I have a service running inside Kubernetes proxied by multiple nginx pods that do SSL termination and basic authentication. I am seeing connection timeouts to the endpoint on a subset of traffic.

I have made a series of unique requests to the endpoint, and see some requests on each nginx pod, so each is receiving traffic. The ones that timeout however are nowhere to be found in the nginx logs.

The service itself is exposed via a GCE LB which directs traffic to the nginx pods, which themselves forward traffic to elasticsearch.


The issue was being caused by an Instance Group on GCE not having the appropriate tags assigned to it

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.