-
Notifications
You must be signed in to change notification settings - Fork 221
ci: Harden the image build process #3174
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
Conversation
b29095d to
ba44f1a
Compare
ba44f1a to
e6230fc
Compare
e6230fc to
1e13657
Compare
1e13657 to
64ba43d
Compare
64ba43d to
167f632
Compare
|
Test of the CI build here, used Not certain of the best way to test the equivalent release build. |
joestringer
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Minor nit but LGTM. I reviewed mainly just by doing diff -u between the original workflow and each new workflow, and between each new workflow here to identify any discrepancies.
- Add a comment to clearly indicate the untrusted code checkout - Split the release and dev CI steps into two separate workflows - Explicitly specify the permissions of the GitHub token in the workflows, to minimize the impact of token compromise. - Split the individual workflows into separate jobs, such that the build happens in a different job to the push, which will allows the separataion of the untrusted code build process from the authenticated push of the built artifact. Signed-off-by: Feroz Salam <[email protected]> Reported-by: Cédric Van Rompay <[email protected]>
167f632 to
533473f
Compare
tklauser
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Test of the CI build here, used
pull_requestas the trigger to test: https://github.com/cilium/cilium-cli/actions/runs/21437055364/job/61730258277?pr=3174Not certain of the best way to test the equivalent release build.
So far we've been testing changes to release workflows only when cutting a release anyway, making adjustments as we hit issues. This approach is not optimal but I can't think of a good way to test this without too much complication. So IMO it is fine to merge this untested and follow-up with fixes in case we hit issues on the next release.
Thanks to @cedricvanrompay for reporting this.