1u0 commented on a change in pull request #9470: [FLINK-13380][k8s] Improve the
usability of Flink session cluster on Kubernetes
URL: https://github.com/apache/flink/pull/9470#discussion_r315612373
##########
File path: docs/ops/deployment/kubernetes.md
##########
@@ -166,18 +249,34 @@ kind: Service
metadata:
name: flink-jobmanager
spec:
+ type: ClusterIP
ports:
- name: rpc
port: 6123
- name: blob
port: 6124
- - name: query
- port: 6125
- name: ui
port: 8081
selector:
app: flink
component: jobmanager
{% endhighlight %}
+`jobmanager-rest-service.yaml`. Optional service, which is used to add a
`NodePort` on the jobmanager rest service so that user could submit job via
`<public-node-ip>:<node-port>`.
Review comment:
My reasoning, that there maybe two types of readers of this docs:
* people like devs, who have a test K8s cluster and want to setup
test/experimental Flink cluster. For this, I assume they can run `kubectl
port-forward|proxy` and also create resources by `kubectl`. For them the
existing means should be already enough to make Flink JM available from a local
dev machine. Showing them method that exposes `NodePort` would be like giving a
poor example;
* people like ops, that operate production K8s cluster. In that case they
are probably knowledgeable enough how to expose a service properly.
There are some blog posts in internet, for example [this
one](https://medium.com/google-cloud/kubernetes-nodeport-vs-loadbalancer-vs-ingress-when-should-i-use-what-922f010849e0)
that give some information what are the different ways and in which situations
to make service available outside of K8s cluster.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
