Kubernetes Helm Artifactory port inaccessible

Question

I'm trying to install Artifactory using Minikube & Helm 3

I'm following instructions here with some other bits I've found here and there: https://artifacthub.io/packages/helm/jfrog/artifactory-oss

I think I might be missing something obvious about exposing ports.

As far as I can tell, you need an 'Ingress' enabled?

So:

minikube addons enable ingress

After running the install,

$ cat namespace-artifactory.json
{
  "apiVersion": "v1",
  "kind": "Namespace",
  "metadata": {
    "name": "artifactory-oss",
    "labels": {
      "name": "artifactory-oss"
    }
  }
}

$ kubectl create -f namespace-artifactory.json

$ helm upgrade --install artifactory-oss --set artifactory.postgresql.postgresqlPassword=postgres_password --namespace artifactory-oss jfrog/artifactory-oss
Release "artifactory-oss" does not exist. Installing it now.
NAME: artifactory-oss
LAST DEPLOYED: Mon Apr 12 16:20:53 2021
NAMESPACE: artifactory-oss
STATUS: deployed
REVISION: 1
TEST SUITE: None
NOTES:
Congratulations. You have just deployed JFrog Artifactory OSS!


$ kubectl get namespace
NAME                   STATUS   AGE
artifactory-oss        Active   99s
default                Active   6d
kube-node-lease        Active   6d
kube-public            Active   6d
kube-system            Active   6d
kubernetes-dashboard   Active   4d22h


$ kubectl get svc --namespace artifactory-oss
NAME                                  TYPE           CLUSTER-IP       EXTERNAL-IP   PORT(S)                      AGE
artifactory-oss-artifactory           ClusterIP      10.98.138.1      <none>        8082/TCP,8081/TCP            100s
artifactory-oss-artifactory-nginx     LoadBalancer   10.104.240.114   <pending>     80:32112/TCP,443:30681/TCP   100s
artifactory-oss-postgresql            ClusterIP      10.100.119.202   <none>        5432/TCP                     100s
artifactory-oss-postgresql-headless   ClusterIP      None             <none>        5432/TCP                     100s


$ kubectl --namespace artifactory-oss get pods
NAME                                                 READY   STATUS    RESTARTS   AGE
artifactory-oss-artifactory-0                        1/1     Running   0          2m23s
artifactory-oss-artifactory-nginx-75b745c955-ffgql   1/1     Running   0          2m23s
artifactory-oss-postgresql-0                         1/1     Running   0          2m23s


$ kubectl logs --namespace artifactory-oss artifactory-oss-artifactory-0
... Server configuration reloaded on localhost:8046
... Server configuration reloaded on localhost:8049
... Server configuration reloaded on :8082
... frontend (jffe) service initialization completed in 40.24 seconds. Listening on port: port 8070

So now it looks like it worked, maybe, except I can't access any ports.

The cluster IPs are different. Seems odd.
No external IPs. Seems odd.

I opened ports in the 8000-9000 range using firewalld:

public (active)
  target: default
  icmp-block-inversion: no
  interfaces: ens192
  sources:
  services: dhcpv6-client ssh
  ports: 8000-9000/tcp
  protocols:
  masquerade: no
  forward-ports:
  source-ports:
  icmp-blocks:
  rich rules:

But none of the 80** ports are hitting anything.

Anyone familiar with this install, or know what I'm doing wrong?

Thanks

------------EDIT 1:

kubectl --namespace artifactory-oss describe service artifactory-oss-artifactory-nginx

Name:                     artifactory-oss-artifactory-nginx
Namespace:                artifactory-oss
Labels:                   app=artifactory
                          app.kubernetes.io/managed-by=Helm
                          chart=artifactory-11.13.0
                          component=nginx
                          heritage=Helm
                          release=artifactory-oss
Annotations:              meta.helm.sh/release-name: artifactory-oss
                          meta.helm.sh/release-namespace: artifactory-oss
Selector:                 app=artifactory,component=nginx,release=artifactory-oss
Type:                     LoadBalancer
IP Families:              <none>
IP:                       10.104.240.114
IPs:                      10.104.240.114
Port:                     http  80/TCP
TargetPort:               80/TCP
NodePort:                 http  32112/TCP
Endpoints:                172.17.0.2:80
Port:                     https  443/TCP
TargetPort:               443/TCP
NodePort:                 https  30681/TCP
Endpoints:                172.17.0.2:443
Session Affinity:         None
External Traffic Policy:  Cluster
Events:                   <none>

--------EDIT 2:

 minikube tunnel

Status:
        machine: minikube
        pid: 1380280
        route: 10.96.0.0/12 -> 192.168.49.2
        minikube: Running
        services: [artifactory-oss-artifactory-nginx]
    errors:
                minikube: no errors
                router: no errors
                loadbalancer emulator: no errors

Can you post logs of artifactory-oss-artifactory-nginx Pod as well? — AndD, Apr 13 '21 at 07:42
Not yet, and I don't know what the issue is, yet. Everything seems to be up. I think maybe the problem is that K8s is making a new IP for the cluster, but I'm on a single machine, and I need it to use the host IP, because new random IPs might not be whitelisted on the DNS. Just a guess though. — djb, Apr 22 '21 at 14:07

score 0 · Answer 1 · answered Apr 14 '21 at 20:50

0

Look into why your nginx ingress is <pending>.

kubectl --namespace artifactory-oss describe service artifactory-oss-artifactory-nginx

We are using this in production and it requires an nginx ingress engine installed. Might need more in minikube than just the ingress addon

answered Apr 14 '21 at 20:50

lance.johnsn

467
3
9

I tried, and pasted the output above. I am looking at https://stackoverflow.com/questions/44110876/kubernetes-service-external-ip-pending which might have an answer, but still not sure. tunnel doesn't seem to do much. – djb Apr 21 '21 at 14:04

Kubernetes Helm Artifactory port inaccessible

1 Answers1