You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A common convention for helm charts is to document breaking changes in the readme of the chart that is being upgraded.
Having this would be very useful because it's hard to scan commits across a whole repo for changes that may be relevant. When users are applying helm charts through, say, terraform, the diffs are not always visible to people.
I ended up scanning through flux commits yesterday to find the major bump to see if there was anything to beware of, only to be bitten by a much more significant breaking change in 2.2.0 (#144) that effectively ended up taking down a whole cluster when the unnoticed change uninstalled the kustomization crds.
Describe the solution you'd like.
An "Upgrading" section in the chart README ala e.g.:
Sorry about the negative tone here. I do think the charts are in general great so thank you for all the work that goes into them. It would just simplify the downstream pickup quite a bit to have changes documented.
The text was updated successfully, but these errors were encountered:
At the moment we have the change log in
inhttps://artifacthub.io/packages/helm/fluxcd-community/flux2 click on changelog - but yes would be better to have a readme or changelog section in the repo
Is your feature request related to a problem ?
A common convention for helm charts is to document breaking changes in the readme of the chart that is being upgraded.
Having this would be very useful because it's hard to scan commits across a whole repo for changes that may be relevant. When users are applying helm charts through, say, terraform, the diffs are not always visible to people.
I ended up scanning through flux commits yesterday to find the major bump to see if there was anything to beware of, only to be bitten by a much more significant breaking change in 2.2.0 (#144) that effectively ended up taking down a whole cluster when the unnoticed change uninstalled the kustomization crds.
Describe the solution you'd like.
An "Upgrading" section in the chart README ala e.g.:
Describe alternatives you've considered.
NONE
Additional context.
Sorry about the negative tone here. I do think the charts are in general great so thank you for all the work that goes into them. It would just simplify the downstream pickup quite a bit to have changes documented.
The text was updated successfully, but these errors were encountered: