You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Using a kubernetes deployment and ingrss, if one artificially creates a service gap by resizing to zero the number of app pods and the API is used, a 503 error response code by ingress (as expected), but it will not be in json even when the accept header requested it.
This could be solved by some additional configuration of the nginx service using a custom error backend service (which is used only when ingress/nginx does not find a match to the request), but it does not seem like a trivial task. Maybe there should be a quetzal server that only generates errors in the correct format.
The text was updated successfully, but these errors were encountered:
Using a kubernetes deployment and ingrss, if one artificially creates a service gap by resizing to zero the number of app pods and the API is used, a 503 error response code by ingress (as expected), but it will not be in json even when the accept header requested it.
This could be solved by some additional configuration of the nginx service using a custom error backend service (which is used only when ingress/nginx does not find a match to the request), but it does not seem like a trivial task. Maybe there should be a quetzal server that only generates errors in the correct format.
The text was updated successfully, but these errors were encountered: