ceremony/drills/README.md
cluade-reviewer-andresmgsl e61bb91476
All checks were successful
CI / test (pull_request) Successful in 3m2s
CI / release-exercise (pull_request) Successful in 11s
CI / self-guards (pull_request) Successful in 8s
CI / action-exercise (pull_request) Successful in 6s
CI / docs-sync-exercise (pull_request) Successful in 6s
Refs guard / refs-not-closing (pull_request) Has been skipped
labels / labels (pull_request) Successful in 8s
docs(runner-probes): its own document, an arming procedure, and the evidence boundary made consistent (#202)
@codex-reviewer-andresmgsl's four gaps.

1. BRANCH UPDATED TO CURRENT MAIN. The commit's parent was pre-#204 dad99dd, so
   its green 22-file run did not contain the six test files and rules that
   landed with the sync. Merged main in — no rewrite — and re-verified against
   the 28-file suite the operator would actually receive.

2. WHO MAY RESET IT is now a section, and it says operator-owned until ruled
   otherwise, with content reset separated from archive/delete/admin. The
   access policy is set when the repo is created, which is the operator's step,
   so the two belong together. Flagged for @andres rather than assumed.

3. AN EXECUTABLE ARMING PROCEDURE replaces "install whatever the probe needs":
   fork ref and canonical SHA, caller stubs pinned to it, BOTH
   CEREMONY_SELF_REF carriers rewritten, the event invoked recorded by name,
   and what reset removes afterwards. It reuses the drill rehearsal's fork-ref
   pattern rather than inventing a floating pin, including its rule against
   ever creating a tag-shaped branch on heavy-duty/ceremony.

4. THE EVIDENCE CONTRADICTION IS RESOLVED. "Write results into an issue in this
   repo" and "no probe touches ceremony's board" could not both be followed in
   a file where "this repo" reads as ceremony. The job now writes raw results
   into the PROBE repo, and a human carries the issue URL and run number to the
   ceremony issue. The probe workflow holds no credential and no code path that
   can write to ceremony, which is what makes the two rules compatible.

Placement: the operational contract moves to docs/RUNNER-PROBES.md, with a
short cross-link in drills/README.md beside the disposal rule it excepts — the
exception stays visible where the dangerous habit lives, and neither document
grows a second top-level heading.

test/run.sh 28/28 on the updated branch; shellcheck 0.10.0, actionlint,
self-ref, marker, vendored and changelog-armed clean.

Refs #202
2026-08-05 13:40:28 +00:00

120 lines
6.6 KiB
Markdown

# Drills
What a drill means in this repo: an **end-to-end rehearsal of both doors of
the release workflow on a disposable repo**. The contract suite proves every
decision offline — facts → decide → notes against fixtures, the merge door's
step sequence replayed in release-exercise.yml — but the doors themselves
only ever run live: gating on a real push event, the tag create, the
publish, the `-dev` re-arm (release.yml's "what is honestly untested"). The
drill is where they run live *before* a version rests on them.
## The rehearsal
1. Create a scratch **private** repo. It is disposable by design — but the
disposal is split, because the builder cannot perform the delete: at the
end the builder **archives** it (`PATCH /repos/{owner}/{repo}` with
`archived: true`, inside the `repo` scope every fleet identity holds),
and **deleting it is the operator's step**`delete_repo` is
deliberately absent from bot tokens, fleet doctrine and not a
misconfiguration, so no builder that will ever run a drill can do it.
Do not retry the delete and do not wait on it: both 0.2.0 drills ended
at that wall independently (#135) — one builder held its release draft
in `state:building` re-trying a 403 that cannot succeed, the other wrote
a record asserting a delete that had not happened. **Cleanup gates
nothing** — not ready-for-review, not the review panel, not the merge.
The archived leftover is safe to leave: private, no consumers, and
outside `heavy-duty/ceremony`'s ref namespace — the namespace the
"never a branch named like the tag" rule below protects.
2. Install the docs/CONSUMERS.md caller stubs, pinned to a fork ref carrying
the release candidate tree. The candidate's `CEREMONY_SELF_REF` is by
construction the tag this release has not created yet, so the consumer
path cannot resolve directly from the candidate. Rewrite that pin to a
canonical candidate SHA in every carrier on the fork ref, and record the
fork ref and rewritten pin in the drill record.
Never create a branch named like the tag on `heavy-duty/ceremony` to
paper over this deadlock: it would shadow the tag for every consumer
until someone remembers to delete it. The 0.1.0 drill (#11) is the
worked example of this standing fork-ref shape.
3. Give it a fixture `VERSION` / `CHANGELOG.md` / `changelog.d/` /
`drills/` in the armed state (`X.Y.Z-dev`, the fragments directory with
its `README.md` marker plus at least one fragment for the ceremony to
consume).
4. Exercise both doors, one probe at a time:
1. a merge-door ceremony publishes exactly one release and re-arms main
to `-dev`;
2. a mislabeled ordinary PR is a green NOTICE no-op;
3. a bare-version PR without the `release` label refuses;
4. a re-run of the completed ceremony refuses;
5. a tag-door release from a manual tag;
6. a mismatched tag refuses.
Every refusal must refuse **creating nothing** — a probe that leaves a
tag or a release behind on a refusal path is a failed probe.
## The record
One file per version, `drills/X.Y.Z.md` — the shape the siblings use: what
was run, where, the result of each probe, failures written down plainly. The
record is the evidence; the scratch repo is the evidence's scaffolding. The
record names the scratch repo by full `owner/name` and states its disposal
state **as its author observed it when the record was written** — archived
and pending the operator's delete, or deleted only if the author genuinely
performed the delete. Never a disposal the author did not observe: the
record is the only thing that survives the drill, and 0.2.0's record shipped
its first draft asserting a cleanup that had not happened (#135) — false
evidence in the one file whose job is to be evidence.
A record has one of three shapes. A **rehearsal** records the disposable-repo
run above. **Doors unchanged** records the mechanically checked claim below
when a new rehearsal would execute the same bytes as the last one. **WAIVED**
records a maintainer's judgement under the standing paragraph below. If the
doors-unchanged conditions do not all hold, the release owes a rehearsal or a
waiver; the narrower shape is never a substitute for either.
## Doors unchanged
The builder may assert that no disposable-repo rehearsal is owed only when
all three conditions below hold at the candidate head. The release PR's panel
verifies the claim like any other evidence, and if any reviewer rules a full
drill owed, that verdict wins.
1. `git diff <last-rehearsed-tag>..HEAD -- <release-path>` contains no change
except the `CEREMONY_SELF_REF` pin line in
`.github/workflows/release.yml`.
2. The release path is exactly the output of
`.github/scripts/release-path.sh`: `.github/workflows/release.yml`, `bin/`,
`lib/version.sh`, `lib/decide.sh`, `lib/facts.sh`, and
`lib/changelog.sh`. The script is the record author's copy-paste source;
its contract test keeps this inline list and the workflow's direct and
transitive dependencies in agreement.
3. The last rehearsed tag's own record is a full rehearsal, its release is
published, and `main` was re-armed to `-dev` after it.
The baseline is the last **rehearsed** tag, never merely the previous tag. A
previous-tag baseline could chain one doors-unchanged assertion from another
while the doors drift a small diff at a time; the last-rehearsed anchor makes
any accumulated release-path change force a new rehearsal.
The record carries all three measurements as observed at its candidate head,
never copied from an earlier record. `drills/0.4.1.md` and
`drills/0.5.0.md` are the worked examples; the latter's amendment from a
predicted empty `lib/` diff to the observed `lib/ruling.sh` delta is why each
candidate is measured afresh (#233). Re-running its stricter baseline now is
also the path-enumeration proof: `git diff 0.4.0 0.5.0 -- <release-path>` is
only the `CEREMONY_SELF_REF` pin, while adding `lib/ruling.sh` makes the diff
non-empty even though neither release door reads that file (#217, #237).
`actions/drill-recorded` refuses any bare-version tree whose record is
missing or blank. A waived drill is still a record: the file says WAIVED and
why — a maintainer's call, visible and reviewable in the release PR's diff,
never a silent skip.
---
**Standing runner probes are not drills.** The disposal rule above — builder
archives, operator deletes — is for the disposable scratch repo a drill runs
in. `heavy-duty/ceremony-runner-probe` is the opposite shape: it stands, and
archiving it is the failure mode that made all three previous drill repos
unavailable. See [docs/RUNNER-PROBES.md](../docs/RUNNER-PROBES.md) (#202).