I can't get the nginx controller to route based on the hostname. The YAML below doesn't work - traffic goes to the default back-end / I get a 404. However, if I remove the value for host, the ingress controller successfully routes traffic to my-service. The service works successfully if I place it behind a load balancer but I want to have multiple services working for different host names so I want to use an ingress controller and use a single IP. Thoughts?
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: ingress-nginx
annotations:
ingress.kubernetes.io/rewrite-target: /
spec:
rules:
- host: test1.mydomain.com
http:
paths:
- path: /
backend:
serviceName: my-service
servicePort: 80
The yaml looks slightly different than the rewrite example located here. The yaml is valid and kubectl apply or create should work but not produce the results you are expecting. Do you need the rewrite annotation or could you remove it and the back end service will respond without issue? If you don't need to rewrite anything try removing the yaml to just look like:
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: ingress-nginx
spec:
rules:
- host: test1.mydomain.com
http:
paths:
- backend:
serviceName: my-service
servicePort: 80
Related
I need to service gitlab, nexus and jupyterhub based on URL using one open port using k8s ingress.
If the path is written as "/" when create ingress, it works normally, but if you write "/nexus" like this, a problem occurs during the redirection process.
Have any of you solved the same problem? Please help.
my ingress.yaml as below
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
annotations:
nginx.ingress.kubernetes.io/rewrite-target: /$2
creationTimestamp: "2022-04-06T05:56:40Z"
generation: 7
name: nexus-ing
namespace: nexus
resourceVersion: "119075924"
selfLink: /apis/extensions/v1beta1/namespaces/nexus/ingresses/nexus-ing
uid: 4b4f97e4-225e-4faa-aba3-6af73f69c21d
spec:
ingressClassName: nginx
rules:
- http:
paths:
- backend:
serviceName: nexus-service
servicePort: 8081
path: /nexus(/|$)(.*)
pathType: ImplementationSpecific
status:
loadBalancer:
ingress:
- ip: 172.30.1.87
That's a problem with nexus itself. Your ingress works as intended, and you cannot do more from this side.
The problem here is that nexus webpage, i.e. index.html, requests resources in such a way that it's looking at the wrong place. You can see this by opening the network tab and inspecting the request URL of the missing statics.
To see what I mean, let's examine the below HTML image tags.
<img id="1" src="./statics/some-image.svg" alt="some image" />
<img id="2" src="/statics/some-image.svg" alt="some image" />
You can see that the first one, is using relative path, and would work with your configuration since the request URL would be relative to the location in the browser and then the nexus part gets stripped by the ingress controller.
However, the second one is using absolute path, so it will not have the nexus part in the request URL and the ingress controller will not be able to route it to the correct service.
This is a common problem when stripping path prefixes. It only works fully when the application you are serving when stripping a prefix is correctly configured.
In your case this means, checking the documentation of the services, if you have any way to influence this.
It may be more straight forward to route based on hostname instead of path. I.e nexus.myhost.com. For that, you would need a domain and point the corresponding A records to your ingress services IP / use a wildcard record.
I solve this problem by myself
I edited my pc hosts file
172.30.1.87 nexus.k8s.io
172.30.1.87 gitlab.k8s.io
I edited each Ingress in same service namespace
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
annotations:
nginx.ingress.kubernetes.io/rewrite-target: /
namespace: nexus
spec:
ingressClassName: nginx
rules:
- host: nexus.k8s.io
http:
paths:
- backend:
serviceName: nexus-service
servicePort: 8081
path: /
status:
loadBalancer:
ingress:
- ip: 172.30.1.87
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
annotations:
nginx.ingress.kubernetes.io/rewrite-target: /
name: gitlab-ingress
namespace: gitlab
spec:
ingressClassName: nginx
rules:
- host: gitlab.k8s.io
http:
paths:
- backend:
serviceName: gitlab-webservice
servicePort: 8181
path: /
status:
loadBalancer:
ingress:
- ip: 172.30.1.87
connect test
ingress Hostname + ingress Controller Nodeport
I am trying to setup Prometheus in k8 cluster, able to run using helm. Accessing dashboard when i expose prometheus-server as LoadBalancer service using external ip.
Same does not work when I try to configure this service as ClusterIP and making it as backend using ingress controller. Receiving 404 error, any thoughts on how to troubleshoot this?
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: ops-ingress
annotations:
#nginx.org/server-snippet: "proxy_ssl_verify off;"
nginx.ingress.kubernetes.io/rewrite-target: /$2
spec:
rules:
- http:
paths:
- path: /prometheus(/|$)(.*)
backend:
serviceName: prometheus-server
servicePort: 80
with above ingress definition in place, url “http://<>/prometheus/ getting redirected to http://<>/graph/ and then 404 error page getting rendered. When url adjusted to http://<>/prometheus/graph some of webcontrols gets rendered with lots of errors on browser console.
Prometheus might be expecting to have control over the root path (/).
Please change the Ingress to prometheus.example.com and it should work fine. (Changing it to a subdomain)
Please change your Ingress configuration file, add host field:
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: ops-ingress
annotations:
#nginx.org/server-snippet: "proxy_ssl_verify off;"
nginx.ingress.kubernetes.io/rewrite-target: /$2
spec:
rules:
- host: prometheus.example.com
http:
paths:
- path: /prometheus(/|$)(.*)
backend:
serviceName: prometheus-server
servicePort: 80
then apply changes executing command:
$ kubectl aply -f your_ingress_congifguration_file.yaml
The host header field in a request provides the host and port
information from the target URI, enabling the origin server to
distinguish among resources while servicing requests for multiple
host names on a single IP address.
Please take a look here: hosts-header.
Ingress definition: ingress.
Useful information: helm-prometheus.
Useful documentation: ingress-path-matching.
I have an EKS cluster and currently use path based routing. I previously posted this thread, so all my configuration is on there: Kubernetes: 502 (Bad getaway)
My ingress controller is from there: https://raw.githubusercontent.com/kubernetes/ingress-nginx/nginx-0.24.1/deploy/mandatory.yaml
Here is my ingress:
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: simple-fanout-example
namespace : default
annotations:
kubernetes.io/ingress.class: nginx
nginx.ingress.kubernetes.io/rewrite-target: /$2
spec:
rules:
- host: domain.com
http:
paths:
- path: /bleble(/|$)(.*)
backend:
serviceName: bleble-svc
servicePort: 8080
- path: /hello-world
backend:
serviceName: hello-world-svc
servicePort: 8080
This works absolutly fine, until you try to go anywhere that is not the domain.com/bleble or domain.com/hello-world. The services need to refer to each other, and the uri they request is just domain.com, which obviously doesn't work because the service is at domain.com/nameoftheservice.
The 2 problems are :
The service name is bleble, so we decided the path should be /bleble. I saw on this documentation (kubernetes.github.io/ingress-nginx/examples/rewrite) that adding (/|$)(.*) would allow rewrite. We want to use url as domain.com/bleble/swagger or domain.com/bleble/clients, ... But when we type those, it brings back to what was on /bleble
I need bleble to get an info from hello-world. Right now, instead of going from domain.com/bleble to domain.com/hello-world, it goes from domain.com/bleble to domain.com. it seems to only be able to call the host name, and not the path.
I've tried doing a single ingress resource and have nginx.ingress.kubernetes.io/app-root : /bleble but this doesn't seem to be working. I've tried the annotation as well nginx.ingress.kubernetes.io/rewrite-target: /coretest
Thanks for the help!
I'm moving to kubernetes using traefik as my Ingress Controller.
I have a single backend that should respond to 3000+ websites. Depending on the host, I need to add a custom header to the request before proxy passing it to the backend.
I can use the ingress.kubernetes.io/custom-request-headers annotation to add a custom header to the request but it's an annotation for the whole Ingress, so I would need to create 3000+ Ingresses, one for each website.
Is there another way to do this? Creating 3000+ Ingresses is the same thing as creating one Ingress with 3000+ rules?
Yes, you need to create one Ingress object per one host, if you want different headers her host.
You can do it by Traefik:
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: traeffic-custom-request-header
annotations:
ingress.kubernetes.io/custom-request-headers: "mycustomheader: myheadervalue"
spec:
rules:
- host: custom.configuration.com
http:
paths:
- backend:
serviceName: http-svc
servicePort: 80
path: /
Also, the same thing you can do by Nginx Ingress Controller.
It has the support for configuration snipper. Here is an example of using it to set a custom header per Ingress object:
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: nginx-configuration-snippet
annotations:
nginx.ingress.kubernetes.io/configuration-snippet: |
more_set_headers "Request-Id: $request_id";
spec:
rules:
- host: custom.configuration.com
http:
paths:
- backend:
serviceName: http-svc
servicePort: 80
path: /
BTW, you can use several different ingress controllers on your cluster, so it does not need to migrate everything to only one type of Ingress.
Is it possible to have a fallback service for Kubernetes ingresses in the event that none of the normal pods are live/ready? In other words, how would you go about presenting a friendly "website down" page to visitors if all pods crashed or went down somehow?
Right now, a page appears that says "default backend - 404" if that happens.
Here's what we tried, to no avail:
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: example-ingress
spec:
backend:
serviceName: website-down-service
servicePort: 80
rules:
- host: example.com
http:
paths:
- path: /
backend:
serviceName: example-service
servicePort: 80
For reference, we're testing locally with Minikube and deploying to the cloud on Google's Container Engine.
If using Nginx then default backend annotation should do the trick, sample:
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: ingress-name
namespace: your-namespace
annotations:
kubernetes.io/ingress.class: nginx
nginx.ingress.kubernetes.io/default-backend: fallback-backend
spec:
<your spec here>
For the Nginx Ingress Controller there is a flag --default-backend-service, which currently points to the service showing the "default backend - 404" message. Just replace it with the service you want. See https://github.com/kubernetes/ingress/tree/master/controllers/nginx#command-line-arguments
If you're using another Ingress Controller, I expect it to have a similar option.