apps is null despite ForecastleApp resource present about forecastle HOT 13 CLOSED

This is probably because of #105 @gargath @rasheedamir

from forecastle.

@ahmedwaleedmalik Unfortunately I currently don't have time to revisit this.
The discussion here has given me enough context, though, to figure this out when I do, in case I still have problems.

Please go ahead and close this and thank you very much for the insights.

from forecastle.

@ahmedwaleedmalik This is fixed for me now. I think the readme should be updated to reflect that CRDEnabled is not true by default. Thanks for the help!

from forecastle.

Sorry @gargath ! we missed the issue; were you able to fix or not? you are welcome to ask us on slack

from forecastle.

I did not, I'm afraid. I've parked it for now as I haven't had time to look into it further.
Any pointers would be appreciated, but I'm also happy to reach out on Slack.

Could the problem be that I have only got a ForecastleApp and no Ingress annotations?

from forecastle.

I didn't use helm. Never touch the stuff. ;-)
The CRD is present and I have successfully added the above with kubectl apply.
It's just not showing up on the landing screen.

from forecastle.

+1 on this except I do use Helm, however the CRD IS deployed in my cluster.

Any combination of namespaceSelector results in: "Looking for forecastle apps in the following namespaces: []"

from forecastle.

@gargath @challett try omitting the instanceName field. It adds an additional filter for instance name to look for and in your case i am assuming no such instance exists. https://github.com/stakater/Forecastle#running-multiple-instances-of-forecastle

Let me know if this solves your issue.

from forecastle.

Hey @ahmedwaleedmalik thanks for the reply. In my testing to try to resolve this I did remove instanceName to no avail.

I am setting the following configs: customApps, namespaceSelector.matchNames, title.

I've also tried namespaceSelector.any: true instead of matchNames but I'm seeing that the default is true anyways.

from forecastle.

@challett it would be of great help if you could share the complete deployment files that you are using as is. I'll try that at my end on Monday. Will get it fixed accordingly.

P.s. share the values if you are using helm chart.

from forecastle.

forecastle:
  labels:
    group: com.stakater.platform
    provider: stakater
    version: "v1.0.50"
  namespace: default
  image:
    name: stakater/forecastle
    tag: "v1.0.50"
  deployment:
    replicas: 1
    revisionHistoryLimit: 2
    annotations: {}
  config:
    namespaceSelector:
      matchNames:
      - namespace1
      - namespace2
      - namespace3
    title: Company Dashboard
    customApps:
      - name: "Some Custom App"
        icon: https://logo.svg
        url: https://google.com
        group: Apps
  proxy:
    enabled: false
  openshiftOauthProxy:
    enabled: false
  service:
    annotations: {}
    expose: "false"
  ingress:
    hosts:
    - host: dashboard.internal.url
      paths:
      - /
    enabled: true
    annotations:
      kubernetes.io/ingress.class: nginx
      kubernetes.io/protocol: http
      kubernetes.io/ssl-redirect: "true"
  route:
    enabled: false
    annotations: {}
    additionalLabels: {}
    host: ""
    port:
      targetPort: http
    wildcardPolicy: None
    tls:
      termination: edge
      insecureEdgeTerminationPolicy: Redirect

from forecastle.

@challett there was an issue with namespace selector and it's fixed with #118 now. Please can you try that workflow again ? I tried this with the values that you gave and it's working fine. Also, you'll have to add crdEnabled: true in your config so that it can poll for CRDs.

@gargath can you please verify this at your end so that we can close this issue ?

from forecastle.

That's a good point @challett. I'll revisit the readme and update it accordingly.
Closing this issue for now.

from forecastle.