decisions(secops): record gitea cred rotation + SSH auth migration + history scrub

The token embedded in the recipe clone's gitea remote was copytree'd into the
world-readable per-run tree. SSH push keys the auth to ~/.ssh, so no secret is
written into any .git/config. API calls still use the token.

.gitignore

@@ -3,6 +3,8 @@
 *.key
 *.pem
 !docs/**/*.pem
+# holds a live provider API key at .provider.tinfoil.options.apiKey (F-redfix-2)
+config.json
 # python
 __pycache__/
 *.pyc

.gitmodules

@@ -1,3 +1,3 @@
 [submodule "secrets"]
 	path = secrets
-	url = https://git.autonomic.zone/recipe-maintainers/cc-ci-secrets.git
+	url = git@git.autonomic.zone:recipe-maintainers/cc-ci-secrets.git

machine-docs/BACKLOG-redfix.md

@@ -51,10 +51,352 @@ hold). Concrete fix designs from M1 evidence:
       doing the migration (major rewrite — official discourse image is launcher-based, likely
       infeasible cleanly). Lean (a)+tracked-upstream; may need operator input (DEFERRED?) — assess in M2.
 
+### M3 — post-VETO remediation (F-redfix-4)
+- [x] **keycloak warm-state slot collision** — FIXED at `redfix-m2-harness`@`b5f2b10`. `canonical_ns()` is
+      now the one namespace behind both the canonical's domain and its warm-state slot; live-warm provider
+      → `canon-<recipe>` slot, disjoint from the reconciler's `<recipe>/`. Plus a naming-independent
+      `_assert_slot_not_foreign()` guard. Unit suite 315→325; clearing condition re-run green on cc-ci
+      (each `restore()` returns its own stack's volumes; reconciler `last_good` survives). Verify per
+      STATUS-redfix.md "Gate: M2 RE-CLAIMED".
+- [ ] **B-redfix-5 — reconciler's post-`undeploy` warmsnap calls are outside the upgrade's `try/except`**
+      (NOT blocking; NOT part of F-redfix-4's clearing condition; recorded so it is not silently dropped).
+      In `warm_reconcile.py` there are **two** such sites, both after `abra.undeploy(domain)` and both
+      outside the `try/except` that guards the upgrade:
+      **(a) upgrade path** `:512-514` — `abra.undeploy` → `wait_undeployed` → `warmsnap.snapshot(...)`;
+      **(b) rollback path** `:534-536` — `abra.undeploy` → `wait_undeployed` → `warmsnap.restore(...)` →
+      `deploy_version(last_good)` at `:537`.
+      If either raises for ANY reason (absent/corrupt snapshot, foreign slot, docker error, "no volumes
+      found") the exception propagates, the following `deploy_version` never runs, and live keycloak is left
+      **undeployed**. Site (a) is on the *normal* upgrade path — it does not need a rollback to fire.
+      **Reachability, corrected (wake #42):** F-redfix-4's fix does not remove the trigger, it *adds* one —
+      `_assert_slot_not_foreign()` is a new raise inside BOTH `snapshot()` (`warmsnap.py:158`) and
+      `restore()` (`:209`). It is unreachable on cc-ci **today** because the live slot holds no
+      `snapshot/meta.json` (`read_meta` → `None`, and an unclaimed slot is free to claim) — *not* because
+      F-redfix-4 is closed. A pre-fix (`07fc6d4`) canonical seed is the one state that writes a foreign
+      domain into the live slot; see the MERGE PRECONDITION in STATUS-redfix.md.
+      **Consequence, corrected (wake #44, A-redfix-4) — worse than recorded above, still not blocking.**
+      The wedge is neither manual-recovery-only (my `68b51d5`) nor reboot-healed (A-redfix-2). A weekly
+      `nightly-sweep.timer` (`OnCalendar=Sun *-*-* 03:00:00`, `Persistent=true`) re-invokes
+      `warm_reconcile.py keycloak` via `nightly_sweep.roll_warm_infra()` (`:60`), so once armed the wedge
+      **alternates DOWN/UP every 7 days, indefinitely**: sweep N wedges at site (a); sweep N+1 takes the
+      fresh-deploy branch (`:471-479`, no `warmsnap`) and comes back up on the old version; sweep N+2
+      wedges again. It is also **silent** — `roll_warm_infra()` discards the subprocess rc (`:59-62`) so
+      `nightly-sweep.service` reports `Result=success`, and site (a) is upstream of every `write_alert()`
+      (`:493,500,503,539`). A weekly outage of the shared SSO provider would surface nowhere. Verified
+      first-hand against the node, not adopted from the finding.
+      **Not armed, and not self-arming:** `origin/main` — the tree the sweep executes (`CCCI_REPO=/etc/cc-ci`)
+      — has `WARM_CANONICAL = False`; only `07fc6d4` and `b5f2b10` have `True`, and `b5f2b10` ships the
+      `canonical_ns()` fix so its seeds land in `canon-keycloak/`. The binding rule is **never deploy
+      `07fc6d4`**.
+      Remedy sketch: wrap both sites so a warmsnap failure still redeploys `last_good` (or, if restoring
+      data is judged mandatory before redeploy, alert loudly + leave a breadcrumb rather than dying mid-
+      rollback), and make `roll_warm_infra()` propagate a non-zero rc so the failure is at least visible.
+      Needs a decision on which is safer for a DB-backed app after a forward migration — that
+      trade-off is why this is filed, not fixed inline. **Fixing it now would move the merge target off
+      `b5f2b10`, against which the M2 PASS was given, so it stays deferred for the operator.**
+- [ ] **B-redfix-6 — `canonical_ns()` docstring says "the 15 existing canonicals"; the real number is 17,
+      and the invariant is not a count** (COSMETIC; docs-only; no behaviour change). At
+      `redfix-m2-harness`@`b5f2b10`, `runner/harness/canonical.py:52` reads "zero blast radius on the 15
+      existing canonicals". Verified on live disk: `/var/lib/ci-warm/` has 20 slots, **17** carrying a
+      `canonical.json` (spared: `alerts`, `keycloak`, `traefik` — the reconciler dirs). More importantly the
+      guarantee is *structural*, not numeric: `WARM_DOMAINS` is a singleton (`{"keycloak"}`), 21 recipes are
+      enrolled, so exactly one re-keys (`keycloak -> canon-keycloak`) and the other 20 satisfy
+      `canonical_ns(r) == r`. The docstring's count will rot again on the next enrollment.
+      **Deliberately NOT fixed inline:** amending it would move the branch tip off `b5f2b10`, the exact sha
+      the M2 PASS was granted against and that every drift sweep pins. Fold into the next commit that moves
+      the branch for a substantive reason (e.g. B-redfix-5), rewording to the structural form.
+      Corrected in STATUS-redfix.md prose (which is not sha-pinned) as of wake #17.
+- [ ] **B-redfix-7 — orphaned non-Nix clone at `/etc/cc-ci` on the node stores the Gitea bot password in
+      plaintext in a world-readable `.git/config`** (OPERATOR CALL; out of redfix scope; found wake #18 while
+      checking a claim in re-confirmation #16). Facts, all read-only-verified on cc-ci:
+      - `/etc/cc-ci/.git/config` is mode **644** `root:root` and its `origin` URL embeds
+        `https://autonomic-bot:<password>@git.autonomic.zone/...` — the bot's Gitea credential at rest in
+        cleartext. (Value deliberately not reproduced here.)
+      - **Severity is LOW, not nil — but my first rationale for that was WRONG** (corrected wake #19 after
+        Adversary re-confirmation #17; filed Adversary-side as **A-redfix-1**). I originally argued "no
+        non-root *login* users, so only root can read it." That is the wrong axis: a process needs no login
+        shell to run as uid 1000, and mode 644 permits **any** uid to read. Verified directly —
+        `setpriv --reuid=1000 … /etc/cc-ci/.git/config` **succeeds** in the host namespace.
+      - **What actually contains the blast radius is mount-namespace isolation**, re-derived first-hand:
+        exactly ONE non-root process shares pid1's mount ns (`dbus-daemon`, uid 4). Every other non-root
+        process is containerized — notably the uid-1000 Quarkus `java` that IS the internet-facing
+        warm-keycloak: its `ns/mnt` is `4026533305` vs pid1's `4026531841`, and
+        `ls /proc/<pid>/root/etc/cc-ci` → **No such file or directory**. Mode-permits ≠ reachable.
+      - **Therefore the containment is load-bearing on a fragile property** and dies silently if a non-root
+        host-namespace daemon ever appears, or if `/etc` is bind-mounted into any container. That fragility —
+        not the current reachability — is the reason to fix it.
+      - `/etc/cc-ci` is a **real directory, not a `/nix/store` symlink**, and **no systemd unit references
+        it** — i.e. undeclared, unmanaged server state that nothing runs from. It violates the standing
+        "keep server state Nix-declared and reversible" rule.
+      **NOT actioned by me, deliberately.** The Gitea bot credential is a Class-A1 EXTERNAL infra input
+      (plan §4.4) — not mine to rotate or invent. And I did not create `/etc/cc-ci`, so it is not mine to
+      delete. Remedies for the operator, cheapest first: (a) `chmod 600` the config; (b) move the credential
+      to a `credential.helper` / netrc outside the repo; (c) if the clone is genuinely orphaned, remove it and
+      let Nix own any needed checkout. Rotating the bot password is worth considering regardless, since it has
+      sat in cleartext on disk.
+- [ ] **B-redfix-8 — the live Gitea bot password was committed to this repo, pushed to `origin/main`, and is
+      served to the UNAUTHENTICATED PUBLIC INTERNET** (**SEVERITY HIGH — URGENT operator rotation, not
+      deferrable**; found + redacted wake #19; public-exposure confirmed wake #20).
+      **Public exposure — independently verified wake #20 (not taken from the Adversary's report):** a plain
+      `urllib.request.urlopen` (no auth handler, no netrc, no git credential helper) of
+      `…/recipe-maintainers/cc-ci/raw/commit/14c7dee/machine-docs/BACKLOG-redfix.md` returns **HTTP 200, 33408
+      bytes, with the cleartext password in the body** (sanity-checked: body starts `# BACKLOG`, contains the
+      A-redfix-1 `Repro` block — a real fetch, not an error page). Cross-check that the fetch is intact:
+      `git cat-file -s 14c7dee:machine-docs/BACKLOG-redfix.md` → `33408`; a raw blob at a fixed commit sha is
+      immutable, so the served size MUST equal the object size. If your probe returns 33408, that is the leak,
+      not an error page. (Both figures were recorded as `33080` until 2026-07-09T02:24Z — a transcription slip
+      caught by the Adversary and re-measured against the blob; see REVIEW re-confirmation #22.) The mirror is public. So the leaked
+      credential is now **world-readable to anyone on the internet, with no account**, permanent in history,
+      replicated to every clone, AND push-capable to `recipe-maintainers/*`.
+      **STILL THE LIVE CREDENTIAL — measured, not assumed (wake #23, 2026-07-09).** Public *reachability*
+      (HTTP 200) and the value being *unrotated* are two different claims; until wake #23 only the first had
+      ever been measured, and "unrotated" was carried forward by repetition. Direct check: the current
+      `GITEA_PASSWORD` from `/srv/cc-ci/.testenv` is present verbatim in the `14c7dee` blob → **True**. So the
+      bytes the public internet serves ARE the password the harness authenticates with today; the exposure is
+      **not inert**. Operator re-check without printing the secret (verified to run, wake #23 — copy verbatim):
+
+          python3 -c 'import hashlib,re;v=re.search(r"^GITEA_PASSWORD=(.*)$",open("/srv/cc-ci/.testenv").read(),re.M).group(1).strip().strip(chr(34)).strip(chr(39));print(hashlib.sha256(v.encode()).hexdigest()[:16])'
+
+      → prints `3fcea78925015fc9` **while still unrotated**. A different digest ⇒ rotation happened ⇒
+      `14c7dee` is inert ⇒ this item can be closed. (The digest commits to the leaked value without
+      republishing it; `sha256` of the raw password, first 16 hex chars.) HEAD (`main`) and the redaction
+      commit `e99e2b3` were re-fetched publicly and are **clean** — only the historical commit `14c7dee`
+      serves it. This lifts A-redfix-1/B-redfix-8 from LOW (host-local, mount-ns-contained) to **HIGH**:
+      mount-namespace isolation is irrelevant once the same secret is on the open web.
+      While documenting A-redfix-1, `machine-docs/BACKLOG-redfix.md` gained a "repro" line that inlined the
+      **actual bot password** as a `grep` pattern. Introduced by commit **`14c7dee`**, which is **on
+      `origin/main`** — i.e. pushed to `git.autonomic.zone/recipe-maintainers/cc-ci`.
+      **This strictly escalates A-redfix-1.** That finding was a secret in one 0644 file on one host,
+      reachable only from pid1's mount namespace. This moved the same secret into a **git repository**, which
+      is: replicated to every clone (both loops' clones and the node's `/etc/cc-ci`), readable by anyone with
+      read access to the mirror, and — the sharp edge — the credential grants **push** access to
+      `recipe-maintainers/*`, so it now sits inside a repo it can write to.
+      **Done:** redacted from `HEAD` (wake #19); the repro now greps `autonomic-bot:` instead, verified to
+      return the same `1`, so the finding lost no verifiability.
+      **NOT done, and cannot be by me:** the secret remains in **history at `14c7dee` forever**. Excising it
+      needs a history rewrite + `--force`, which the standing rules forbid ("Never `--force`"), and which
+      would break both loops' clones. Redaction at HEAD stops propagation; it does **not** undo disclosure.
+      **⇒ Rotate the `autonomic-bot` Gitea password — URGENT.** It is a live, world-readable, push-capable
+      credential; every hour it stays valid it can be used by anyone who has fetched that public URL. It is a
+      Class-A1 external input, so only the operator can rotate it. **Rotation is the ONLY remediation that
+      actually closes this** — a history rewrite is both forbidden here (`--force`) and insufficient anyway,
+      because a value already served publicly must be presumed captured/cached/crawled and cannot be
+      un-published. Once rotated, `14c7dee` becomes inert. The re-issued credential must NOT go back into a
+      remote URL — see A-redfix-1's remedy ladder (`credential.helper` / netrc outside the repo).
+      **Process lesson (mine):** my pre-commit guard did catch this, but I had chained it with `;` instead of
+      `&&`, so the commit proceeded anyway and I pushed on top of the leak. A guard whose failure does not
+      halt the pipeline is decoration. Fixed by making the check `&&`-gated before `git commit` in wake #19+.
+
+- **B-redfix-9 — harden `CCCI_SKIP_FETCH` staging: it copies credentialed `.git/config` into a world-readable
+  run tree.** *(DEFERRED — post-merge; outside redfix DoD; no gate impact. Filed wake #54.)*
+  `fetch_recipe`'s staging branch (`runner/run_recipe_ci.py:348-353`) does
+  `shutil.copytree(~/.abra/recipes/<recipe>, <run-dir>)`. The canonical clones carry the bot password AND a
+  live oauth2 token in their remotes but are shielded by `/root` = `0700`; the run tree `/var/lib/cc-ci-runs`
+  is `0755`, so the copy **loses the protection of its source** — 133 world-readable cred-bearing copies today
+  (78 password, 117 token, overlapping). **CORRECTED (wake #55): this regenerates WEEKLY, not just on hand-runs.**
+  The `manual-*` prefix does not mean hand-run: the autonomous `nightly-sweep.timer` runs `run_recipe_ci.py`
+  outside Drone (`nightly_sweep.py:88` sets `CCCI_SKIP_FETCH="1"`), and `run_id()` labels every non-Drone run
+  `manual-<pid>` (`run_recipe_ci.py:318-319`). So the sweep itself is the generator and re-arms the exposure
+  each fire. (My earlier "0 production runs carry it, all `manual-*` = leftovers" mis-read `manual-*`; the
+  normal Drone fetch path IS clean — clean URL + per-command `http.extraHeader` token, landed `9b33fdf` — but
+  the sweep does not take that path.)
+  **Fix:** after `copytree`, strip userinfo from the copied remote (`git -C <dest> remote set-url origin <clean>`),
+  or exclude `.git/config` from the copy and re-init the remote clean; plus `chmod 0750 /var/lib/cc-ci-runs`.
+  **Do not** rely on rotating the password: rotation without this re-exposes the *new* value on the next staged run.
+
 ## Adversary findings
 
 (Adversary-owned — do not edit.)
 
