Skip to content

ci: Split up network fetches and retry with Justfile targets#2187

Open
cgwalters wants to merge 1 commit intobootc-dev:mainfrom
cgwalters:provisioning-retries
Open

ci: Split up network fetches and retry with Justfile targets#2187
cgwalters wants to merge 1 commit intobootc-dev:mainfrom
cgwalters:provisioning-retries

Conversation

@cgwalters
Copy link
Copy Markdown
Collaborator

Our CI is long and flakes are painful, especially with merge queues. Let's take a heavy hammer and refactor all network-fetched resources into a clearly separate stage, and then we have a fully enforced retry loop for those that has even longer timeouts in CI.

Assisted-by: OpenCode (claude-sonnet-4-6@default)

@bootc-bot bootc-bot Bot requested a review from jeckersb May 5, 2026 15:08
@cgwalters cgwalters enabled auto-merge May 5, 2026 15:12
@cgwalters cgwalters added the ci/tier-1 Run CI for tier-1 OS (centos-10) only label May 5, 2026
Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request refactors the image provisioning process by splitting it into two distinct stages: a network-dependent 'fetch' stage and a network-isolated 'configure' stage. This change allows for independent retries of network-heavy operations via a new 'build-fetch' command in the Justfile, improving CI reliability against transient failures. A logic error was identified in the new 'provision-configure.sh' script where the root user's configuration was being deleted immediately after creation due to the order of operations and the use of symbolic links for the home directory.

Comment thread hack/provision-configure.sh Outdated
@cgwalters cgwalters disabled auto-merge May 5, 2026 15:17
@cgwalters cgwalters marked this pull request as draft May 5, 2026 15:17
Our CI is long and flakes are painful, especially with merge queues.
Let's take a heavy hammer and refactor all network-fetched resources
into a clearly separate stage, and then we have a fully enforced
retry loop for those that has even longer timeouts in CI.

Assisted-by: OpenCode (claude-sonnet-4-6@default)
Signed-off-by: Colin Walters <walters@verbum.org>
@cgwalters cgwalters force-pushed the provisioning-retries branch from c03b509 to 43b67e2 Compare May 5, 2026 16:58
@cgwalters cgwalters marked this pull request as ready for review May 5, 2026 17:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ci/tier-1 Run CI for tier-1 OS (centos-10) only

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant