For your information: we are currently using Tyk on-premises as our corporate API Gateway that manages and proxy traffic from our applications to the services deployed on our own Kubernetes on-premise cluster. Tyk itself has been deployed as a service on this Kubernetes cluster.
Since one week ago we’re experiencing intermitent problems with one of the services deployed on our Kubernetes cluster that seems to be unreachable.
We get thousands of lines like this on Tyk-Gateway log:
time=“Apr 6 06:45:57” level=error msg=“http: proxy error: dial tcp 10.110.67.165:8080: connect: connection refused” api_id=d8fa13f8f9fd403744e43ce1dcf78875 org_id=5ce5448fe63d98000104b382 server_name=“pricingnewis-service.sales.svc.cluster.local:8080” user_id="****d480" user_ip=“10.68.18.47, 10.244.0.0” user_name=“[email protected]”
We only managed to fix it redeploying again this service on our Kubernetes cluster every now and then.
Please, do you know what could be causing this problem ? Apart from monitoring Tyk_Gateway logs is there any other way for debugging this kind of problems ) ?