+### A-redfix-1 [adversary] — `/etc/cc-ci/.git/config` is 0644 and embeds the Gitea bot password in cleartext (severity LOW, but for a different reason than B-redfix-7 states)
+
+Independently confirmed the Builder's B-redfix-7. Out of redfix scope; **no VETO, does not reopen the phase**
+(pre-existing undeclared server state, not created by redfix, not covered by any DoD item). Filed so the
+severity *rationale* is right, because the wrong rationale would let the mitigation evaporate silently.
+
+**Repro (read-only, from any shell with `ssh cc-ci`):**
+
+    stat -c '%a %U:%G %n' /etc/cc-ci/.git/config      # -> 644 root:root
+    grep -c 'autonomic-bot:' /etc/cc-ci/.git/config    # -> 1   (cleartext bot password in origin URL)
+    # [REDACTED by Builder, wake #19] the line above originally inlined the LIVE bot password as the
+    # grep pattern. Matching on the username is an equivalent repro and leaks nothing. See B-redfix-8.
+    stat -c '%a %n' / /etc /etc/cc-ci /etc/cc-ci/.git  # -> 755 on every parent: world-traversable
+
+**Why LOW — the correct reason.** B-redfix-7 argues "no non-root *login* users exist (no uid>=1000 with a
+real shell)". That is the wrong axis: the risk is any non-root *code execution*, login shell or not. The
+mode genuinely permits it —
+
+    setpriv --reuid=1000 --regid=1000 --clear-groups cat /etc/cc-ci/.git/config   # -> prints the password
+
+— so the file is readable by uid 1000. What actually holds the severity down is **mount-namespace
+isolation**, which neither of us had checked: *every* non-root process on the box is containerized and its
+`/etc` is not the host's. I enumerated it directly (compare `/proc/<pid>/ns/mnt` against `/proc/1/ns/mnt`):
+of all non-root processes, exactly **one** lives in the host mount namespace — `dbus-daemon`, uid 4
+(`messagebus`), not network-facing. Everything else (incl. the uid-1000 `java` = the warm-keycloak Quarkus
+container) is in its own namespace and **cannot** reach host `/etc/cc-ci`; verified:
+`ls /proc/<keycloak-pid>/root/etc/cc-ci/.git/config` -> *No such file or directory*.
+
+**Correction to my own probe.** My first pass read the uid-1000 `java` process off `ps`, saw `setpriv` as
+uid 1000 print the password, and was one step from concluding "the internet-facing Keycloak can steal the
+Gitea bot credential." It cannot. `setpriv` ran in the **host** namespace; the real uid-1000 process does
+not. Mode-permits != reachable-by-a-real-process. Recording the near-miss so it is not re-derived as fact.
+
+**Why it still matters.** The mitigation is incidental, not designed. It fails the moment anyone (a) runs a
+non-root daemon in the host namespace, or (b) bind-mounts `/etc` into any container. Contrast the 28 other
+`.git/config` files carrying the same cleartext credential under `/root`: those are protected by design —
+`/root` is `0700`, and the control probe confirms it
+(`setpriv --reuid=1000 … cat /root/.abra/recipes/drone/.git/config` -> *Permission denied*).
+`/etc/cc-ci` is the sole copy whose parents are world-traversable.
+
+**ESCALATION (wake #20, Adversary) — the disclosure is PUBLIC and unauthenticated; severity is no longer LOW.**
+My own commit `14c7dee` inlined the live password as a grep *value* and pushed it to `origin/main`. The
+Builder redacted `HEAD` (`e99e2b3`) and filed B-redfix-8 — correct, and it stops forward propagation — but
+the secret is permanent in **history** at `14c7dee`, and I have now established the fact neither of us had
+checked: **`git.autonomic.zone/recipe-maintainers/cc-ci` is a public mirror.** An **unauthenticated** HTTP
+GET (no creds, plain `urllib`, git's `insteadOf` cred-injection explicitly bypassed) of
+`…/raw/commit/14c7dee/machine-docs/BACKLOG-redfix.md` returns **200 with the cleartext password in the body**.
+So the credential is: in git history permanently, replicated to every clone, **served to anyone on the
+internet**, and it grants **push** to `recipe-maintainers/*`. That is a live, world-readable, write-capable
+credential — the worst quadrant, and strictly worse than the on-disk 0644 file A-redfix-1 originally
+described. This reclassifies the finding from LOW to **HIGH/urgent**. Only rotation clears it; a
+history-rewrite + force-push is forbidden by the standing rules and would not help once the value is already
+public. My scratchpad copy of the config was purged.
+
+**Operator action (agreed with Builder: not ours).** Rotate the `autonomic-bot` Gitea token (Class-A1
+external input, plan §4.4), then `chmod 600 /etc/cc-ci/.git/config` — or delete `/etc/cc-ci`, which is an
+orphaned real dir (not a `/nix/store` symlink) referenced by no systemd unit. `/root/.git-credentials` is
+already `0600`. **Closable only by me, after re-test.**
+
+### [adversary] F-redfix-4 — keycloak enrollment is collision-free in DOMAIN but NOT in warm-state: the live-warm reconciler and the new data-warm canonical share one per-recipe snapshot slot — **CLOSED @2026-07-09T00:18Z (VETO CLEARED)**
+
+**CLOSED by Adversary re-test.** Fixed at `redfix-m2-harness`@`b5f2b10` (parent `07fc6d4`): `canonical_ns()`
+is now the single namespace behind BOTH the canonical's domain and its warm-state slot, so a live-warm
+provider gets slot `canon-<recipe>/`, disjoint from the reconciler's `<recipe>/`. Plus a
+naming-independent `_assert_slot_not_foreign()` guard on snapshot AND restore.
+
+My cold re-test (full evidence in REVIEW-redfix.md @2026-07-09T00:18Z): the published clearing condition is
+met verbatim — slots disjoint, `restore(canon)` and `restore(live)` each return their OWN stack's volumes,
+reconciler `last_good` survives, foreign snapshot *and* foreign restore both refused. Beyond the Builder's
+own checks I added four: (a) the canon mariadb volume is byte-identical across the destructive restore
+round-trip (`4271926745 166164480`, 386 files); (b) **mutation testing** — reverting `canonical_ns()` reds 4
+of the new tests, removing the guard reds 2, so the 315→325 test delta is not vacuous; (c) every
+`snapshot`/`restore`/`app_dir` caller now passes an explicit slot, no bare recipe survives; (d) all 21
+enrolled recipes still resolve to their existing on-disk dirs (`registry_path("bluesky-pds")` is
+character-identical at parent and fix) — zero blast radius, no migration needed.
+
+Consequences 1–3 resolved. Consequence 4 (`prune_stale`) is now structural: `<recipe>/` never gains a
+`canonical.json`, verified in a scratch root. Enrollment retained (`WARM_CANONICAL = True`) — no silent
+de-enrollment. The two false "can never touch each other" comments are gone.
+
+Residual **B-redfix-5** (reconciler rollback `restore()` outside the upgrade's `try/except`) is NOT part of
+this finding's clearing condition and is **not** blocking: I confirmed it is verbatim present at parent
+`07fc6d4`, so it predates the enrollment. F-redfix-4 made it *reachable*; that path is now closed.
+
+<details><summary>Original report (as filed 2026-07-08T23:56Z)</summary>
+
+### [adversary] F-redfix-4 — original text — **OPEN, BLOCKING (VETO)**
+
+**Severity:** BLOCKS the phase's keycloak DoD item and must be fixed before the operator merges
+`redfix-m2-harness`. Worst case is an outage of the live shared OIDC provider that `lasuite-*`/`drone`
+depend on — the exact hazard the original canon §2.B de-enrollment exception existed to prevent,
+resurrected in a different namespace. Fails closed (raises), so **no silent data corruption**.
+
+**What the M2 fix does (and it does work, as far as it goes).** `canonical.canonical_domain()` now routes
+any recipe in `warm.WARM_DOMAINS` to `warm-canon-<recipe>`, and `tests/keycloak/recipe_meta.py` flips
+`WARM_CANONICAL = True`. The two stacks are genuinely distinct at the docker layer — verified on cc-ci:
+
+```
+canonical_domain(keycloak) = warm-canon-keycloak.ci.commoninternet.net
+WARM_DOMAINS[keycloak]     = warm-keycloak.ci.commoninternet.net
+stack_volumes(CANON) = ['warm-canon-keycloak_..._mariadb', 'warm-canon-keycloak_..._providers']
+stack_volumes(LIVE)  = ['warm-keycloak_..._mariadb',       'warm-keycloak_..._providers']
+```
+
+**The defect.** Warm *state* is keyed by RECIPE, not by domain:
+`warmsnap.snap_dir(recipe) = $CCCI_WARM_ROOT/<recipe>/snapshot` and
+`canonical.registry_path(recipe) = $CCCI_WARM_ROOT/<recipe>/canonical.json`.
+So both stacks now share **one** snapshot slot, `/var/lib/ci-warm/keycloak/snapshot/`, which already
+holds the live reconciler's sibling `last_good`. Two producers, two consumers, one slot:
+
+| | producer | consumer |
+|---|---|---|
+| live-warm | `warm_reconcile.py:512` `snapshot(recipe, warm-keycloak…)` (stateful=True, pre-upgrade) | `warm_reconcile.py:534` `restore(recipe, warm-keycloak…)` (health-gate rollback) |
+| data-warm | `canonical.seed_canonical` → `warmsnap.snapshot(recipe, warm-canon-keycloak…)` (via `run_recipe_ci.py:1047` `promote_canonical`, **no `WARM_DOMAINS` guard**) | `run_recipe_ci.py:896` `restore(recipe, warm-canon-keycloak…)` (quick-FAIL canonical rollback) |
+
+`warmsnap.snapshot()` atomically **replaces** the slot; `warmsnap.restore()` reads `meta.json` by recipe
+and then requires every recorded volume to exist in the *target* stack. Cross-stack names never match,
+so restore raises `SnapshotError` instead of cross-writing data.
+
+Consequences, in descending certainty:
+
+1. **Deterministic — canonical known-good destroyed.** Every stateful reconciler upgrade of live keycloak
+   overwrites the canonical's snapshot. The canonical's WC4 quick-FAIL rollback (`run_recipe_ci.py:896`)
+   then raises `SnapshotError` and cannot roll back.
+2. **Deterministic — reverse direction.** After a promote seeds the canonical, the slot holds canon volumes.
+3. **Race, high impact — live SSO outage.** The reconciler's window between its pre-upgrade `snapshot()`
+   and its rollback `restore()` spans `deploy latest` + `wait_healthy` (`health_timeout: 900`). A nightly-sweep
+   `promote_canonical(keycloak)` landing in that window replaces the slot with canon volumes. The rollback
+   then does `abra.undeploy(live)` → `wait_undeployed` → `warmsnap.restore(...)` → **raises**. `restore` sits
+   OUTSIDE the `try/except` that guards the upgrade, so the exception propagates and `deploy_version(last_good)`
+   never runs — **live keycloak is left undeployed**, taking SSO down for `lasuite-*`/`drone`.
+4. **Latent — `prune_stale()` invariant now false.** Its docstring promises it "Leaves the live-warm reconciler
+   dirs (keycloak/traefik — they have a `last_good`, no `canonical.json`) untouched." Once keycloak is seeded it
+   *has* a `canonical.json`; if `WARM_CANONICAL` is ever flipped back to False, `prune_stale` matches it and
+   `shutil.rmtree(app_dir("keycloak"))` deletes the live reconciler's `last_good`.
+
+**Why M2's verification could not have caught this.** The enrollment's data path never executed: on cc-ci,
+`/var/lib/ci-warm/keycloak/` contains **only** `last_good` — no `canonical.json`, no `snapshot/` — while a normal
+canonical (`/var/lib/ci-warm/cryptpad/`) has both. The `warm-canon-keycloak_*` volumes exist, so the promote
+*deployed*, but `seed_canonical` never ran (registry-advance is deliberately deferred to the operator's merge).
+The first-ever keycloak seed will therefore happen post-merge, in production, unexercised.
+
+**The shipped code asserts the opposite.** `canonical.py` docstring: "a separate stack/domain that can never
+touch the live provider"; `recipe_meta.py`: "separate deployments that can never touch each other… structurally
+impossible." Both are false for warm state. That claim is what I falsified.
+
+**Repro (cold, non-destructive — writes only to a scratch `CCCI_WARM_ROOT`; never touches the live stack).**
+Uses the real idle `warm-canon-keycloak` stack and idle `warm-custom-html` as a stand-in for the live stack
+(the live one cannot be undeployed to snapshot it):
+
+```sh
+ssh cc-ci
+git clone -q --branch redfix-m2-harness <cc-ci remote> /tmp/p8 && cd /tmp/p8/runner
+CCCI_WARM_ROOT=/tmp/p8w /nix/store/jag2131a95gw6ng7grig9pj3dn2q8vrv-python3-3.12.8-env/bin/python3 - <<'PY'
+import sys; sys.path.insert(0, ".")
+from harness import warmsnap as ws, canonical as c
+CANON, STANDIN = c.canonical_domain("keycloak"), "warm-custom-html.ci.commoninternet.net"
+print(ws.snap_dir("keycloak"))                       # one slot, domain-blind
+ws.snapshot("keycloak", CANON,   version="canon-known-good")
+ws.snapshot("keycloak", STANDIN, version="live-last-good")   # clobbers it
+print(ws.read_meta("keycloak")["domain"])            # -> warm-custom-html…  (canon known-good GONE)
+ws.restore("keycloak", CANON)                        # -> SnapshotError
+PY
+```
+
+EXPECTED (observed @2026-07-08T23:55Z):
+```
+/tmp/p8w/keycloak/snapshot                       <- SAME slot for BOTH domains
+read_meta(keycloak).domain = warm-custom-html.ci.commoninternet.net   <- canon known-good DESTROYED
+SnapshotError -> snapshot volume warm-custom-html_ci_commoninternet_net_content
+                 absent from current stack ['warm-canon-keycloak_..._mariadb', 'warm-canon-keycloak_..._providers']
+```
+Post-probe the node was verified untouched: `/var/lib/ci-warm/keycloak/` still `last_good` only, all three
+volumes intact (159M / 8.0K / 40K, file counts unchanged), live `warm-keycloak…/realms/master` → 200. Scratch removed.
+
+**Proposed remedy (Builder's to choose — mine to file, not to make).** Key warm state by the *stack/domain*
+rather than the bare recipe for recipes in `WARM_DOMAINS` — e.g. `app_dir()` takes the domain, or the canonical
+seeds under `<recipe>-canon/`. Then: fix `prune_stale`'s now-false invariant, and correct the two "can never
+touch each other" comments. A guard alone (skip `seed_canonical` for `WARM_DOMAINS` recipes) would silently
+de-enroll keycloak and re-open the DoD item, so it is not sufficient.
+
+**Clears the VETO when:** the two stacks provably use disjoint warm-state paths, and a seeded keycloak canonical
+survives a live-reconciler stateful upgrade (and vice versa) — demonstrated by re-running the repro above and
+seeing each `restore()` return its OWN stack's volumes.
+
+</details>
+
+---
+
 ### [adversary] F-redfix-1 — discourse migration INCOMPLETE: dangling image-less `sidekiq` in compose.smtpauth.yml (R011 lint regression + breaks SMTP-auth deploys) — **CLOSED @2026-06-18T07:06Z**
 
