- 
                Notifications
    You must be signed in to change notification settings 
- Fork 1.4k
Description
I'm currently trying to run this application in Kubernetes, but with no luck. I fixed a couple of problems with the deployments in my forked version: https://github.com/pulnara/ftgo-application
I'm using a Minikube Kubernetes cluster and the kubernetes-deploy-all.sh script. After executing it, pods responsible for Kafka, MySQL etc. are successfully running, but the ones for business logic (like accounting service, consumer service etc.) not. After a while (and multiple restarts) they're eventually going into the CrashLoopBackOff status, since all the time they don't meet the liveness probe - getting "connection refused" for the Spring Boot's /actuator/health endpoint.
This is my result of the kubectl get pods command:
NAME                                         READY   STATUS    RESTARTS   AGE
ftgo-accounting-service-55bd94d788-sh62d     0/1     Running   4          11m
ftgo-api-gateway-79948fdfbc-mxndr            0/1     Running   5          11m
ftgo-cdc-service-58fcf454b5-fdf67            0/1     Running   5          11m
ftgo-consumer-service-59fd7cd94d-wstrr       0/1     Running   4          11m
ftgo-dynamodb-local-0                        1/1     Running   0          13m
ftgo-dynamodb-local-init-4kpf5               1/1     Running   0          11m
ftgo-dynamodb-local-init-mntsj               0/1     Error     0          13m
ftgo-kafka-0                                 1/1     Running   0          13m
ftgo-kitchen-service-868476cb5c-rpxfc        0/1     Running   4          11m
ftgo-mysql-0                                 1/1     Running   0          13m
ftgo-order-history-service-b84cb6f49-8qtv6   0/1     Running   5          11m
ftgo-order-service-58b776fc7c-q4l22          0/1     Running   4          11m
ftgo-restaurant-service-5fdf567678-ffdvc     0/1     Running   4          11m
ftgo-zookeeper-0                             1/1     Running   0          13m
This are events from the ftgo-accounting-service pod's description:
Events:
  Type     Reason     Age                    From               Message
  ----     ------     ----                   ----               -------
  Normal   Scheduled  5m52s                  default-scheduler  Successfully assigned default/ftgo-accounting-service-55bd94d788-sh62d to minikube
  Normal   Pulled     5m25s                  kubelet            Successfully pulled image "msapatterns/ftgo-accounting-service:latest" in 8.6026105s
  Normal   Pulled     2m55s                  kubelet            Successfully pulled image "msapatterns/ftgo-accounting-service:latest" in 2.0851995s
  Normal   Created    2m54s (x2 over 5m25s)  kubelet            Created container ftgo-accounting-service
  Normal   Started    2m53s (x2 over 5m24s)  kubelet            Started container ftgo-accounting-service
  Warning  Unhealthy  68s (x6 over 4m7s)     kubelet            Liveness probe failed: Get "http://172.17.0.9:8080/actuator/health": dial tcp 172.17.0.9:8080: connect: connection refused
  Normal   Killing    67s (x2 over 3m27s)    kubelet            Container ftgo-accounting-service failed liveness probe, will be restarted
  Warning  Unhealthy  49s (x8 over 4m9s)     kubelet            Readiness probe failed: Get "http://172.17.0.9:8080/actuator/health": dial tcp 172.17.0.9:8080: connect: connection refused
  Normal   Pulling    35s (x3 over 5m33s)    kubelet            Pulling image "msapatterns/ftgo-accounting-service:latest"
  Normal   Pulled     31s                    kubelet            Successfully pulled image "msapatterns/ftgo-accounting-service:latest" in 3.8507742s
Did anyone happen to have a similar problem? I tried multiple times and on 2 different OS (Ubuntu, Mac OS), and getting the same error on both. I have no idea how to fix it.