I have set up a k8s two-node cluster (with flannel cidr 10.244.0.0/16) on
Ubuntu for experimental purpose at home. The setup is pretty basic.
I have docker is version 17.06-ce. I installed k8s and the two nodes are
reported as ready.
ube@sage ~/k8s $ kubectl get nodes
NAME STATUS AGE VERSION
parsley Ready 3d v1.7.4
sage Ready 3d v1.7.5
I know the version has some discrepancies as I did a apt-get update on
sage. Then I deployed a simple helloworld webapp, which is a default
Mojolicious webapp listening on port 3000...
This is the service definition:
apiVersion: v1
kind: Service
metadata:
name: helloworld-service
labels:
app: helloworld
spec:
type: NodePort
ports:
- port: 3000
nodePort: 30000
protocol: TCP
selector:
app: helloworld
And,
kube@sage ~/k8s $ kubectl describe service
Name: helloworld-service
Namespace: default
Labels: app=helloworld
Annotations: <none>
Selector: app=helloworld
Type: NodePort
IP: 10.100.172.154
Port: <unset> 3000/TCP
NodePort: <unset> 30000/TCP
Endpoints: 10.244.1.7:3000,10.244.1.8:3000,10.244.1.9:3000
Session Affinity: None
Events: <none>
Then, I fired up a deployment:
kube@sage ~/k8s $ kubectl describe deployments
Name: helloworld-deployment
Namespace: default
CreationTimestamp: Sat, 02 Sep 2017 09:37:51 -0400
Labels: app=helloworld
Annotations: deployment.kubernetes.io/revision=1
Selector: app=helloworld
Replicas: 3 desired | 3 updated | 3 total | 3 available | 0 unavailable
StrategyType: RollingUpdate
MinReadySeconds: 0
RollingUpdateStrategy: 25% max unavailable, 25% max surge
Pod Template:
Labels: app=helloworld
Containers:
helloworld:
Image: helloworld:1.0.0
Port: 3000/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Conditions:
Type Status Reason
---- ------ ------
Available True MinimumReplicasAvailable
Progressing True NewReplicaSetAvailable
OldReplicaSets: <none>
NewReplicaSet: helloworld-deployment-1481581006 (3/3 replicas created)
Events: <none>
And sure, all the pods are run on parsley (sage is the k8s master, but also
plays the role as node):
kube@sage ~/k8s $ kubectl get pods --output=wide
NAME READY STATUS RESTARTS AGE
IP NODE
helloworld-deployment-1481581006-6dgpp 1/1 Running 0 1d
10.244.1.7 parsley
helloworld-deployment-1481581006-j47cj 1/1 Running 0 1d
10.244.1.9 parsley
helloworld-deployment-1481581006-tmbxl 1/1 Running 0 1d
10.244.1.8 parsley
If I am on parsley, any of these work properly:
curl http://parsley:30000/
curl http://127.0.0.1:30000/
But, these do not work:
curl http://sage:30000/
curl http://[ip of sage]:30000/
On sage, all those curl commands listed above don't work...(it just times
out, i.e. after issuing the curl command, it kind of hangs there till it
times out)
I don't know if I understand the documentation correctly. I know a service
is run through kube-proxies which are present on each node, and the port I
specified (30000) should be open for connection, which I checked with
netstat -tupln | grep "30000", and it is the case on both sage and parsley:
kube@sage ~/k8s $ netstat -tupln | grep "30000"
(Not all processes could be identified, non-owned process info
will not be shown, you would have to be root to see it all.)
tcp6 0 0 :::30000 :::* LISTEN
-
And, if I hit this port on either node should direct me to the pod (which I
don't quite understand how this can be done in this case, as all pods are
on one node, and no pods are running on sage). How can I troubleshooting
this? Or I misunderstand the documentation.
I also do not quite understand what the IP of the service means:
10.100.172.154 in the service description.
Appreciate if someone can help me on this. Thanks!
Daniel.
--
You received this message because you are subscribed to the Google Groups
"Kubernetes user discussion and Q&A" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to kubernetes-users+unsubscr...@googlegroups.com.
To post to this group, send email to kubernetes-users@googlegroups.com.
Visit this group at https://groups.google.com/group/kubernetes-users.
For more options, visit https://groups.google.com/d/optout.