 **CLOSED by Adversary re-test.** Builder fixed in PR #4 @9ff5e19 (force-pushed onto 53ba0910): removed the
@@ -107,3 +449,461 @@ ABRA_DIR=$LA script -qec "abra recipe lint -n discourse" /dev/null   # -> R011 X
 **Proposed remedy (recipe PR #4):** remove the orphaned `sidekiq:` block from `compose.smtpauth.yml` (fold
 its `DISCOURSE_SMTP_PASSWORD_FILE` env + `smtp_password` secret into the `app` service, since sidekiq is now
 internal). Re-run discourse cold -> EXPECT R011 OK, level=5. Only the Adversary closes this, after re-test.
+
+### [adversary] F-redfix-2 — live API key sat untracked **and un-gitignored** at the Builder clone's repo root (`config.json`) — one `git add -A` from being pushed to origin — **CLOSED @2026-07-08T23:26Z**
+
+**CLOSED by Adversary cold re-test.** Builder remedied @`8cf08fd`: `config.json` added to the "local secrets /
+env — never commit" block in `.gitignore` (line 7, with a comment naming the finding). My independent
+verification, none of it taking the Builder's word:
+1. **Attack replay from cold** — `git add -A` into a scratch `GIT_INDEX_FILE` seeded from HEAD: staged paths
+   are `main.go` only; `config.json` **not staged**. `git check-ignore -v config.json` → `.gitignore:7`.
+2. **Fix is on origin, not just local** — `git show origin/main:.gitignore` contains `config.json`. A *fresh
+   clone from origin* + dropping the real `config.json` in → ignored ✅, `git add -A` does not stage it ✅.
+   This matters: a local-only .gitignore edit would not protect the next clone.
+3. **Full key never committed** — my original evidence used the 6-char prefix and is now **contaminated**: our
+   own finding/inbox/journal text contains `tk_bhg`, so `-S'tk_bhg'` yields false positives. Re-tested against
+   the *full 51-char value*: `git log --all -S"$KEY"` → **0 commits** in BOTH `cc-ci` and `cc-ci-adv`. Binary
+   search on prefix length: the longest prefix ever committed anywhere is **6 of 51 chars**, in our own
+   documentation — not a usable disclosure. No leak, past or latent.
+4. **No non-git exposure** — dashboard is live (`https://ci.commoninternet.net/` → 200) but
+   `/config.json` → **404** (also 404 on `dashboard.ci.…`); no tracked source reads it (the other
+   `config.json` hits are `/root/.docker/config.json`, unrelated). Perms `-rw-r--r-- loops:users`.
+
+**CORRECTION to my own finding (Builder was right, I was wrong).** I wrote "BOTH Builder clones". There is
+only **one** repo: `/srv/cc-ci` is a symlink → `/srv/cc-ci-orch` (`ls -ld`), so `/srv/cc-ci/cc-ci` and
+`/srv/cc-ci-orch/cc-ci` share `rev-parse --show-toplevel`, the same `.git` inode (3206558) and the same
+`.gitignore` inode (3252849). My `cc-ci-adv` "pair" is the same illusion. A filesystem-wide sweep found
+exactly one `config.json` inside any git repo, and it is now IGNORED. One fix, fully applied — not half.
+
+**Residual, explicitly NOT closed by this:** the key is still on disk **unrotated** (`len=51`, `tk_bhg…`).
+Gitignoring prevents a future commit; it cannot un-expose a value that leaked by another channel. Since the
+full key provably never entered git and is not HTTP-reachable, git is not a reason to rotate. The Builder
+correctly **escalated rotation to the operator rather than deciding it** — that judgement was right, and the
+call remains the operator's.
+
+---
+
+<details><summary>Original report (as filed 2026-07-08T23:12Z)</summary>
+
+### [adversary] F-redfix-2 — original text — **OPEN, NON-BLOCKING**
+
+**Severity:** does NOT block phase `redfix` (out of scope of its Definition of Done — no VETO, DONE stands).
+Latent secret-leak risk in the working environment; worth fixing before any future phase does a broad `git add`.
+
+**What:** `config.json` (1128 B, mtime 2026-06-23T00:50Z) exists at the repo root of BOTH Builder clones —
+`/srv/cc-ci/cc-ci` and `/srv/cc-ci-orch/cc-ci`. It holds a live-shaped inference credential at
+`.provider.tinfoil.options.apiKey` (51 chars, prefix `tk_bhg…` — value not reproduced here). The file is
+**untracked**, but `.gitignore` does **not** cover it: `.gitignore` lists `.testenv`, `*.key`, `*.pem`,
+`runs/`, `.claude/` — no `config.json`. So `git check-ignore config.json` → miss.
+
+Origin is a real pushed remote (`git.autonomic.zone/recipe-maintainers/cc-ci.git`, credentials embedded in
+the remote URL). A single `git add -A` / `git add .` in either clone would stage and then push the key.
+
+**Good news (verified, not assumed):** the key has never been committed —
+`git log --all --oneline -S'tk_bhg'` → empty; `git log --all -- config.json` → empty; `git ls-files` has no
+`config.json` at any path. So this is a *latent* risk, not an existing leak. The Adversary clones
+(`/srv/cc-ci/cc-ci-adv`, `/srv/cc-ci-orch/cc-ci-adv`) do not carry the file at all.
+
+**Repro:**
+```
+cd /srv/cc-ci/cc-ci && git status --porcelain config.json     # -> "?? config.json"
+git check-ignore -v config.json; echo "exit=$?"               # -> exit=1 (NOT ignored)
+git log --all --oneline -- config.json                        # -> empty (never committed)
+```
+
+**Proposed remedy (Builder — repo change, mine to file, not to make):** add `config.json` to `.gitignore`
+under the existing "local secrets / env — never commit" block. Optionally rotate the Tinfoil key if it was
+ever pasted into a log/transcript. I did **not** touch, move, or delete the file — it holds a live-looking
+credential and is not mine to modify.
+
+**Discovery:** independent break-it probe on my "no secrets in the repo / published logs / dashboard"
+standing mandate, run after the phase closed. The Builder's journal @418ec57 independently noticed the same
+file; I verified the exposure surface (gitignore miss + never-committed) from a cold start rather than
+taking that note at face value. Only the Adversary closes this, after re-test.
+
+</details>
+
+### [adversary] F-redfix-3 — M2's discourse evidence shas (`9ff5e19`, `53ba0910`) no longer exist on the mirror; the fix content survives — **CLOSED @2026-07-08T23:24Z (non-blocking, no VETO)**
+
+**Severity:** does NOT block phase `redfix` (DONE stands). Evidence-durability defect in the *record*, not in
+the fix. Filed so a future auditor of redfix does not conclude "the discourse fix was withdrawn."
+
+**What.** `STATUS-redfix.md` pins the discourse fix at `9ff5e19` (fix list) and `53ba0910` (WHERE refs) on
+`recipe-maintainers/discourse` branch `discourse-official-image`. As of 2026-07-08 that branch heads at
+`ede6399` and **neither sha resolves**: fetching all 17 `refs/heads/*` + `refs/pull/*/head` into one clone and
+running `git cat-file -t` on each returns *not a valid object name*. The branch was force-pushed/rebased and
+extended by **later** phases (`ede6399` = `refs/pull/5/head`; adds `discourse/postgres:pg18` + `POSTGRES_USER`
+in `pg_backup.sh`). redfix's PR is also no longer "#4" — `refs/pull/4/head` is now `0c4539b7`.
+
+**Why it is CLOSED rather than a VETO.** I re-verified the *content* the M2 PASS actually asserted, at the
+current head: `compose.yml` → `image: discourse/discourse:3.5.3` (official-image migration) and
+`compose.smtpauth.yml` → 0 `sidekiq` occurrences (the F-redfix-1 remedy). Both hold at `ede6399` and at
+`0c4539b7`. The fix is present and re-verifiable; only the pointers rotted. M2 was correct when given.
+
+**Repro.** `git clone https://git.autonomic.zone/recipe-maintainers/discourse && cd discourse && git fetch
+origin 'refs/heads/*:refs/remotes/origin/*' 'refs/pull/*/head:refs/remotes/pr/*' && git cat-file -t 9ff5e19`
+→ fatal. Then `git show origin/discourse-official-image:compose.yml | grep image:` → official image present.
+(Note: `git fetch origin <sha>` and a `--filter=blob:none` clone both give false "absent" signals — use
+reachability from all refs.)
+
+**Lesson for future phases (no action required of the Builder now):** shared recipe branches get rewritten, so
+a sha alone is not durable evidence. Record the *content assertion* (file → expected line) alongside the sha,
+or push a tag. The other three redfix fixes pinned exactly (`4ca7f418`, `a0f2db88`, `4987ba91`), as did cc-ci
+`redfix-m2-harness`@`07fc6d4a` — discourse drifted only because a later phase reused its branch.
+
+### A-redfix-2 [adversary] — **CLOSED 2026-07-09T08:59Z** (accepted at `f64d102`; false half superseded by A-redfix-4 at `e356698`) — STATUS's "no code path redeploys it / recovery is manual" is REFUTED: the B-redfix-5 wedge **does** self-heal on the next unit activation, which makes it look intermittent
+
+> **⚠ PARTIALLY WITHDRAWN by me at wake #44 (2026-07-09T08:5xZ) — see A-redfix-4.**
+> The *first* half (a code path redeploys it → the "recovery is manual" wording is wrong) **stands**.
+> The *second* half — "what blocks self-healing is the **absence of a re-trigger**; `warm-keycloak.service`
+> has no timer / no `Restart=`; it heals on the next **reboot / `nixos-rebuild switch`**" — is **WRONG**.
+> There IS an autonomous re-trigger: `nightly-sweep.timer` (`OnCalendar=Sun *-*-* 03:00:00`, `Persistent=true`)
+> → `nightly_sweep.roll_warm_infra()` → `subprocess warm_reconcile.py keycloak` (`WARM_APPS=["keycloak",…]`).
+> My probe was scoped to the wrong name: `list-timers | grep -i keycloak` is empty because the timer is called
+> `nightly-sweep`, and `git grep warm-keycloak -- nix/` never sees the sweep. **Do not use either command.**
+> Corrected dynamics, evidence, and remediation are in **A-redfix-4**, which supersedes this half.
+
+Severity **LOW-MED**, **no VETO, does not reopen the phase** (operator-facing description of a *deferred*
+item, not a DoD item). Merge target `b5f2b10` unchanged. Filed because the stated mechanism is wrong and the
+wrong mechanism points the operator at the wrong remediation.
+
+**What STATUS-redfix.md (`68b51d5`, lines ~91-94) claims:**
+> "**The wedge does not self-heal.** … A raise at `:514` or `:536` therefore escapes to `SystemExit` with
+> live keycloak already undeployed by the preceding `abra.undeploy()` — **no code path redeploys it.
+> Recovery is manual.**"
+
+**The conclusion "does not self-heal" is right; the reason given is wrong.** There *is* a code path that
+redeploys it, and it runs automatically.
+
+**Repro (static, from any clone; no node access needed):**
+
+    git show b5f2b10:runner/warm_reconcile.py | sed -n '468,479p'   # the fresh-deploy branch
+    git show b5f2b10:runner/harness/abra.py   | sed -n '295,298p'   # undeploy does NOT remove the .env
+    sed -n '38,45p' nix/modules/warm-keycloak.nix                   # oneshot, RemainAfterExit, no Restart=
+    git grep -n "warm-keycloak" -- nix/ | grep -iE "timer|OnCalendar|Restart="   # -> empty
+
+**Chain, each leg verified:**
+1. `abra.undeploy()` runs only `abra app undeploy` — it does **not** remove `~/.abra/servers/default/<domain>.env`.
+2. So on the next `reconcile("keycloak")`: `current_version(domain)` still resolves from `TYPE=` in that
+   `.env`, while `is_deployed(domain)` is `False` (no docker services for the stack).
+3. `warm_reconcile.py:471-479` therefore takes the **fresh-deploy branch**: `target = current or latest` →
+   `deploy_version(...)` → `wait_healthy` → `write_last_good` → `return deployed-fresh:<target>`.
+   **This branch never calls `warmsnap`**, so `_assert_slot_not_foreign` is never reached and the foreign
+   meta cannot block it. Live keycloak comes back up.
+
+**What actually prevents self-healing** is not the absence of a redeploy path but the absence of a
+**re-trigger**: `warm-keycloak.service` is `Type=oneshot`, `RemainAfterExit=true`, `wantedBy=multi-user.target`,
+with **no timer and no `Restart=`** (the module comment says it "converges every activation/boot"). Nothing
+re-runs it between activations.
+
+**Why this matters operationally — the failure is intermittent, not sticky.**
+- The wedge persists only until the next **reboot or `nixos-rebuild switch`**, at which point the unit
+  re-activates, hits the fresh-deploy branch and silently restores live keycloak. "Recovery is manual" will
+  mislead an operator into hand-redeploying something that a routine rebuild already fixed.
+- But the foreign `snapshot/meta.json` is **still in the slot**. So the box heals, then **re-wedges the next
+  time an upgrade is actually due** (`current != latest` → upgrade path → `:514` `snapshot()` → guard raises
+  → `abra.undeploy()` at `:512` has already run). Heal → re-wedge → heal: it will read as a *flake*.
+- Therefore the correct remediation is **remove the foreign `snapshot/` from `/var/lib/ci-warm/keycloak/`**,
+  not "redeploy keycloak". Redeploying treats the symptom and the wedge returns on the next due upgrade.
+
+**Ask of the Builder:** correct the two sentences in STATUS. Suggested: *"The wedge does not self-heal between
+activations — `warm-keycloak.service` is a `oneshot` with no timer and no `Restart=`. The next reboot or
+`nixos-rebuild switch` re-activates it and the fresh-deploy branch (`:471-479`, which never calls `warmsnap`)
+does restore live keycloak — but the foreign meta remains, so it re-wedges on the next due upgrade. Remediation
+is to remove the foreign `snapshot/`, not to redeploy."* The **precondition itself is unchanged and still
+correct**; only the mechanism/remediation sentences are wrong.
+
+### A-redfix-3 [adversary] — **CLOSED 2026-07-09T08:59Z** (accepted + evidence retracted at `f64d102`; STATUS:28-29,564-572 re-read and correct) — STATUS's `main.go` "present in both clones ⇒ written outside git" evidence is void: `/srv/cc-ci` is a **symlink** to `/srv/cc-ci-orch`, so there is one file, not two
+
+Severity **INFO**. No VETO, no gate impact, no DoD impact. The *conclusion* (untracked, not from this phase,
+leave it alone) is fine; the supporting evidence is an artifact.
+
+**What STATUS-redfix.md (`42bc4e4`, lines ~470-472) claims:**
+> "`/srv/cc-ci-orch/cc-ci/main.go` **and** `/srv/cc-ci/cc-ci/main.go` — present in **both** clones, identical
+> size, identical mtime … i.e. written by something outside git (**a git operation cannot leave the same
+> untracked file in two clones**)."
+
+**Repro:**
+
+    ls -la /srv/                                 # cc-ci -> /srv/cc-ci-orch   (symlink)
+    stat -c '%d:%i  %n' /srv/cc-ci-orch/cc-ci/main.go /srv/cc-ci/cc-ci/main.go
+    # -> 2049:3254604 for BOTH; stat -c 'links=%h' -> links=1
+
+Same device, same inode, link count 1: **one file, one clone, seen through a symlink.** The same holds for the
+Adversary clones (`/srv/cc-ci-orch/cc-ci-adv` and `/srv/cc-ci/cc-ci-adv` are inode `3206945`). The "two clones"
+observation carries **zero** information about the writer, and the identical size/mtime are tautological.
+
+**Risk bounded (read-only):** contents are a 281-byte hello-world `net/http` listener; `sha256
+bdbc3bf167cd20f30c00880005f4f994f17f3722660973c9c65c7bf33e81ffaf`; no `go.mod`; **`go` is not installed**;
+**nothing is listening on `:8080`**; `git log --all -- main.go` is empty; not gitignored; unreferenced by any
+tracked file. It cannot execute. Present in the **Builder's** clone only — **absent from both Adversary
+clone paths**, which is consistent with a single stray write into one working tree.
+
+**Ask of the Builder:** drop the "both clones ⇒ outside git" inference (keep the file, keep the flag). Agreed
+on not deleting it: neither of us created it.
+
+---
+
+### A-redfix-4 [adversary] — **CLOSED 2026-07-09T08:59Z** — all three asks applied at `e356698` and cold-re-verified by me at wake #45 (probes deleted; oscillation + silence recorded; precondition re-anchored to "never deploy `07fc6d4`"). The underlying *behaviour* remains open under **B-redfix-5**, which is deferred, not fixed. — the B-redfix-5 wedge has a **weekly autonomous re-trigger** (`nightly-sweep.timer`) and fails **silently**: post-arming, keycloak oscillates down-a-week / up-a-week with `nightly-sweep.service` reporting `Result=success`
+
+Severity **MED** (operator-facing; describes a *deferred* item's blast profile, **not** a DoD item).
+**No VETO. Does not reopen the phase.** M1 + M2 PASS stand. Merge target `b5f2b10` unchanged and still correct.
+Supersedes the "absence of a re-trigger" half of **A-redfix-2** (my own error) and, with it, the text the
+Builder accepted into STATUS-redfix.md at `f64d102`.
+
+**Both prior descriptions of the wedge are wrong, in opposite directions.**
+- Builder (`68b51d5`): "no code path redeploys it. **Recovery is manual.**" → wrong; it recovers unattended.
+- Me (`b358b7e`, A-redfix-2): "no re-trigger … heals on the next **reboot / `nixos-rebuild switch`**" → wrong;
+  it re-triggers **weekly**, and an unrelated `switch` does *not* heal it.
+
+**The re-trigger.** `warm_reconcile.py` has **two** invokers, not one. The second is the sweep:
+
+    nightly-sweep.timer   OnCalendar=Sun *-*-* 03:00:00, Persistent=true
+      └─ nightly-sweep.service → /etc/cc-ci/runner/nightly_sweep.py   (CCCI_REPO=/etc/cc-ci)
+           └─ main():147 roll_warm_infra()          # unconditional after the _another_run_active() guard
+                └─ for app in WARM_APPS = ["keycloak", "traefik"]:   # nightly_sweep.py:39,57
+                     subprocess.run([python, warm_reconcile.py, app])   # nightly_sweep.py:59-61
+
+**Observational proof that the sweep — not `warm-keycloak.service` — reconciles keycloak** (cc-ci, cold):
+
+    systemctl show warm-keycloak.service -p ExecMainStartTimestamp   # → Wed 2026-06-17 17:29:31 UTC (once)
+    stat -c %y /var/lib/ci-warm/keycloak/last_good                   # → 2026-07-05 03:04:51.209 +0000
+    systemctl list-timers --all | grep nightly-sweep                 # → last trigger Sun 2026-07-05 03:04:50
+
+`write_last_good()` is called **only** from `reconcile()`. The slot file was written **18 days after** the unit
+last ran, and **1.2 s after** the sweep fired. ⇒ `reconcile("keycloak")` ran from the sweep. (`journalctl -u
+nightly-sweep.service` retains 1 line — the empty grep is *retention*, not absence. Do not read it as evidence.)
+
+**Why both of our probes missed it — the tell.** The invoking timer's name never contains the app name:
+
+    systemctl list-timers --all | grep -i keycloak        # → EMPTY, yet a timer DOES drive it
+    git grep -n warm-keycloak -- nix/ | grep -iE 'timer|Restart='   # → EMPTY, yet a timer DOES drive it
+
+Both are **unfalsifiable by construction**. STATUS-redfix.md (`f64d102`) currently prescribes *both* as the
+verification commands for "no re-trigger". They must be removed. The sound probe is to grep for the **callee**:
+
+    git grep -n "warm_reconcile" -- runner/ nix/ | grep -v "^runner/warm_reconcile.py"
+
+Same class as the `e3b0c44298fc1c14` empty-input tell (B-redfix-8) and the `restore()`/`docker` ordering tell:
+a probe that returns the "all clear" value for a reason unrelated to the property under test.
+
+**Corrected dynamics** (only once armed — i.e. only if a foreign `snapshot/meta.json` sits in the slot **and**
+an upgrade is pending, `current != latest`):
+
+| sweep | state on entry | branch taken | outcome |
+|---|---|---|---|
+| N   | deployed, upgrade pending | stateful upgrade `:511-514` → `abra.undeploy` → `snapshot()` → guard raises | **keycloak DOWN** |
+| N+1 | not deployed | fresh-deploy `:471-479` → redeploys `current`; **never calls `warmsnap`** | keycloak UP (old version) |
+| N+2 | deployed, upgrade still pending | stateful upgrade again → guard raises | **keycloak DOWN** |
+
+⇒ **keycloak alternates DOWN one week / UP the next**, indefinitely, until the foreign `snapshot/` is deleted
+or `current == latest`. Not a one-shot outage, and not manual-recovery-required.
+
+**And it is silent.** `roll_warm_infra()` captures the subprocess rc and **ignores it** (`nightly_sweep.py:59-63`
+— `print(f"nightly: reconcile {app} rc={rc}")`, no raise), so a wedged reconcile does **not** fail
+`nightly-sweep.service` (`Result=success`). The guard raises at `:514`, which is **outside** the try/except at
+`:520-525` (that is B-redfix-5) and **upstream of every `write_alert()`** — so **no alert is written**. The only
+trace is a journal line `nightly: reconcile keycloak rc=1`, in a unit whose output is not retained.
+**A weekly production outage of the shared SSO provider would surface nowhere.**
+
+**Good news — the trap cannot arm itself.** Arming requires the canonical seed to run from a tree that has the
+keycloak enrollment but **not** the fix. The enrollment is **branch-only**:
+
+    git show origin/main:tests/keycloak/recipe_meta.py | grep WARM_CANONICAL   # → False   (= deployed /etc/cc-ci)
+    git show 07fc6d4:tests/keycloak/recipe_meta.py     | grep WARM_CANONICAL   # → True    (enrollment, NO fix)
+    git show b5f2b10:tests/keycloak/recipe_meta.py     | grep WARM_CANONICAL   # → True    (enrollment + fix)
+
+The deployed checkout `/etc/cc-ci` carries `WARM_CANONICAL = False`, so **no autonomous node activity — including
+Sunday's sweep — can write foreign meta into `/var/lib/ci-warm/keycloak/`.** Verified: the slot holds only
+`last_good` (no `snapshot/`, no `canon-*`), and keycloak is currently deployed and healthy (`_app` + `_db`).
+
+**⇒ The precondition is self-maintaining, and its landmark in STATUS is wrong.** It is not "the slot must be clean
+*at merge time*" (a `git merge` executes nothing). The binding statement is:
+
+> **Never deploy `07fc6d4`** — or any tree carrying the enrollment without `canonical_ns()`. That intermediate
+> state is the *only* thing that can arm the trap. Merging `b5f2b10` ships enrollment **and** fix together, so
+> canonical seeds land in `canon-keycloak/` and the slot never goes foreign.
+
+**Secondary correction (`nixos-rebuild switch`).** `warm-keycloak.service`'s `ExecStart` embeds the runner by
+store path — `/nix/store/…-runner/warm_reconcile.py`, and `warmsnap.py` lives in that **same** derivation
+(`…-runner/harness/warmsnap.py`). So a switch restarts the unit **iff `runner/**` changed**. A switch that
+touches anything else (e.g. sshd config) leaves the store path identical and does **not** re-run reconcile —
+so "heals on `nixos-rebuild switch`" is false in general. Deploying the merge *does* change `runner/**`, which
+means **the merge's own activation switch is itself a reconcile trigger**, alongside the Sunday sweep.
+
+**Asks of the Builder (STATUS-redfix.md text only — no code change, nothing reopens):**
+1. Delete both `grep -i keycloak` / `git grep warm-keycloak -- nix/` "no re-trigger" probes; they cannot fail.
+2. Replace "recovery is manual" **and** "heals on reboot / `nixos-rebuild switch`" with the weekly
+   down/up oscillation above, and state that it is **silent** (`Result=success`, no `write_alert`).
+3. Re-anchor the merge precondition from "slot clean at merge time" to "**never deploy `07fc6d4`**".
+
+Remediation if ever armed is unchanged and still correct: **delete the foreign `snapshot/` from the slot**;
+do not redeploy keycloak (the sweep does that for you, and it is what masks the fault).
+
+---
+
+### A-redfix-1 — ADDENDUM (wake #45, 2026-07-09T08:59Z): the embedded value is the **same live credential as B-redfix-8**, confirmed at value level; and rotation will **not** clean this file
+
+Still **OPEN**. Re-verified cold on cc-ci this wake, without printing the secret:
+
+    ssh cc-ci 'stat -c "%a %U:%G" /etc/cc-ci/.git/config'                     # → 644 root:root
+    ssh cc-ci 'grep -oE "https://[^/@]+:[^/@]+@" /etc/cc-ci/.git/config \
+                 | sed -E "s#https://[^:]+:##; s#@$##" | tr -d "\n" | sha256sum | cut -c1-16'
+
+**EXPECTED / observed: `3fcea78925015fc9`** — byte-identical to the B-redfix-8 rotation sentinel
+(`sha256(GITEA_PASSWORD)[:16]`, re-confirmed this wake from `/srv/cc-ci/.testenv` on the **orchestrator**).
+So A-redfix-1 and B-redfix-8 are **two exposures of one still-unrotated, push-capable credential**, and the
+sentinel matching proves it is **still not rotated**. (Extract-then-hash, never echo. `python3` is absent on
+cc-ci — do not run the orchestrator's python probe over `ssh`; that is the `e3b0c44298fc1c14` empty-input tell.)
+
+**New, and operator-relevant: rotation does not fix this file, and may recreate the exposure.**
+`/etc/cc-ci` is a **manual `git clone`**, not nix-generated — `nix/hosts/cc-ci-hetzner/configuration.nix:7`
+documents `git clone --recursive https://git.autonomic.zone/recipe-maintainers/cc-ci.git /etc/cc-ci`. Nothing
+regenerates `.git/config` on `nixos-rebuild switch`. Therefore:
+- After rotating `GITEA_PASSWORD`, the **old** value persists verbatim in this 0644 file (inert, but it should
+  still be scrubbed — it is the value published at `14c7dee`).
+- If the operator re-clones or re-embeds the **new** password in the remote URL, the 0644 exposure **returns**.
+
+**Recommendation (operator, alongside the B-redfix-8 rotation):** point the remote at a credential-less URL
+(`https://git.autonomic.zone/recipe-maintainers/cc-ci.git`) and authenticate via a `credential.helper` /
+token file with `0600`, or at minimum `chmod 0600 /etc/cc-ci/.git/config`. Do not carry userinfo in the URL.
+
+**Rotation blast radius is small** (checked, so this is not a reason to delay): `GITEA_PASSWORD` is consumed
+only by `scripts/bootstrap-drone-oauth.sh` (a one-off bootstrap). `scripts/recipe-mirror-sync.sh` pushes with
+an **OAuth token**, not the password, so the weekly sweep's mirror sync does **not** depend on it.
+
+**CORRECTION (wake #55): the "small blast radius" reasoning is HALF WRONG, and the OAuth token is ALSO
+exposed.** The above argues rotation is safe because mirror-sync uses a token, not the password — but that
+same `oauth2` token (`recipe-mirror-sync.sh:39`, `https://oauth2:${TOKEN}@…`, PUSH-capable) is itself leaked
+in **117** world-readable `.git/config` copies under `/var/lib/cc-ci-runs` (sentinel `9c44a1aea2ecb389`;
+verified live: `GET /api/v1/user` → 200 `autonomic-bot`/id 64). So there are **two** live push-capable
+credentials to rotate, and rotating `GITEA_PASSWORD` alone leaves the token exposed. The token rotation DOES
+have a real dependent (mirror-sync) — so it must be re-minted and the new value delivered to mirror-sync's
+secret source, not merely revoked. See STATUS-redfix.md steps 3–4 (updated) for the combined remedy.
+
+### A-redfix-1 — ADDENDUM (wake #53, 2026-07-09T10:15Z): the "sole copy" claim is FALSIFIED — **78** world-readable copies, root-caused to a git `insteadOf` rewrite that regenerates them on every CI run
+
+The original finding asserted (see above, line ~47): "`/etc/cc-ci` is the **sole** copy whose parents are
+world-traversable." I disbelieved that universal and probed it cold on cc-ci. **It is false.** Enumerating
+every cred-bearing `.git/config` and testing each for uid-1000 readability (extract-then-hash, never echo):
+
+    ssh cc-ci 'find / -xdev -name config -path "*/.git/*" 2>/dev/null | while read f; do
+      grep -q "autonomic-bot:" "$f" || continue
+      setpriv --reuid=1000 --regid=1000 --clear-groups cat "$f" >/dev/null 2>&1 && echo "$f"; done'
+    # → 78 world-readable copies: 68 under /var/lib/cc-ci-runs/*/abra/recipes/*/.git/config,
+    #   8 in /nix/store/*-source/.git/config (0444, read-only fs), /tmp/v/.git/config, /etc/cc-ci/.git/config
+
+All 78 carry the **same** live credential — `sha256(pw)[:16] = 3fcea78925015fc9`, identical to the B-redfix-8
+sentinel (empty-input control `e3b0c44298fc1c14` to prove the extractor fires). So the exposure is **78×**, not
+1×, and `/etc/cc-ci` is not special.
+
+**Root cause (grep the callee, not the value).** `run_recipe_ci.py:360` clones a **clean, credential-less**
+URL (`https://git.autonomic.zone/{src}.git`). The userinfo is injected by a global git rewrite in
+`/root/.gitconfig`:
+
+    url.https://autonomic-bot:<pw>@git.autonomic.zone/.insteadOf = https://git.autonomic.zone/
+
+Every per-run recipe clone the harness makes therefore bakes the cleartext password into its `.git/config` at
+mode **0644**, under `/var/lib/cc-ci-runs/<rid>/abra/recipes/<recipe>/` (world-traversable parents). This is
+the **generator**: the 68 run-dir copies accrete one-per-recipe-per-run and **regenerate autonomously on the
+next CI run** — not just on a manual re-clone as the prior addendum supposed.
+
+**Consequence for the remedy.** The prior recommendation (`chmod 0600 /etc/cc-ci/.git/config`, or delete
+`/etc/cc-ci`) fixes **1 of 78** and does nothing about regeneration. A durable operator fix must remove the
+credential from git's URL layer, e.g.:
+- replace the `insteadOf`-with-userinfo rewrite in `/root/.gitconfig` with a `credential.helper` (token in a
+  `0600` file), so harness clones carry **no** userinfo; and
+- scrub the existing run-dir + `/etc/cc-ci` + `/tmp/v` copies (the `/nix/store` ones are 0444 on a read-only
+  store and clear on GC).
+
+Otherwise the next `!testme`/nightly-sweep run re-exposes the (even freshly-rotated) value at 0644.
+
+**Scope note — NOT a gate, NOT a VETO, DONE stands.** The dashboard "no secrets" invariant was independently
+**re-verified holding** this wake: `/runs/<rid>/abra/recipes/<recipe>/.git/config` → **404** on the live
+dashboard (blocked by the `len(parts)==2` guard + the `_RUN_FILES` allow-list; positive control
+`/runs/985/results.json` → 200, `has_cred=False`). This exposure is **host-local filesystem only** (any local
+uid, incl. containers sharing the host mount), which is squarely the operator-only A-redfix-1/B-redfix-8
+territory. It **widens** their remediation scope; it does not reopen any D-gate or DoD item.
+
+### A-redfix-1 — ADDENDUM (wake #55, 2026-07-09T10:4xZ): wake-#53 `insteadOf` root cause RETRACTED; real generator = the sweep's `CCCI_SKIP_FETCH` copytree; a SECOND credential (oauth2 token) also exposed
+
+Adjudicated the Builder's wake-#54 rebuttal cold. Corrections to my own wake-#53 addendum above:
+
+- **RETRACT the `insteadOf` root cause.** Reproduced first-hand on the node (git 2.47.2): a clone
+  whose transport `url.<t>.insteadOf`-rewrites still stores the **original** URL; `insteadOf` injects
+  no userinfo when the clone URL is credential-less. My "regenerates via insteadOf" was wrong.
+- **Real generator:** canonical `/root/.abra/recipes/*/.git/config` carry the cred (safe at rest under
+  `/root` = `0700`); `run_recipe_ci.py:348-353` `CCCI_SKIP_FETCH=1` `copytree`s them into
+  `/var/lib/cc-ci-runs` (`0755`), dropping the protection.
+- **The exposure IS autonomously regenerated — weekly, by the sweep** (correcting the Builder's
+  "manual-* = hand-runs" claim). `run_id()`→`"manual"` for any non-Drone run; `nightly_sweep.py:88`
+  runs `run_recipe_ci.py` outside Drone with `CCCI_SKIP_FETCH=1`, so the sweep's runs are the
+  `manual-<pid>` dirs. Freshest copies dated 2026-07-05 03:37–03:59Z = the sweep's last fire.
+- **SECOND credential found (Builder census missed it).** `/var/lib/cc-ci-runs` holds **123**
+  world-readable cred-bearing configs: **68** `autonomic-bot/3fcea78925015fc9` (B-redfix-8 password)
+  **+ 55** `oauth2/9c44a1aea2ecb389` (a distinct OAuth2 token, sourced from bluesky-pds,
+  custom-html-tiny, gitea, mumble). Remedy must scrub/rotate **both**.
+
+Still **OPEN**, still **operator-scope**, **no gate / no DoD impact, no VETO**. Endorse the Builder's
+proposed remedy (strip userinfo from canonical origins + scrub + `chmod 0750 /var/lib/cc-ci-runs`)
+with the two amendments above (both credentials; canonical-origin fix is required because a one-time
+scrub is re-armed by the next sweep). Full reasoning + reproductions in REVIEW-redfix.md wake #55.
+
+### B-redfix-8 — ADDENDUM (wake #58, 2026-07-09): exposure is **two published commits**, not one; and the `oauth2` token is **not** in git history
+
+Triggered by the Adversary's wake-#57 break-it probe (`5431252`), which verified the token does not leak to
+the dashboard / weekly reports / Drone log API — but did **not** probe the **public git mirror**, which is the
+surface B-redfix-8 is actually about. Closing that gap surfaced a defect in my own STATUS.
+
+**Method.** Exhaustive scan of the whole object db (`git cat-file --batch-all-objects` → **3099 blobs**),
+literal-value match, `awk`-free (no `awk` on the orchestrator; an `awk`-based pipeline silently yields an
+empty blob list and a **vacuous** `0 hits`). Token pulled root-only from a world-readable run-dir
+`.git/config` into a `chmod 600` file, verified `sha256[:16]=9c44a1aea2ecb389` (matches the Adversary's
+wake-#55 sentinel), `shred`-removed after. Positive control: the known-leaked password **must** be found.
+
+**Results.**
+- **password** (`sha16 3fcea78925015fc9`) → **2** blobs / **2** published commits:
+  `14c7dee`(blob `fd21fcb8`, 33408 B) and `223cc16`(blob `bcc31b55`, 34353 B). Both are ancestors of
+  `origin/main` ⇒ both mirror-served. Neither reachable from HEAD. `e99e2b3` and HEAD are clean.
+  ⇒ STATUS's "**only** the historical commit `14c7dee` serves it" was **wrong and understated the exposure**.
+  A history scrub targeting `14c7dee` alone would leave `223cc16` serving the live credential. STATUS corrected.
+- **oauth2 token** (`sha16 9c44a1aea2ecb389`) → **0** of 3099 blobs. The 271 blobs matching the *string*
+  `oauth2:` are `recipe-mirror-sync.sh:39` `https://oauth2:${TOKEN}@…` variable interpolation and machine-docs
+  prose citing the sha16 sentinel — no literal value.
+
+**Consequence for the operator remedy.** The two credentials are **not symmetric**:
+- **password** — exposed on the **public mirror** (permanent, unauthenticated, in every clone) **and** on the
+  filesystem. Rotation is urgent; scrubbing must cover **both** commits, and cannot recall existing clones.
+- **oauth2 token** — exposed on the **filesystem only** (world-readable `.git/config` copies, A-redfix-1),
+  regenerated weekly by the sweep. Rotation + `chmod`/userinfo-strip suffices; no history rewrite needed.
+
+**Guard for future probes.** Any scan of this repo reporting **0** password-bearing blobs is a **broken probe**
+(missing `awk`, empty sentinel `e3b0c44298fc1c14`, or names-only grep of tracked paths). The correct answer
+is **2**. Always run the positive control.
+
+Still **operator-scope, outside DoD, no gate impact.** `## DONE` stands; no VETO.
+
+### B-redfix-8 — ADDENDUM 2 (wake #58b, 2026-07-09): blob-count figure was brittle; the invariant is 2-and-0, not a total
+
+The Adversary's wake-#58 adjudication (`63f7001`) independently CONFIRMED both results (token `0` blobs,
+password exactly `2` — `fd21fcb8@14c7dee` + `bcc31b55@223cc16`, both mirror-served ancestors) — but reported
+its total as **3061 blobs** where I reported **3099**. Neither is wrong; the *total* is simply not an invariant.
+
+Root-caused, three independent sources of drift:
+1. **New commits add blobs.** Same host now reads **3104** (my wake-#58 commits landed after the scan).
+2. **`--batch-all-objects` counts unreachable objects** that `git rev-list --all` omits — **49** of them here.
+   Reachable-only is **3026**.
+3. **`git clone /local/path` hardlinks the entire object store**, so a *local* clone inherits those unreachable
+   objects (verified: my cold clone also reads 3104). A clone from the **remote** gets only reachable objects —
+   which is why the Adversary's number is lower, and why it is the right scope for a *mirror-exposure* question.
+
+**Consequence.** My STATUS quoted `# → 3099 lines` as the expected output of the documented reproduction. A
+future agent running that command will see a different number, and could read the mismatch as either a broken
+probe or a changed repo — both wrong. STATUS corrected: the reproduction now asserts `wc -l > 100` as a sanity
+floor and names the **durable invariants — password ⇒ exactly 2 blobs, token ⇒ exactly 0** — which held across
+all four scans (3099 / 3104 / 3026 / 3061) precisely because they are scope-independent.
+
+Lesson, same shape as the `awk` near-miss: **an incidental figure recorded as if it were an assertion becomes a
+future false alarm.** Record the invariant, not the measurement.
+
+No new exposure, no gate impact, operator-scope. `## DONE` stands; no VETO.

machine-docs/DECISIONS.md

@@ -1609,3 +1609,52 @@ trim job observed → adequate; revisit only if a cap is ever needed.
   `dashboard/dashboard.py` + `tests/unit/test_dashboard.py` would be reformatted — confirmed pre-existing
   at HEAD f68f1c5, outside the settings diff. Flagged for the dashboard owner / orchestrator; not fixed
   here (narrow scope).
+
+## Warm state is keyed by STACK NAMESPACE, not by recipe (phase redfix, F-redfix-4, 2026-07-09)
+
+**Settled.** `/var/lib/ci-warm/<ns>/` is a slot owned by exactly ONE deployed stack. `ns` comes from
+`canonical.canonical_ns(recipe)` for the data-warm canonical and `warmsnap.live_slot(recipe)` for the
+live-warm reconciler. A recipe that is both (only `keycloak` today, via `warm.WARM_DOMAINS`) gets
+`canon-<recipe>` for its canonical, so the two never share a slot. The canonical's **domain and its slot
+derive from the same `canonical_ns()`**, so they cannot drift apart — that coupling is the invariant, not
+the `canon-` string. Every slot `<ns>` maps 1:1 to the stack at `warm.stable_domain(<ns>)`.
+
+Rejected alternatives:
+- **Skip `seed_canonical` for `WARM_DOMAINS` recipes.** Silently de-enrolls keycloak and re-opens the DoD
+  item ("keycloak enrolled … verified green"). The Adversary named this as not acceptable; agreed.
+- **Key every slot by the full domain / stack name** (the "purest" model). Correct, but it relocates the
+  reconciler's `last_good` for keycloak+traefik and `canonical.json` + `snapshot/` for 15 existing
+  canonicals on a live node, requiring a migration shim for a phase whose mandate is to fix red, not
+  re-architect. The chosen scheme is the same model applied only where two stacks actually contend, and
+  needs no migration (keycloak's canonical had never been seeded).
+- **Guard-only (`_assert_slot_not_foreign` alone).** Turns silent destruction into a loud failure but leaves
+  keycloak's canonical unable to ever seed. Kept as defence in depth BEHIND the slot separation, not as the
+  fix.
+
+Consequence for future enrollments: adding a recipe to `warm.WARM_DOMAINS` automatically namespaces its
+canonical on both axes. `prune_stale()`'s "reconciler dirs are never pruned" invariant is now structural
+(a `<recipe>/` dir never gains a `canonical.json`) rather than an incidental property that de-enrollment
+could falsify.
+
+## 2026-07-10 — Gitea credential rotation + git auth moved to SSH (operator-directed)
+
+Both leaked `autonomic-bot` credentials were rotated and the exposures scrubbed (was B-redfix-8 /
+A-redfix-1 / B-redfix-9, previously operator-scope):
+- **Account password** rotated (old value now returns HTTP 401). It had been world-readable on the
+  public mirror in commits `14c7dee` + `223cc16`. History was rewritten (`git filter-repo --replace-text`)
+  to redact it and force-pushed; `main` went `bfa5396 -> 0d8adba` (tree byte-identical, history-only).
+  All clones + node `/etc/cc-ci` were reset to the rewritten `main`. Backup of pre-rewrite repo saved at
+  `~/cc-ci-backups/pre-scrub-*` (mirror + bundle). **Caveat:** the orphaned commits are still fetchable
+  by exact SHA on the public mirror until git.autonomic.zone runs a site-admin GC (value is dead, so inert).
+- **`bridge_gitea_token`** (the oauth2 token, sops `secrets.yaml`) rotated: fresh PAT minted, re-encrypted,
+  deployed via `nixos-rebuild ?submodules=1`, old token + all 8 other stale PATs revoked. Only one PAT
+  remains: `cc-ci-bridge-*` (id 36).
+
+**Git auth is now SSH, not HTTP.** git.autonomic.zone's Gitea only honours *preemptive* HTTP Basic auth;
+credential-helper (challenge-response) always 401s on private repos — do not rely on it. Remotes are SSH
+(`git@git.autonomic.zone:...`, port 2222 via `~/.ssh/config`); dedicated keys `cc-ci-orchestrator-20260710`
+(orchestrator, `~/.ssh/autonomic-bot-gitea-ed25519`) and `cc-ci-node-20260710` (node, `/root/.ssh/...`).
+`.gitmodules` secrets URL is SSH. `recipe-mirror-sync.sh` pushes over SSH (no token written into recipe
+`.git/config` -> fixes the copytree leak). The bot **password stays in `/srv/cc-ci/.testenv` for API use only**
+(comments / PR / repo creation); the weekly Thursday upgrade run picks it up by sourcing `.testenv`.
+Node run-tree `/var/lib/cc-ci-runs` chmod 0750; 144 world-readable cred-bearing config copies scrubbed.

machine-docs/DEFERRED.md

@@ -427,3 +427,29 @@ reachable via the operator/dev STAGES escape — production drone runs always ru
   key on stdout (write to a run-scoped sidecar the test reads), or register the minted key in the harness
   redaction set so even the RAW log is scrubbed. Low priority (RAW log is access-controlled; key is ephemeral).
 - **Filed by:** Builder, phase prevb (acknowledging Adversary [F-prevb-C]).
+
+## B-redfix-5 — warm reconciler's rollback `restore()` is outside the upgrade's `try/except`
+
+- **What:** in `runner/warm_reconcile.py`, **two** post-`abra.undeploy()` warmsnap calls sit OUTSIDE the
+  `try/except` guarding the upgrade — **(a)** the upgrade path `:512-514` (`undeploy` → `wait_undeployed` →
+  `warmsnap.snapshot(...)`) and **(b)** the unhealthy-rollback path `:534-536` (`undeploy` →
+  `wait_undeployed` → `warmsnap.restore(...)` → `deploy_version(last_good)` at `:537`). If either raises for
+  any reason (absent/corrupt snapshot, foreign slot, docker error, "no volumes found") the exception
+  propagates, the following `deploy_version` never runs, and live keycloak is left **undeployed**, taking
+  down the shared OIDC provider `lasuite-*`/`drone` depend on. Site (a) fires on the *normal* upgrade path —
+  no rollback required.
+- **Why deferred, not fixed:** structural gap, present at `07fc6d4` and predating the keycloak enrollment.
+  Not part of F-redfix-4's published clearing condition; the Adversary explicitly concurred it is **not a
+  VETO** and was "correctly filed rather than silently fixed".
+- **Reachability, corrected (wake #42).** The earlier text here said F-redfix-4 "supplied the only
+  *reachable* trigger and that is now closed". That is wrong in one direction: `b5f2b10` *adds* a raise path
+  — `_assert_slot_not_foreign()` is called inside BOTH `warmsnap.snapshot()` (`:158`) and `warmsnap.restore()`
+  (`:209`). What makes the composition unreachable **today** is that cc-ci's live slot
+  `/var/lib/ci-warm/keycloak/` holds no `snapshot/meta.json` at all (`read_meta` → `None`; an unclaimed slot
+  is free to claim) — not the closure of F-redfix-4. The one state that writes a foreign domain into the
+  live slot is a **pre-fix (`07fc6d4`) canonical seed**. Recorded as a MERGE PRECONDITION in STATUS-redfix.md.
+- **Needed from operator:** decide the safe behaviour for a DB-backed app after a forward migration —
+  (a) redeploy `last_good` anyway even though data was not restored, or (b) die loudly + leave a breadcrumb
+  rather than start on unrestored data. That trade-off is a real safety call, not a mechanical fix, which is
+  why it was not settled inside a remediation commit.
+- **Filed by:** Builder, phase redfix (from Adversary F-redfix-4 consequence 3).

machine-docs/STATUS-redfix.md

@@ -7,7 +7,456 @@ gitea, keycloak) → isolate → root-cause → classify (flake vs genuine; reci
 warm-machinery vs load) → FIX each (recipe PR or harness improvement) → verify green. No standing
 exceptions. Nothing merged.
 
-## DONE — 2026-06-18T07:09Z
+## DONE — 2026-07-09T00:18Z
+
+Phase `redfix` COMPLETE. All six canon-sweep failures investigated in isolation, root-caused, classified,
+**FIXED — each via a recipe PR or a harness improvement — and verified green**; no recipe left as a standing
+exception; nothing merged (operator merges). Every gate has a fresh Adversary PASS in REVIEW-redfix.md and
+there is **no standing VETO**:
+- **M1 PASS** @ 2026-06-18T01:18Z (investigation/classification cold-verified).
+- **M2 PASS** @ 2026-07-09T00:18Z (F-redfix-4 remedy cold-verified; VETO CLEARED; supersedes the
+  2026-06-18T07:06Z M2 PASS, which was given against parent `07fc6d4`).
+
+Adversary findings F-redfix-1/2/3/4 are all **CLOSED**; no open blocking finding.
+
+The four `A-redfix-*` findings are also non-blocking and are accounted for here, so nothing dangles:
+- **A-redfix-1** (`/etc/cc-ci/.git/config` is 0644 and embeds the Gitea bot password) — **OPEN**, out of redfix
+  scope, pre-existing undeclared server state, no DoD item. Mirrors my B-redfix-7. Not created by this phase.
+  **It is the SAME credential as B-redfix-8, not merely a similar one** — verified first-hand 2026-07-09T09:1xZ:
+  the password extracted from the `.git/config` remote URL and the `GITEA_PASSWORD` in the orchestrator's
+  `/srv/cc-ci/.testenv` both hash to `3fcea78925015fc9` (neither is the empty-string artifact
+  `e3b0c44298fc1c14`). **Two exposures, one credential, still unrotated.** Folded into the B-redfix-8 remedy
+  below, because rotating the password does **not** scrub this file.
+- **A-redfix-2** (the B-redfix-5 wedge's self-heal mechanism) — **PARTIALLY WITHDRAWN by the Adversary at
+  wake #44.** Its first half stands (a code path *does* redeploy keycloak — the fresh-deploy branch). Its
+  second half ("no re-trigger; heals on reboot") is **false** and is superseded by A-redfix-4.
+- **A-redfix-3** (the `main.go` "both clones" evidence) — **ACCEPTED**; that evidence is retracted below.
+  `/srv/cc-ci` is a symlink to `/srv/cc-ci-orch`. The file's origin remains unexplained; risk inert.
+- **A-redfix-4** (a weekly `nightly-sweep.timer` *does* re-trigger reconcile; the wedge is **silent** and
+  **recurring**) — **ACCEPTED and independently re-verified against the node**, not taken on authority. The
+  B-redfix-5 precondition below is rewritten accordingly: two unfalsifiable probes deleted, the wedge
+  described as a weekly alternating DOWN/UP outage that raises no alert, and the precondition re-anchored
+  from "slot clean at merge time" to **"never deploy `07fc6d4`"**. The trap is **not armed** and **cannot be
+  armed by autonomous node activity** (the tree the sweep runs has `WARM_CANONICAL = False`).
+
+  **MECHANISM CORRECTED (wake #54; Adversary REVIEW wake #54, re-verified first-hand by the Builder).** The
+  sweep does **not** run `origin/main`. `nightly-sweep.service` ExecStart is a nix-store wrapper that sets
+  `CCCI_REPO=/etc/cc-ci` and execs `cc-ci-run "$CCCI_REPO/runner/nightly_sweep.py"`, so the tree that
+  executes is the **deployed checkout `/etc/cc-ci`** (branch `main` @ `d11f8f5`, clean). The guarantee is a
+  property of **what is deployed to `/etc/cc-ci`**, so a future **deploy** — not a merge to `origin/main` —
+  is what could arm the wedge. Conclusion unchanged (trap disarmed); only the named tree was wrong. Verify:
+
+      ssh cc-ci 'git -C /etc/cc-ci rev-parse --abbrev-ref HEAD; grep -n "^WARM_CANONICAL" /etc/cc-ci/tests/keycloak/recipe_meta.py;
+        git -C /etc/cc-ci merge-base --is-ancestor 07fc6d4 HEAD && echo ARMED || echo disarmed'
+      # EXPECTED: main / 16:WARM_CANONICAL = False / disarmed   (a real assignment, not the line-15 comment)
+
+All four corrected operator-facing *descriptions*, not DoD items. No VETO at wake #43 (`b358b7e`) or wake #44
+(`84a1666`, 2026-07-09T08:47Z). **Note on provenance:** A-redfix-2's false half entered this file at `f64d102`
+because I adopted the Adversary's stated mechanism *and its two probe commands* without re-deriving them. The
+probes returned the all-clear for a reason unrelated to the property under test (they grep the app name;
+the timer is named `nightly-sweep`). Both are now deleted, and every claim in the rewritten block below was
+re-checked first-hand.
+
+## ⚠ OPERATOR ACTION REQUIRED — B-redfix-8 (HIGH, open, outside DoD)
+
+Phase DoD is met, but **one HIGH item is open and only an operator can close it.** It does not block `##
+DONE` (no DoD item depends on it) and is recorded in full in BACKLOG-redfix.md.
+
+**WHAT.** The live Gitea bot password (`GITEA_PASSWORD` in `/srv/cc-ci/.testenv`) was committed to
+`machine-docs/BACKLOG-redfix.md`, pushed to `origin/main`, and is served **in cleartext to the
+unauthenticated public internet** by the public mirror. It is push-capable to `recipe-maintainers/*`. HEAD
+and the redaction commit `e99e2b3` are clean. Redaction cannot unpublish it — the value is permanent in
+history and in every clone.
+
+**CORRECTED (wake #58, 2026-07-09):** an earlier version of this entry said "**only** the historical commit
+`14c7dee` serves it." That **understated the exposure**. An exhaustive scan of every blob in the object db
+(`git cat-file --batch-all-objects`, literal-value match) finds the password in **two distinct
+blobs**, carried by **two published commits** — both ancestors of `origin/main`, so both are mirror-served:
+
+| commit | blob | `machine-docs/BACKLOG-redfix.md` size |
+|---|---|---|
+| `14c7dee` | `fd21fcb8ac02bff9917531c8175a29e450c41b19` | 33408 |
+| `223cc16` | `bcc31b55860f782056ba3ccdf13165191bd329b5` | 34353 |
+
+Neither blob is reachable from HEAD. Scrubbing only `14c7dee` would leave `223cc16` serving the credential.
+
+**WHERE.** `git.autonomic.zone/recipe-maintainers/cc-ci` →
+`/raw/commit/14c7dee/machine-docs/BACKLOG-redfix.md` **and**
+`/raw/commit/223cc16/machine-docs/BACKLOG-redfix.md`
+
+**HOW to confirm exposure** (unauthenticated fetch; a raw blob at a fixed sha is immutable, so served size
+must equal object size):
+
+    git cat-file -s 14c7dee:machine-docs/BACKLOG-redfix.md    # → 33408
+    git cat-file -s 223cc16:machine-docs/BACKLOG-redfix.md    # → 34353
+
+An unauthenticated fetch of **either** raw URL returning **HTTP 200** at the matching byte count is the
+leak, not an error page.
+
+**Exhaustive-scan reproduction** (names/counts only; never prints the value). The `awk`-free form matters —
+`awk` is absent on the orchestrator and an `awk`-based blob list yields an **empty** list, i.e. a vacuous
+`0 hits`:
+
+    git cat-file --batch-all-objects --batch-check='%(objecttype) %(objectname)' \
+      | grep '^blob ' | cut -d' ' -f2 > /tmp/blobs        # assert  wc -l > 100  before trusting
+    # then literal-match each blob against a chmod-600 sentinel; positive control MUST return 2
+
+**Do not assert an exact total blob count** — it is host- and time-dependent, and is *not* the invariant.
+Observed legitimately: **3099** (Builder wake-#58 scan), **3104** (same host, after that wake's commits),
+**3026** (reachable-only), **3061** (Adversary's cold clone, wake #58). Three sources of drift: (a) new
+commits add blobs; (b) `--batch-all-objects` counts **unreachable** objects (49 here) that `git rev-list`
+omits; (c) `git clone /local/path` **hardlinks the whole object store**, so a local clone inherits those
+unreachable objects while a clone from the *remote* receives only reachable ones. All four scans agree on
+the only figures that matter.
+
+**Positive control is mandatory.** The durable invariants, stable across every scan scope above:
+**password ⇒ exactly 2 blobs; token ⇒ exactly 0.** Any scan of this repo that reports **0** password-bearing
+blobs is a **broken probe**, not a clean repo.
+
+**The `oauth2` token (the second exposed credential) does NOT appear in git history.** Same exhaustive scan,
+same run, literal match against the live token (`sha256[:16]=9c44a1aea2ecb389`, extracted root-only to a
+`chmod 600` file, then `shred`-removed): **0 blobs** — independently reproduced by the Adversary at wake #58
+against its own cold clone. The 271 history blobs matching the string
+`oauth2:` are `scripts/recipe-mirror-sync.sh:39`'s `https://oauth2:${TOKEN}@…` **variable interpolation**
+plus machine-docs **prose citing the sha16 sentinel** — no literal value. So the token's exposure is
+**filesystem-only** (A-redfix-1: world-readable `.git/config` copies); it is **not** on the public mirror.
+The password's exposure is **both** filesystem **and** public mirror. The two credentials differ in blast
+radius; the operator remedy must not treat them as symmetric.
+
+**HOW to confirm it is still the live credential** (commits to the value without republishing it).
+Run this **on the orchestrator** — `/srv/cc-ci/.testenv` lives there, **not** on `cc-ci`:
+
+    python3 -c 'import hashlib,re;v=re.search(r"^GITEA_PASSWORD=(.*)$",open("/srv/cc-ci/.testenv").read(),re.M).group(1).strip().strip(chr(34)).strip(chr(39));print(hashlib.sha256(v.encode()).hexdigest()[:16])'
+
+**EXPECTED.** Prints `3fcea78925015fc9` while still unrotated. **A different digest ⇒ rotated ⇒ `14c7dee`
+is inert ⇒ close B-redfix-8** — *but only if the probe was sound.* Two known-bad probes yield a different
+digest without any rotation, and must NOT be used to close this item:
+
+- **`e3b0c44298fc1c14` is the empty-input tell, not a rotation.** Running the command over `ssh cc-ci`
+  fails to open `.testenv`, hashes the empty string, and prints `sha256("")[:16]` = `e3b0c44298fc1c14`
+  (check: `printf '' | sha256sum`). Fix the probe; do not close B-redfix-8.
+- **Do not conflate the three digests.** `3fcea78925015fc9` = sha256(credential *value*)[:16] — the only
+  rotation sentinel. `1994fd8d…` = sha256(the whole served *blob*) — immutable, changes never.
+  `e3b0c44298fc1c14` = sha256(empty) — a broken probe.
+
+Exposure confirmed at **value level** (Adversary, 2026-07-09T07:34Z): the currently-valid, push-capable
+password appears **verbatim** in the publicly served body, not merely "a blob that once held a secret."
+
+**REMEDY (operator).** Rotate the Gitea bot password, then update **both** places that hold it. The credential
+is a Class-A1 external infra input (plan §4.4) — the Builder must not rotate or invent it. History rewrite is
+not available to the Builder either (`--force` is forbidden by the standing rules), so rotation is the only
+action that actually revokes the exposure.
+
+1. Rotate the bot password in Gitea.
+2. Update `GITEA_PASSWORD` in `/srv/cc-ci/.testenv` **on the orchestrator**.
+3. **Rewrite `/etc/cc-ci/.git/config` on `cc-ci` (A-redfix-1) — rotation does NOT scrub it.** That file is
+   `644 root:root` and embeds the same password in the `origin` URL's userinfo. `/etc/cc-ci` is a **manual
+   `git clone`**, not nix-generated (`nix/hosts/cc-ci-hetzner/configuration.nix:7` documents the clone), so no
+   `nixos-rebuild` regenerates or scrubs it. Strip the userinfo from the URL rather than re-embedding the new
+   password — re-embedding recreates the exposure verbatim:
+
+       ssh cc-ci 'git -C /etc/cc-ci remote set-url origin \
+         https://git.autonomic.zone/recipe-maintainers/cc-ci.git && chmod 600 /etc/cc-ci/.git/config'
+
+   **CORRECTION (wake #54→#55, Builder, 2026-07-09): `/etc/cc-ci/.git/config` is 1 of many world-readable
+   copies of TWO live credentials — step 3 alone fixes 1.** The "sole copy" claim above is **withdrawn**.
+   Scope re-derived first-hand at wake #55 (a full-FS, per-file census — my wake-#54 count of 78 was BOTH
+   undercounted AND single-credential; Adversary flagged both, wake #55). Two distinct live credentials, each
+   its own `sha256(pw)[:16]` sentinel (empty-input control `e3b0c44298fc1c14`):
+
+   | credential | sentinel | world-readable copies (uid-1000-readable) |
+   |---|---|---|
+   | `autonomic-bot` **password** (B-redfix-8) | `3fcea78925015fc9` | 78 (68 run-dir + 8 `/nix/store` 0444 + `/tmp/v` + `/etc/cc-ci`) |
+   | `oauth2` **Gitea token** (NEW, wake #55) | `9c44a1aea2ecb389` | 117 (`/var/lib/cc-ci-runs`), sourced from 20 canonical clones |
+
+   Per-file under `/var/lib/cc-ci-runs`: **62 carry BOTH**, 16 password-only, 55 token-only → **133 distinct
+   world-readable cred-bearing files** there. **The `oauth2` token is LIVE and PUSH-CAPABLE** — verified
+   `GET /api/v1/user` → 200 `login=autonomic-bot id=64`, and `recipe-mirror-sync.sh:39` pushes with it
+   (`https://oauth2:<TOKEN>@…`). So **both** secrets must be rotated, not just `GITEA_PASSWORD`. Re-census:
+
+       ssh cc-ci 'find / -xdev -name config -path "*/.git/*" 2>/dev/null | while read f; do
+         setpriv --reuid=1000 --regid=1000 --clear-groups cat "$f" >/dev/null 2>&1 || continue
+         grep -qE "autonomic-bot:|oauth2:" "$f" && echo "$f"; done | wc -l'
+       # EXPECTED: ~135 today (133 in /var/lib + /tmp/v + /etc/cc-ci); after remediation: 0 (ignoring 0444 /nix/store)
+
+   Legacy list (password only), retained for the step-3 command below — the `origin` in these carries the
+   password sentinel `3fcea78925015fc9`:
+   **68** under `/var/lib/cc-ci-runs/manual-*/abra/recipes/*/`, **8** in `/nix/store` (0444, read-only, clears
+   on GC), **1** `/tmp/v`, **1** `/etc/cc-ci`. Verify:
+
+       ssh cc-ci 'find / -xdev -name config -path "*/.git/*" 2>/dev/null | while read f; do
+         grep -q "autonomic-bot:" "$f" || continue
+         setpriv --reuid=1000 --regid=1000 --clear-groups cat "$f" >/dev/null 2>&1 && echo "$f"; done | wc -l'
+       # EXPECTED: 78   (after remediation: 0, ignoring /nix/store)
+
+4. **Scrub the copies AND stop regeneration (A-redfix-1, widened).** The generator is the **credentialed
+   remotes in the canonical clones** `/root/.abra/recipes/*/.git/config` (password on `origin`, oauth2 token
+   on the `gitea`/mirror remote), which `run_recipe_ci.py:348-353`'s `CCCI_SKIP_FETCH=1` `shutil.copytree`
+   copies into the world-traversable run tree (`/var/lib/cc-ci-runs`, `0755`). Canonicals are shielded by
+   `/root` = `0700`; the run-dir copies are not.
+   **REGENERATES WEEKLY (corrected wake #55).** The `manual-*` run dirs are NOT hand-run leftovers: the
+   autonomous `nightly-sweep.timer` runs `run_recipe_ci.py` outside Drone (`nightly_sweep.py:88` sets
+   `CCCI_SKIP_FETCH="1"`), and `run_id()` labels any non-Drone run `manual-<pid>` (`run_recipe_ci.py:318-319`).
+   So the sweep re-creates these copies every fire (freshest copies dated 2026-07-05 03:37–03:59Z = the
+   `LastTriggerUSec` 2026-07-05 03:04:50Z sweep). **A one-time scrub is re-exposed on the next sweep unless
+   the canonical origins are stripped or `/var/lib/cc-ci-runs` is hardened to `0750` durably.** Strip **both**
+   credentials at the source, then scrub the copies:
+
+       ssh cc-ci 'for d in /root/.abra/recipes/*/; do r=$(basename "$d");
+         for rem in origin gitea upstream; do
+           git -C "$d" remote get-url "$rem" >/dev/null 2>&1 || continue
+           git -C "$d" remote set-url "$rem" "$(git -C "$d" remote get-url "$rem" | sed -E "s#://[^@/]+@#://#")"; done; done
+       rm -rf /tmp/v
+       find /var/lib/cc-ci-runs -path "*/.git/config" \( -exec grep -lq "autonomic-bot:" {} \; -o -exec grep -lq "oauth2:" {} \; \) -delete
+       chmod 0750 /var/lib/cc-ci-runs'
+
+   Stripping the userinfo does **not** break the clone: the sweep only ever *fetches*, and the mirror serves
+   this repo anonymously (that is precisely what B-redfix-8 exploits). Verified 2026-07-09T09:1xZ —
+   `GIT_TERMINAL_PROMPT=0 git -c credential.helper= ls-remote https://git.autonomic.zone/recipe-maintainers/cc-ci.git HEAD`
+   → `f0372b8… HEAD`, rc=0, no credential. The sibling `cc-ci-secrets` submodule URL (`.git/config:15`)
+   already carries no userinfo, so this matches the existing pattern.
+
+   **Independently cold-verified by the Adversary (REVIEW-redfix.md wake #46, `3f27cc6`), which attempted to
+   refute it three ways and failed on all three** — so this step is safe to run on the node:
+   - **No push through `origin`.** The only `git push` in the deployed tree (`scripts/recipe-mirror-sync.sh:84-85`)
+     `remote add`s a `gitea` OAuth-token remote inside a *recipe* clone; it never touches `/etc/cc-ci`'s origin.
+   - **The `secrets` submodule survives the strip.** Its URL already carries no userinfo and
+     `cc-ci-secrets.git` is anonymously fetchable (`ls-remote` rc=0), so `submodule update` still works.
+   - **Nothing auto-pulls the checkout.** `/etc/cc-ci` is clone-once + local `nixos-rebuild switch`; the sweep
+     never `git pull`s it. (`/root/.git-credentials` exists but no `credential.helper` is wired at global or
+     system scope, so it is inert — the anonymous `ls-remote` results above are genuinely credential-free.)
+
+**Rotation blast radius is small — no operational reason to delay.** `GITEA_PASSWORD` is consumed by exactly
+one script, `scripts/bootstrap-drone-oauth.sh` (`git grep -ln GITEA_PASSWORD -- .` returns that script, `docs/
+install.md`, and machine-docs prose; nothing else in the runner). In particular `scripts/recipe-mirror-sync.sh`
+pushes with an OAuth **token** read from `/run/secrets/bridge_gitea_token` (`:30`), not this password, so
+mirror sync is unaffected by the rotation.
+
+**HOW to confirm BOTH exposures are closed** (each prints a digest; a value other than `3fcea78925015fc9` on
+both, and not the empty-string tell `e3b0c44298fc1c14`, means rotated):
+
+    # exposure 2 — the embedded copy; `python3` is absent on cc-ci, so use sha256sum
+    ssh cc-ci 'sed -n "s#^\s*url = https://[^:]*:\([^@]*\)@git\.autonomic\.zone/recipe-maintainers/cc-ci\.git#\1#p" \
+      /etc/cc-ci/.git/config | head -1 | tr -d "\n" | sha256sum | cut -c1-16'
+
+**EXPECTED after step 3:** the command prints **nothing** (no userinfo left to match) — an empty result here
+is the fix, whereas an `e3b0c44298fc1c14` from the *step-2* probe is a broken probe. Do not confuse them.
+
+**Merge target: `redfix-m2-harness` @ `b5f2b10`** (not `07fc6d4` — that tip carries the F-redfix-4 defect).
+The earlier `## DONE` of 2026-06-18T07:09Z was **withdrawn** on 2026-07-09 under the standing VETO
+F-redfix-4 and is re-asserted here only after the remedy was Adversary-verified. Details of that cycle,
+and the still-open non-blocking B-redfix-5, are below.
+
+One deferred, non-blocking item remains recorded (NOT part of any DoD item, NOT a standing finding):
+**B-redfix-5** in BACKLOG-redfix.md — in `warm_reconcile.py`, TWO post-`abra.undeploy()` calls sit outside
+the upgrade's `try/except`: `warmsnap.snapshot()` on the upgrade path (`:514`) and `warmsnap.restore()` on
+the rollback path (`:536`). If either raises, live keycloak is left **undeployed**, and because both sit
+upstream of every `write_alert()` call site the failure is **silent**. Pre-dates the enrollment (present at
+`07fc6d4`). Adversary concurred it is not a VETO. Its *consequence* is worse than first recorded — see the
+merge precondition below: reached only via a foreign live-slot meta, which nothing on the node can create
+today, but if ever reached it produces a weekly alternating SSO outage that reports `Result=success`.
+
+**MERGE PRECONDITION for `b5f2b10`.** The F-redfix-4 fix *adds* a raise path to both of those calls —
+`warmsnap._assert_slot_not_foreign()` (`warmsnap.py:158` in `snapshot()`, `:209` in `restore()`). It raises
+iff the live slot's `snapshot/meta.json` records a `domain` other than the one passed. Composed with
+B-redfix-5, a foreign live-slot meta wedges live keycloak: on the **upgrade** path this fires on *every*
+stateful auto-upgrade, not only on rollback. It is **unreachable on cc-ci today** because the live slot holds
+no snapshot at all (`read_meta` → `None` → "a slot with no snapshot yet is free to claim"). It is unreachable
+for that reason — **not** because F-redfix-4 is closed.
+
+**THE BINDING RULE: never deploy `07fc6d4`.** (Re-anchored 2026-07-09 per Adversary **A-redfix-4**; the
+earlier "verify the slot is clean at merge time" framing named the wrong landmark — *a `git merge` executes
+nothing*.) The only state that creates a foreign live-slot meta is a **canonical keycloak seed executed from
+a tree that carries the enrollment but not `canonical_ns()`**. Exactly one such tree exists: `07fc6d4`
+(`tests/keycloak/recipe_meta.py` → `WARM_CANONICAL = True`, no `canonical_ns()`). `b5f2b10` ships enrollment
+**and** fix together, so its canonical seeds land in `canon-keycloak/` and the live slot never goes foreign.
+The precondition is therefore **self-maintaining across the merge** — deploying the merge target cannot arm
+the trap; only deploying the intermediate `07fc6d4` can.
+
+**No autonomous node activity can arm it.** The tree the weekly sweep executes is `CCCI_REPO=/etc/cc-ci` =
+`origin/main`, where `WARM_CANONICAL = False`. Only `07fc6d4` and `b5f2b10` have `True`.
+
+**If it is ever armed, the wedge is SILENT and RECURRING — a weekly alternating outage of the shared SSO
+provider.** (Corrected 2026-07-09 per **A-redfix-4**, which withdrew the "no re-trigger / heals on reboot"
+half of A-redfix-2. My own earlier "recovery is manual" (`68b51d5`) was also wrong. Re-derived independently
+against the node — see HOW below.) `warm_reconcile.py` has **two** invokers, and the second is a weekly timer:
+
+    nightly-sweep.timer  (OnCalendar=Sun *-*-* 03:00:00, Persistent=true)
+      └─ nightly_sweep.main():147 → roll_warm_infra() → subprocess warm_reconcile.py keycloak
+         (WARM_APPS = ["keycloak", "traefik"], nightly_sweep.py:39,57-61)
+
+Once armed, with `current != latest`: **sweep N** → `abra.undeploy()` → `snapshot()` guard raises → keycloak
+**DOWN**. **Sweep N+1** (7 days later) → `is_deployed` is `False` → the **fresh-deploy branch**
+(`:471-479`), which **never calls `warmsnap`** → keycloak **UP** on the old version. **Sweep N+2** → upgrade
+path → **DOWN** again. **Alternating weeks, indefinitely**, because the foreign `snapshot/meta.json` is never
+removed. (`abra.undeploy()` (`abra.py:295-297`) runs only `abra app undeploy -n` and does **not** remove the
+app `.env`, so `current_version` stays resolvable — this is why the fresh-deploy branch is the one taken.)
+
+**And it raises no alert.** `roll_warm_infra()` captures the subprocess rc and discards it
+(`nightly_sweep.py:59-62` — `rc` is only ever printed), so `nightly-sweep.service` still reports
+`Result=success`. The raise at `:514` is outside the only `try/except` (`:520-525`, which wraps
+`deploy_version` + `wait_healthy` — *that gap is B-redfix-5*) and upstream of **every** `write_alert()` call
+site (`:493,500,503,539`). `reconcile()`'s sole caller `main():556` does not catch either. A weekly SSO
+outage would surface **nowhere**. (All line numbers here are `b5f2b10` coordinates, matching the rest of this
+block; on `main` the same sites sit two lines earlier.)
+
+**REMEDIATION if ever armed: delete the foreign `snapshot/` from the slot** (or fix B-redfix-5).
+**Redeploying keycloak does NOT fix it** — the next sweep re-wedges. This is the one part of the old text
+that was right, and it is the part that matters operationally.
+
+**HOW to verify** (from a clone at `b5f2b10`). Note the probe greps the **callee**, never the app name: the
+invoking timer is called `nightly-sweep`, so `list-timers | grep -i keycloak` and `git grep warm-keycloak --
+nix/` are both empty **while a timer drives it weekly**. Those two probes were in this file until 2026-07-09
+and are unfalsifiable by construction; they have been deleted, not fixed.
+
+    git grep -n warm_reconcile -- runner/ nix/            # SOUND probe: finds both invokers
+    git show redfix-m2-harness:runner/harness/abra.py     | sed -n '295,297p'   # undeploy: no .env removal
+    git show redfix-m2-harness:runner/warm_reconcile.py   | sed -n '471,479p'   # fresh-deploy branch, no warmsnap
+    ssh cc-ci 'systemctl cat nightly-sweep.timer | grep -iE "OnCalendar|Persistent"'
+    ssh cc-ci 'systemctl list-timers --all nightly-sweep.timer --no-pager'
+
+**EXPECTED:** `git grep warm_reconcile` prints both `nix/modules/warm-keycloak.nix:26` (the oneshot unit) and
+`runner/nightly_sweep.py:60` (the weekly sweep's subprocess); `undeploy()` is a single
+`_run(["app","undeploy",domain,"-n"], …)` with no `.env` handling; `:472` is `if not deployed:` returning
+`deployed-fresh:{target}` at `:479` with no `warmsnap` between; the timer prints
+`OnCalendar=Sun *-*-* 03:00:00` + `Persistent=true`.
+
+**Observational proof the sweep — not the unit — drives reconcile** (independent of reading any code;
+observed 2026-07-09):
+
+    ssh cc-ci 'systemctl show warm-keycloak.service -p ExecMainStartTimestamp
+               stat -c "%y %n" /var/lib/ci-warm/keycloak/last_good
+               systemctl list-timers --all nightly-sweep.timer --no-pager'
+
+**EXPECTED:** unit last ran `Wed 2026-06-17 17:29:31 UTC`; `last_good` mtime `2026-07-05 03:04:51.209`; timer
+last trigger `Sun 2026-07-05 03:04:50 UTC`. The slot was written **18 days after the unit last ran and 1.2 s
+after the timer fired**, and `write_last_good()` is reachable **only** from `reconcile()` (call sites
+`:478,484,491,527`, all inside it) ⇒ `reconcile("keycloak")` executed from the sweep.
+
+**Also:** `warm-keycloak.service`'s `ExecStart` resolves to `/nix/store/…-runner/warm_reconcile.py`, and
+`harness/warmsnap.py` lives in that **same** derivation. So `nixos-rebuild switch` re-runs reconcile **iff
+`runner/**` changed** — an unrelated switch heals nothing, while the merge's own activation switch *is* a
+reconcile trigger (alongside the Sunday sweep).
+
+**Blast radius: keycloak only.** `warm.WARM_DOMAINS == {"keycloak"}` and `keycloak` is the only `SPECS` entry
+with `stateful: True` (`traefik` is `stateful: False` — version-rollback-only, never snapshots). No other
+warm unit reaches either call site.
+
+**HOW to verify the trap is not armed** (on cc-ci; `python3` is absent there, so `ls`/`cat`):
+
+    ssh cc-ci 'ls -A /var/lib/ci-warm/keycloak/; cat /var/lib/ci-warm/keycloak/snapshot/meta.json 2>&1'
+
+**EXPECTED (observed 2026-07-09T08:5xZ):** `last_good` only, and `cat` reports `No such file or directory` —
+no `snapshot/`, hence no `meta.json`; no `/var/lib/ci-warm/canon-keycloak` either; live keycloak `200`. **A
+`meta.json` whose `"domain"` is anything other than `warm-keycloak.ci.commoninternet.net` means the trap is
+armed:** delete that foreign `snapshot/` (or fix B-redfix-5) **before the next Sunday sweep**, not merely
+before the merge. A `meta.json` recording `warm-keycloak.ci.commoninternet.net` is self-consistent and passes
+the guard.
+
+---
+
+### F-redfix-4 remedy (the reason DONE was withdrawn and re-asserted) — 2026-07-09
+
+**WHAT.** F-redfix-4 is fixed at `redfix-m2-harness` @ **`b5f2b10`** (parent `07fc6d4`, the sha the M2 PASS
+was given against). Warm state is now keyed by a **stack namespace**, not a bare recipe:
+`canonical.canonical_ns(r)` is the single source from which BOTH the canonical's domain and its warm-state
+slot derive. A live-warm provider (`r in warm.WARM_DOMAINS`) gets ns `canon-<r>` → domain
+`warm-canon-keycloak.ci.commoninternet.net` (**unchanged**) + slot `/var/lib/ci-warm/canon-keycloak/`.
+Every other recipe keeps ns `<r>` (the invariant is structural — `r not in warm.WARM_DOMAINS` — not a
+property of any particular count). `WARM_DOMAINS == {"keycloak"}` is a singleton, so the re-key is provably
+`keycloak -> canon-keycloak` and nothing else: of the 21 enrolled recipes the other 20 satisfy
+`canonical_ns(r) == r`, so **zero existing canonical slots change**. On live disk that is 17 seeded
+canonicals (those carrying a `canonical.json`), all of which remain in `prune_stale`'s `keep` set.
+
+Addresses all four consequences: (1)+(2) slots disjoint, neither deployment can replace the other's
+known-good; (3) the outage race is removed with the shared slot; (4) `prune_stale`'s reconciler-dir
+invariant is now structural — `<recipe>/` never gains a `canonical.json`.
+
+Also: `warmsnap._assert_slot_not_foreign()` refuses to snapshot/restore a slot recorded against a different
+domain (defence in depth; fails before the destructive swap, not at the next restore). The two comments
+asserting the deployments "can never touch each other" are corrected (`canonical.py`,
+`tests/keycloak/recipe_meta.py`). `WARM_CANONICAL = True` is retained — keycloak stays enrolled, no
+skip-guard, no silent de-enrollment.
+
+**MIGRATION: none required.** On cc-ci `/var/lib/ci-warm/keycloak/` contains only `last_good` (the
+canonical was never seeded), so no existing file changes slot. Verify: `ls -A /var/lib/ci-warm/keycloak/`
+→ `last_good` only.
+
+**WHERE.** cc-ci repo, branch `redfix-m2-harness` @ `b5f2b10`. Files: `runner/harness/warmsnap.py`,
+`runner/harness/canonical.py`, `runner/warm_reconcile.py`, `runner/run_recipe_ci.py`,
+`tests/keycloak/recipe_meta.py`, `tests/unit/test_warmsnap.py`, `tests/unit/test_canonical.py`.
+
+**HOW to verify (1) — unit suite, cold clone, no docker needed.**
+
+    git clone -q --branch redfix-m2-harness <cc-ci remote> /tmp/v && cd /tmp/v
+    /nix/store/x188l04r3gfkh18gy1dpf05fv3kkrgs7-python3-3.12.8-env/bin/python3 -m pytest tests/unit -q
+
+EXPECTED: `325 passed` (baseline at parent `07fc6d4` is `315 passed`; +10 F-redfix-4 regression tests).
+The 10 include `test_live_and_canonical_slots_are_disjoint`, `test_slot_maps_1to1_to_its_stack`,
+`test_registry_path_of_live_warm_provider_is_not_the_reconciler_dir`,
+`test_prune_stale_deenrolled_provider_spares_reconciler_last_good`,
+`test_snapshot_refuses_to_clobber_another_domains_slot`, `test_restore_refuses_a_foreign_slot`.
+
+**HOW to verify (2) — the VETO's clearing condition, on the real node.** Non-destructive: writes only to a
+scratch `CCCI_WARM_ROOT` + one throwaway docker volume. Uses the real idle `warm-canon-keycloak` stack for
+the canonical side and a throwaway `warm-fakelive_…_data` volume as the live-warm stand-in (the live
+`warm-keycloak` stack cannot be undeployed to snapshot it).
+
+    ssh cc-ci
+    docker volume create warm-fakelive_ci_commoninternet_net_data
+    MP=$(docker volume inspect -f '{{.Mountpoint}}' warm-fakelive_ci_commoninternet_net_data); echo x > $MP/live.txt
+    git clone -q --branch redfix-m2-harness <cc-ci remote> /tmp/v && mkdir -p /tmp/vw/keycloak
+    echo '10.7.1+26.6.2' > /tmp/vw/keycloak/last_good
+    cd /tmp/v/runner && CCCI_WARM_ROOT=/tmp/vw /nix/store/jag2131a95gw6ng7grig9pj3dn2q8vrv-python3-3.12.8-env/bin/python3 - <<'PY'
+    import sys; sys.path.insert(0, ".")
+    from harness import warmsnap as ws, canonical as c
+    LIVE, CANON = ws.live_slot("keycloak"), c.canonical_slot("keycloak")
+    CANON_DOM, LIVE_DOM = c.canonical_domain("keycloak"), "warm-fakelive.ci.commoninternet.net"
+    print(ws.snap_dir(LIVE), ws.snap_dir(CANON), c.registry_path("keycloak"), sep="\n")
+    ws.snapshot(CANON, CANON_DOM, version="canon-known-good")
+    ws.snapshot(LIVE,  LIVE_DOM,  version="live-last-good")     # used to clobber the canonical
+    print("restore(canon) ->", ws.restore(CANON, CANON_DOM)["volumes"])
+    print("restore(live)  ->", ws.restore(LIVE,  LIVE_DOM )["volumes"])
+    print("last_good:", open("/tmp/vw/keycloak/last_good").read().strip())
+    try: ws.snapshot(CANON, LIVE_DOM); print("FAIL: foreign snapshot allowed")
+    except ws.SnapshotError as e: print("foreign REFUSED:", str(e)[:60])
+    PY
+    docker volume rm warm-fakelive_ci_commoninternet_net_data; rm -rf /tmp/vw /tmp/v
+
+EXPECTED (observed 2026-07-09):
+
+    /tmp/vw/keycloak/snapshot                     <- live slot
+    /tmp/vw/canon-keycloak/snapshot               <- canonical slot: DISJOINT
+    /tmp/vw/canon-keycloak/canonical.json         <- registry NOT in the reconciler's dir
+    restore(canon) -> ['warm-canon-keycloak_ci_commoninternet_net_mariadb',
+                       'warm-canon-keycloak_ci_commoninternet_net_providers']
+    restore(live)  -> ['warm-fakelive_ci_commoninternet_net_data']
+    last_good: 10.7.1+26.6.2                      <- survived the canonical seed
+    foreign REFUSED: warm slot 'canon-keycloak' holds the known-good of ...
+
+Each `restore()` returns its OWN stack's volumes (the clearing condition). Pre-fix, the same script prints
+one shared slot `/tmp/vw/keycloak/snapshot` for both and `restore(canon)` raises `SnapshotError`.
+
+**Node integrity after my own run of the above (2026-07-09).** Real warm root untouched
+(`/var/lib/ci-warm/keycloak/` = `last_good` only); `warm-canon-keycloak` mariadb digest byte-identical
+across the restore round-trip (`1201440268 48846` before and after, 391 files / 2 files); live
+`warm-keycloak…/realms/master` → **200** throughout; throwaway volume removed; scratch removed.
+
+**Scope.** keycloak only. M1 classifications and the discourse / mattermost-lts / gitea / bluesky-pds /
+mumble fixes are untouched by this commit and remain Adversary-verified (M1 PASS @2026-06-18T01:18Z; the
+five non-keycloak M2 fixes content-verified through re-confirmation #7).
+
+**Known residual, NOT in F-redfix-4's clearing condition** (filed as B-redfix-5 in BACKLOG-redfix.md, not
+blocking): `warm_reconcile.py`'s rollback `warmsnap.restore()` still sits outside the `try/except` that
+guards the upgrade, so any restore failure (e.g. a corrupt/absent snapshot) leaves live keycloak undeployed
+after `abra.undeploy()`. The F-redfix-4 race that made this reachable is gone; the pre-existing robustness
+gap is not. Recorded, not silently dropped.
+
+The prior DONE text is retained verbatim below as the historical record of what was claimed on 2026-06-18.
+
+---
+
+## (HISTORICAL — withdrawn 2026-07-09 under VETO F-redfix-4; superseded by the DONE above) DONE — 2026-06-18T07:09Z
 
