Timeout issue with kong control plane and data plane

Installation/Setup
1

We have kong deployed in Openshift - control plane and data plane separately - with both connecting to a common DB running on a separate virtual machine (RHEL 7.9 and PostgreSQL 13.5)

The pods/instances are running fine, but occasionally, the GUI login has intermittent timeouts and also any admin api calls return a 502. This behaviour is not consistent, but does happen.

Even when logged into the admin UI or Kong Manager, sometimes, when you click on any of the sections, you get “Error, something went wrong” and upon re-try, it works.

Anyone experiencing a similar pattern in a container/VM based setup?
Is there any fix to this problem?

We are on kong 1.5.x

Is there any way to setup a heartbeat or stay alive mode between the DB and the kong instances to see if this will prevent the timeout issue?

Any help is appreciated

2

The issue was related to the NGIX WORKER PROCESSES definition. I had set the workers to 8 for a 1 CPU core instance. The issue was resolved by reducing the NGINX_WORKER_PROCESSES to 2 (1 should also be fine, but with 2, we are not experiencing the issues that we had)