search

fnproject / fn-helm

Helm Chart for Fn
Apache License 2.0
56 stars 24 forks source link

fn-fnlb stuck in CrashLoopBackOff #24

Closed ImVexed closed 6 years ago

ImVexed commented 6 years ago

Installing on a fresh Kubernetes single node cluster, 20GB claim, using Flannel.

Everything installs normally however the pfn-fn-fnlb pod gets stuck in a CrashLoopBackOff.

Output from kubectl get all --namespace fn

NAME                                    READY     STATUS             RESTARTS   AGE
pod/pfn-fn-api-zd6t9                    1/1       Running            0          1h
pod/pfn-fn-flow-depl-765cd56bbd-fqlkq   1/1       Running            1          1h
pod/pfn-fn-fnlb-depl-56756b5f4c-xkgjj   0/1       CrashLoopBackOff   21         1h
pod/pfn-fn-ui-54c5fcf7f8-kldnw          2/2       Running            0          1h
pod/pfn-mysql-84d45c4687-4p7g5          1/1       Running            0          1h
pod/pfn-redis-6fc56c95c9-9sqzk          1/1       Running            0          1h

NAME                  TYPE           CLUSTER-IP       EXTERNAL-IP   PORT(S)                       AGE
service/pfn-fn-api    LoadBalancer   10.110.76.62     <pending>     80:32636/TCP                  1h
service/pfn-fn-flow   ClusterIP      10.99.126.55     <none>        80/TCP                        1h
service/pfn-fn-ui     LoadBalancer   10.107.154.129   <pending>     3000:32169/TCP,80:32502/TCP   1h
service/pfn-mysql     ClusterIP      10.110.141.33    <none>        3306/TCP                      1h
service/pfn-redis     ClusterIP      10.107.48.48     <none>        6379/TCP                      1h

NAME                        DESIRED   CURRENT   READY     UP-TO-DATE   AVAILABLE   NODE SELECTOR   AGE
daemonset.apps/pfn-fn-api   1         1         1         1            1           <none>          1h

NAME                               DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/pfn-fn-flow-depl   1         1         1            1           1h
deployment.apps/pfn-fn-fnlb-depl   1         1         1            0           1h
deployment.apps/pfn-fn-ui          1         1         1            1           1h
deployment.apps/pfn-mysql          1         1         1            1           1h
deployment.apps/pfn-redis          1         1         1            1           1h

NAME                                          DESIRED   CURRENT   READY     AGE
replicaset.apps/pfn-fn-flow-depl-765cd56bbd   1         1         1         1h
replicaset.apps/pfn-fn-fnlb-depl-56756b5f4c   1         1         0         1h
replicaset.apps/pfn-fn-ui-54c5fcf7f8          1         1         1         1h
replicaset.apps/pfn-mysql-84d45c4687          1         1         1         1h
replicaset.apps/pfn-redis-6fc56c95c9          1         1         1         1h

kubectl describe pod fn-fnlib shows the Events:

Events:
  Type     Reason   Age                From                   Message
  ----     ------   ----               ----                   -------
  Warning  BackOff  1m (x379 over 1h)  kubelet, cheapsamples  Back-off restarting failed container
ImVexed commented 6 years ago

Some research seems to show that other similar error messages are the result of not specifying an ingress, which I didn't. May not be the root cause but it could point in the right direction.

peterj commented 6 years ago

@ImVexed do you have the logs from pfn-fn-fnlb pod? (e.g. kubectl logs pfn-fn-fnlb -n fn)

ImVexed commented 6 years ago

We've moved onto a different FaaS provider.