 Phase `redfix` COMPLETE. All six canon-sweep failures investigated in isolation, root-caused,
 classified, **FIXED — each via a recipe PR or a harness improvement — and verified green**; no recipe
@@ -26,6 +475,29 @@ M1 read-only crash gone; bluesky-pds recipe PR #4 @4987ba9 (caddy `${STACK_NAME}
 200 (was 000). gitea/bluesky end-to-end canonical advance is operator-merge-gated (fix proven by
 chaos-deploy; published tags don't carry it pre-merge) — consistent with "nothing merged", not a shrug.
 
+### Evidence addendum 2026-07-08 (re: F-redfix-3) — discourse sha pins have rotted; verify by CONTENT
+
+The discourse shas cited below (`9ff5e19`, `53ba0910`) **no longer resolve on the mirror**. A *later* phase
+force-pushed and extended the shared branch `discourse-official-image` (now `ede6399` = `refs/pull/5/head`;
+redfix's PR is no longer `#4`). This is branch drift by later work, **not** a retraction of the redfix fix,
+and it does not disturb the M2 PASS — which was given against the shas that existed on 2026-06-18. The other
+three recipe pins and the harness pin are exact and still resolve: mattermost-lts `ci/pg-restore`@`4ca7f418`,
+gitea `ci/app-ini-writable`@`a0f2db88`, bluesky-pds `ci/warm-routing-alias`@`4987ba91`, cc-ci
+`redfix-m2-harness`@`07fc6d4a`.
+
+Historical sha lines below are left **unedited on purpose** — they record what was verified when. Use this
+durable **content assertion** instead of the shas to re-verify discourse at any future head:
+
+    git clone https://git.autonomic.zone/recipe-maintainers/discourse && cd discourse
+    git show origin/discourse-official-image:compose.yml | grep -m1 'image:.*discourse'
+    git show origin/discourse-official-image:compose.smtpauth.yml | grep -c sidekiq
+
+EXPECTED: `image: discourse/discourse:3.5.3` (the official-image migration = M2's claim) and `0` (the
+F-redfix-1 orphaned-sidekiq removal). Both re-confirmed by the Builder at `ede6399` on 2026-07-08, and by the
+Adversary at `ede6399` and `refs/pull/4/head`@`0c4539b7`. (Caveat when reproducing absence of a sha: a
+`--filter=blob:none` clone and `git fetch origin <sha>` both yield false "absent" signals — test reachability
+from all `refs/heads/*` + `refs/pull/*/head` instead.)
+
 ---
 
 ## Phase: M1 — investigate + isolate + classify (IN PROGRESS)
@@ -235,6 +707,33 @@ in the two sections above ("M1 results table" and "HOW the Adversary cold-verifi
 Evidence logs on cc-ci: `/tmp/redfix-{discourse,mattermost-lts,mumble,mumble2,bluesky-pds,gitea2}.log`.
 Reasoning/dead-ends in JOURNAL-redfix.md. Node left clean (see "Node state left clean" above).
 
+## Untracked file `main.go` — NOT produced by this phase (operator: unexplained)
+
+**WHAT.** An untracked `main.go` (281 bytes, a Go "Hello, World!" `net/http` server on `:8080`) sits in the
+repo root. It is not in any commit on any ref, is not gitignored, and is unrelated to every redfix DoD item.
+No `go.mod` accompanies it. I did not create it and have not committed, modified, or deleted it.
+
+**WHERE.** `/srv/cc-ci-orch/cc-ci/main.go` — **one file in one clone.**
+
+**RETRACTED (2026-07-09, Adversary A-redfix-3):** an earlier version of this entry claimed the file was
+"present in **both** clones … i.e. written by something outside git." **That evidence is void.** `/srv/cc-ci`
+is a **symlink to `/srv/cc-ci-orch`**, so `/srv/cc-ci/cc-ci/main.go` and `/srv/cc-ci-orch/cc-ci/main.go` are
+the **same inode** — one file seen twice, not two clones. The "a git operation cannot leave the same untracked
+file in two clones" inference therefore carries **no information** about the file's origin, which remains
+**unexplained**. Independently re-derived: `ls -la /srv/` shows `cc-ci -> /srv/cc-ci-orch`, and
+`stat -c '%i %h %n'` on both paths prints the same inode `3254604` with `links=1`.
+
+**HOW to verify.** `git log --all --oneline -- main.go` → empty (in no commit).
+`git status --short` → `?? main.go`.
+`ls -la /srv/ | grep cc-ci` → `cc-ci -> /srv/cc-ci-orch` (the symlink that voids the "both clones" claim).
+`stat -c '%i %h %n' /srv/cc-ci/cc-ci/main.go /srv/cc-ci-orch/cc-ci/main.go` → same inode, `links=1`.
+
+**Risk: inert** (Adversary-confirmed). 281-byte hello-world `net/http` listener; no `go.mod`; `go` is **not
+installed**; **nothing listening on `:8080`**; in no commit on any ref; unreferenced by any tracked file.
+
+**EXPECTED.** Left in place, untracked. Per guardrails I do not delete or commit files I did not create;
+flagging for the operator rather than acting. It affects no gate, no DoD item, and no verdict.
+
 ## Blocked
 
 (none)

scripts/recipe-mirror-sync.sh

@@ -36,7 +36,10 @@ RECIPE_DIR="${HOME}/.abra/recipes/${RECIPE}"
 }
 TOKEN="$(tr -d '[:space:]' <"${TOKEN_FILE}")"
 API="https://${GITEA_HOST}/api/v1"
-MIRROR_PUSH="https://oauth2:${TOKEN}@${GITEA_HOST}/${NAMESPACE}/${RECIPE}.git"
+# Push over SSH (git@:2222, key in ~/.ssh) so no credential is ever written into the recipe
+# clone's .git/config — that config gets copytree'd into the world-readable per-run tree
+# (CCCI_SKIP_FETCH staging), which previously leaked the token. API still uses the token below.
+MIRROR_PUSH="git@${GITEA_HOST}:${NAMESPACE}/${RECIPE}.git"
 auth=(-H "Authorization: token ${TOKEN}")
 
 # Ensure the recipe is cloned (abra recipe fetch clones from the catalogue → upstream).