We added a health check to the JobWorker template, and a health check on the API container would be useful as well.
The previous JobWorker issue has lots of useful notes: adeutscher/RedShirt.Example.JobWorker#26
In particular, I'm keeping in mind a Reddit comment that I agreed with. The comment brought up the spectre of a pattern of overzealous health checks causing cascading container failures. A failing container health check suggests that the container should be restarted, and a transient issue with a downstream database server (or frankly anything that couldn't be solved with a restart) shouldn't trigger that. A healthy container and a healthy application are not the same thing.
We added a health check to the JobWorker template, and a health check on the API container would be useful as well.
The previous JobWorker issue has lots of useful notes: adeutscher/RedShirt.Example.JobWorker#26
In particular, I'm keeping in mind a Reddit comment that I agreed with. The comment brought up the spectre of a pattern of overzealous health checks causing cascading container failures. A failing container health check suggests that the container should be restarted, and a transient issue with a downstream database server (or frankly anything that couldn't be solved with a restart) shouldn't trigger that. A healthy container and a healthy application are not the same thing.