-
Notifications
You must be signed in to change notification settings - Fork 123
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
upgrade of the trifecta components #2151
Comments
Thanks for the incredible detail @johndietz !! One question: Could I find the external-dns here: https://github.com/kubefunk-net/gitops/blob/main/registry/clusters/mgmt-funknet/components/external-dns/application.yaml |
pull request for step 1: https://github.com/gitops-ing/gitops/pull/1 |
All green / health confirmed after ugrade 👍 . Proceeding to step 2 |
pull request for step 2: https://github.com/gitops-ing/gitops/pull/2 |
pull request for virtual clusters (within step 2): https://github.com/gitops-ing/gitops/pull/3 |
What is your feature idea?
ingress-nginx, cert-manager, and external-dns applications need to upgrade to the latest helm charts.
when you provision a new kubefirst management platform (any cloud), the version of these apps will be outdated from latest available. for ease of documentation in this example, we'll assume a github org of
kubefunk.net
and a mgmt cluster namedmgmt-funknet
step 1: the mgmt cluster
the charts for these 3 components on the mgmt stack are defined at:
https://github.com/kubefunk-net/gitops/blob/main/registry/clusters/mgmt-funknet/components/ingress-nginx/application.yaml#L14
https://github.com/kubefunk-net/gitops/blob/main/registry/clusters/mgmt-funknet/components/ingress-nginx/application.yaml#L14
https://github.com/kubefunk-net/gitops/blob/main/registry/clusters/mgmt-funknet/components/cert-manager/application.yaml#L12
upgrading each to their latest chart, and then refreshing their app in argocd, and confirming health is step 1.
step 2: physical and virtual workload clusters
those upgraded components can also be applied to new clusters created by this mgmt stack. to upgrade physical cluster provisioning, you'll need to update:
https://github.com/kubefunk-net/gitops/blob/main/templates/workload-cluster/30-ingress-nginx.yaml#L14
https://github.com/kubefunk-net/gitops/blob/main/templates/workload-cluster/30-cert-manager.yaml#L14
https://github.com/kubefunk-net/gitops/blob/main/templates/workload-cluster/30-external-dns.yaml#L12
once you update these in the templates, you can test how well they orchestrate by going to kubefirst console on this stack, creating a physical cluster, and confirming health on these 3 resources on the new physical cluster in argocd.
repeat these steps for virtual clusters templating:
https://github.com/kubefunk-net/gitops/blob/main/templates/workload-vcluster/30-ingress-nginx.yaml#L14
https://github.com/kubefunk-net/gitops/blob/main/templates/workload-vcluster/30-cert-manager.yaml#L14
https://github.com/kubefunk-net/gitops/blob/main/templates/workload-vcluster/30-external-dns.yaml#L12
test with creation of a virtual cluster and confirming health in argocd.
step 3:
whatever changeset was required to get this stack operational, needs to be promoted up to the gitops-template repository, so that mgmt clusters will begin with the right charts and templates. there will be corresponding files for each of the files listed in steps 1 and 2 in the gitops-template. for example:
kubefunk stack:
gitops-template civo github:
all of the cloud stacks need these updates in the gitops-template. to test, provision a new stack using your gitops-template branch, and confirm that you can provision a new mgmt cluster, and then confirm that kubefirst instance can create a new physical and virtual cluster that provisions to 100% health in argocd in its registry.
Why is it needed?
cloud native never sleeps
Is this missing feature preventing you from using kubefirst?
Code of Conduct
The text was updated successfully, but these errors were encountered: