question

SiegfriedHeintze-9929 avatar image
0 Votes"
SiegfriedHeintze-9929 asked vipullag-MSFT commented

How to run ingress on AKS Virtual Nodes?


How can I run the ingress controller described here on the virtual nodes described here?

Apparently the yaml needs to be enhanced with special nodeselectors and tolerations ... Is that possible with helm?

Would it be possible with helm to deploy ingress, save the yaml for the deployment, delete the deployment, edit the yaml to include the special node selectors and tolerations and apply the yaml?


Mon Dec 21 2020 Afternoon Update:

So the answer makes sense for a production cluster but what about a development cluster? Might ingress run on a virtual node then?

Well let's assume I take your advice and run Ingress (and friends like the cert manager) on non-virtual node...

Is this a possibility? Would one non-virtual node be enough?

Can I run ingress (and the other required stuff like the cert-manager) on that one non-virtual node and have it invoke my application on the virtual node?

So in general can use service discovery (TCP/IP via REST communication) between apps running on the virtual node with apps running non-virtual node? I know that service discovery requires both apps to be in the same namespace and (I believe) same labels. Will the different nodeselector required for apps on the virtual node cause any problem?

Thanks

Siegfried


azure-kubernetes-service
· 1
5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.

@SiegfriedHeintze-9929

Please 'Accept as answer' if the provided information is helpful, so that it can help others in the community looking for help on similar topics.

0 Votes 0 ·

1 Answer

vipullag-MSFT avatar image
0 Votes"
vipullag-MSFT answered

@SiegfriedHeintze-9929

It is not recommended, even if its possible. Its good to run ingress within the cluster so that the latency from ingress to other pods in the less.

Virtual nodes are useful for workloads which are standalone (Ex: running CI, processing jobs etc) with few inputs and less interaction with other pods. If we run virtual nodes always (24 hours in a day), then it will cost more compared to the cost of running the same capacity on the cluster.

Hope this information is helpful.

Please 'Accept as answer' if the provided information is helpful, so that it can help others in the community looking for help on similar topics.

5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.