Bad Gateway on Deployment

[BugRoger]

Only 1 ApiServer causes downtime on deployment?

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[BugRoger]

This was fixed by adding maxSurge and minUnavailable options to the deployment of the apiserver. With only 1 replica the default settings cause a downtime when deploying a new image.

This caused a prominent error in the deployment pipelines: No consumer for text/html. The problem here was actually the parallel deployment of the admin apiserver and concurrently fetching credentials for it in the staging pipeline. Fix now.

In addition this issue surfaced a problem with nginx' default backend. It responds with text/html error pages even if json is requested. Would be great if it responded in json for the various (javascript) clients.

[BugRoger]

See #80