Using NGINX as a Kubernetes Ingress Controller
Vložit
- čas přidán 20. 05. 2024
- The Ingress controller is one of the most critical parts of Kubernetes platform, acting as the entry point for all incoming traffic to applications running on Kubernetes. That’s why it must be built on top of a proven and reliable load balancing technology, such as NGINX. The NGINX Ingress controller combines the benefits of using the Kubernetes control plane to manage load balancing configuration with the performance, reliability, and advanced features of NGINX or NGINX Plus. Additionally, the NGINX Ingress controller integrates NGINX with cloud‑native tools such as Helm and Prometheus, which are rapidly gaining adoption in production Kubernetes environments.
Speaker:
◆ Michael Pleshakov, Platform Integration Engineer, NGINX
This session is for beginning or intermediate Kubernetes users who are looking to deliver applications on Kubernetes in production. Michael shows how to successfully load balance HTTP as well as TCP/UDP applications on Kubernetes with NGINX Ingress controller.
Topics include:
◆ Installation through Kubernetes manifests or Helm
◆ Configuration of load balancing for HTTP and TCP/UDP applications
◆ Monitoring using Prometheus
◆ Troubleshooting in case of problems
◆ Extensions that support advanced requirements
Get Started with NGINX Ingress Controller
⬡ bit.ly/35BHoSi
NGINX Blog: Kubernetes Networking 101
⬡ bit.ly/3NSGCSy
Free eBook: Taking Kubernetes from Test to Production
⬡ bit.ly/3HpvaJL
Chapters:
0:00 - Using NGINX as a K8s Ingress Controller
0:02 - Kubernetes is Great
1:07 - K8s Ecosystem is Great
2:13 - NGINX for Kubernetes
3:03 - Service Discovery
5:09 - Ingress Resource
6:35 - Configuration Management
7:59 - Demo
9:10 - Installation on GKE
9:43 - Regular Install Ingress Controller
10:15 - Install Ingress Controller with Helm
11:18 - Demo
20:05 - Config Generation - ConfigMap
22:09 - Config Generation - Ingress
23:20 - Config Generation - Annotations
24:17 - Snippets - ConfigMap
24:55 - Snippets - Annotations
25:41 - Custom Templates
26:20 - Config Generation Summary
27:09 - Custom Annotations
28:18 - Monitoring with Prometheus
28:43 - TCP/UDP Load Balancing
29:44 - Benefits NGINX Plus Ingress Controller
30:17 - NGINX Ingress Controllers
30:44 - Q&A - Věda a technologie
Michael, Thaaaaank you, guys like you make the world better place by such contributions, keep it up and running :)
Hi, there is a bug when you have two ingress controlles with the same host, if you try to separate in two ingress (each ingress for triangule and circule) the nginx-ingress-controller will create two config file ( default-circule.conf and default-triangule.conf) and both config file will have the same host and the nginx dashboard will show warning in server zone. I think the nginx-ingress-controller should create config file per servername and not per deployment. In may case I have Almost 50 deployments and each app has their ingress.yaml with the same host (obviously diferente ports).
All I need to know about nginx controller. Thanks a lot.
nice video, thank you for share, I have a question, I recently heard at a conference that it was a good practice to use multiple ingress controllers in Kubernetes. Why?
Thank you for this video !!
Great presentation :)
Ngnix ingress cannot fetch the service from different namespace? Any solution for this...
Nice talk, ty Michael
so on which port is the ingress controller listening? 80? I did not see that specified anywhere in the ingress resource yml file.
yeah it should be port 80
and 443 for https
Great introduction video to nginx ingress. Learned so much. Thanks
Thanks dude It is good video presentation. I looking for this kind video. Thanks a lot man
I got 504 gateway error when try to connect edge server. So i need to update port forwarding and proxy annotations is this correct
Unable to access my kube cluster after applying nginx-config.yaml for tcp/udp. I am getting error like
Unable to connect to the server: dial tcp: lookup A13E7B443XYZCVBDS7868822.ak1.cp-abcd-5.eks.amazonaws.com on 172.31.0.2:53: no such host
- I will show you now how to do X
- Me: yes, finally
- But before I do that, let us jump back
- No, no, no , not back, no jumping, just show me
Still not clear about the need for L3/4 load balancer in front of L7 nginx.
Both nginx IC and L7 proxy will be in the same pod, these pods will be replicated by deployment and exposed by the load balancer service type since the service type is Load balancer we need an L3/L4 load balancer provided by a cloud vendor outside the cluster. hope this clarifies.
Great talk. Could you share the helm chart file by any chance?
HI the video is awesome ....could you pleaes mention the git repo used in the video it will be help fill thanks! cheers.
boring should be crisp and clear..
You’re welcome to do it
Lol