I have setup health checks to check if my web apps are healthy, by checking with TCP request in port 80.
I noticed that in random times I get a notification that my web app is down, which is faulty. Please could be due to network bottleneck?
What kind of health checks you use? I preferred TCP over HTTP check, because in my mind TCP is simpler and faster.
Why are you using port 80? Is your server not being served over SSL?
TCP checks will only see if the port is active. It won’t tell you if the application is down.
Yes of course I use HTTPS, but in my mind TCP is faster and simpler than HTTPS, maybe my logic is wrong.
Usually if the server is down (port 80 is down) then the app is down as well.
This doesn’t make much sense. If you only check for the TCP connection, your server might be returning errors that you wouldn’t be able to detect without http/s health checks.
Things to account on health checks:
- Availability of the backend
- Return code of the endpoint to monitor.
Yes, i get to understand that will be optimal, however, in my case all i want to know is if the server is down. I guess i should replace TCP with HTTPS, right?
This topic was automatically closed 15 days after the last reply. New replies are no longer allowed.