docs: warn that release candidate tags are load-bearing, not clutter - #10
Merged
Conversation
Housekeeping said the per-arch {sha} candidate tags were harmless and
could be deleted whenever clutter bothered you. They are not harmless.
publish-images assembles the multi-arch manifest for X.Y.Z / X.Y / X /
latest by referencing those per-arch manifests by digest, so deleting one
breaks docker pull for every tag pointing at it — including :1, the tag
courses are told to pin for a semester.
Audited the live packages: ae8ba79-* and af9af44-* are referenced by the
:1 and :2/:latest indexes respectively, so following the old advice would
have broken both published majors.
The advice also aged badly in the wrong direction: now that `plan` skips
builds for non-releasing commits, every candidate produced from here on
belongs to a release, so none of them are ever safe to delete.
Replaced with the rule plus a command to check a digest against what the
release indexes actually reference.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Updates maintainer documentation to correct prior housekeeping guidance about GHCR per-architecture candidate tags, clarifying that these tags are required inputs to the published multi-arch release manifests and must not be deleted once referenced.
Changes:
- Replaces the “candidate tags are harmless” advice with a warning that
{sha}-amd64/{sha}-arm64candidates are load-bearing once part of a release. - Adds guidance and a registry-query command intended to help determine whether a candidate digest is referenced by any release index.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Comment on lines
+163
to
+175
| REPO=ourplcc/devcontainers/plcc-ng # or plcc-ng-full | ||
| A='application/vnd.oci.image.index.v1+json,application/vnd.docker.distribution.manifest.list.v2+json' | ||
| TOK=$(curl -s "https://ghcr.io/token?scope=repository:$REPO:pull&service=ghcr.io" | jq -r .token) | ||
|
|
||
| # digests that releases depend on | ||
| for t in $(curl -s -H "Authorization: Bearer $TOK" "https://ghcr.io/v2/$REPO/tags/list" \ | ||
| | jq -r '.tags[]' | grep -Ev -- '-amd64$|-arm64$|^pr-'); do | ||
| curl -s -H "Authorization: Bearer $TOK" -H "Accept: $A" \ | ||
| "https://ghcr.io/v2/$REPO/manifests/$t" | jq -r '.manifests[]?.digest' | ||
| done | sort -u | ||
| ``` | ||
|
|
||
| Anything whose digest appears in that list must stay. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Housekeeping said the per-arch {sha} candidate tags were harmless and could be deleted whenever clutter bothered you. They are not harmless. publish-images assembles the multi-arch manifest for X.Y.Z / X.Y / X / latest by referencing those per-arch manifests by digest, so deleting one breaks docker pull for every tag pointing at it — including :1, the tag courses are told to pin for a semester.
Audited the live packages: ae8ba79-* and af9af44-* are referenced by the :1 and :2/:latest indexes respectively, so following the old advice would have broken both published majors.
The advice also aged badly in the wrong direction: now that
planskips builds for non-releasing commits, every candidate produced from here on belongs to a release, so none of them are ever safe to delete.Replaced with the rule plus a command to check a digest against what the release indexes actually reference.