optimize this, in senses less errors.