Replies: 1 comment
-
You may run For dry-run, look at |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
The bootstrap command looks great in the docs, but it's hard to run it for the first time with confidence in an existing cluster. I wish there was a
--dry-run
flag that would do everything except stop at committing the files to the repository.I have a single repository on Gitlab that has all my cluster manifests. The repository has a build directory organized into 3 folders for 3 separate namespaces. Each namespace has its own Gitlab agent, and each Gitlab agent is watching only it's related namespace folder.
ingress-controller
andsealed-secrets
and a few Kubernetes operators live in here)I wanted to just try out Flux on the staging manifests, but the Gitlab docs and the Flux docs don't make it clear how to accomplish this. My fear is that I run
flux bootstrap gitlab
and Flux makes unwanted changes to my production or infrastructure namespaces.flux bootstrap gitlab
does expose apath
option. The Flux docs for this option sound promising:But the Gitlab docs say
Even if both of these docs are accurate, then it seems I will have Flux configuration files deployed in each of the 3 namespaces managed by my repository. But that doesn't feel right to me.
Here is a different person having issues with multiple Gitlab Agents interacting with Flux. Admittedly his issue may be irrelevant, but it contributes to my hesitation to run the bootstrap command as a newcomer.
Ultimately, I wish there was a way for me to see what changes Flux wants to make without Flux making changes. And by changes, I mean changes to the repository (bootstrapping could output the files and not commit them), and also to the cluster (bootstrapping could output the changes it intends to make to the cluster).
Beta Was this translation helpful? Give feedback.
All reactions