Health Check ping path

Rack (Version 2)
1

In my convox.yml file the health ping path is set to /health. However looking in AWS in the load balancer settings I see the ping path is set to /check with port 3000 (classic type load balancer, the other I don’t see any health check settings). I have nothing setup in my app under the port 3000.

44%20AM
44%20AM541×537 36.9 KB

I do see that the response timeout and interval are the same as my convox.yml

Is this the correct behavior? The Convox Docs mentions that

deployment process depends on application Health Checks to determine success.

but it does not mention anything about the health checks of the load balancer to check for a healthy instance.

Are the health settings in convox.yml only for deployment purposes?

I want to understand better what are the settings of my convox.yml doing. Are they just setting health checks for deployment, or they are for instance level health checks (load balancer), or are they checking app level health, or even are they checking at process level for health checks?

2

Hi @moises,
The health check configuration in your convox.yml is used to configure the Load Balancer health check, which is used to mark instances as healthy or unhealthy during deployments, and in general to ensure the appropriate scale is maintained.

I’m not sure why your path is different in your LB than what is configured. Is it possible it was changed in the AWS console afterwards?

Thanks,
Ed

3

Hey @ed_convox,

I have not modified anything in AWS console. I was going through all the settings that convox configures for me in AWS to better understand what is going on behind the convox.yml and I noticed that the path is different in the LB. I have never set the path to /check. What it also gets my attention is that is pinging at port 3000 which is a port I never use.

Could it be a bug?