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
Currently Radix-API sets the correct tekton and pipeline runner tags for new RadixJobs. Instead Radox-operator should control these, and they should be released together.
We should also use tagged versions when running them instead of master-latest
The text was updated successfully, but these errors were encountered:
Radix-tekton is actually part of the radix-pipeline jobs. Extracting the radix-pipeline from the radix operator and merging it to a radix-tekton (with changing the repo name to e.g. radix-pipeline) might be more preferable: Pros
Pipeline logic is not released when not-related radix-operator logic is released
No need to run extra jobs like prepare-pipeline of run-pipeline with providing back info through config maps
Image builders can be put to this repo to reduce maintanence cost
If/when pipeline is considered to run within Tekton pipeline, transition will be easier, with possibility to use Tekton workspaces with cloned sources Cons
After further evaluation I agree this is a reasonable trade-off to merge radix-tekton repo to the radix-operator, as it grown with extra logic like github properties and change analyses. It reduces amount of separate kube jobs, keeping same pipeline steps.
Currently Radix-API sets the correct tekton and pipeline runner tags for new RadixJobs. Instead Radox-operator should control these, and they should be released together.
We should also use tagged versions when running them instead of master-latest
The text was updated successfully, but these errors were encountered: