cc-ci

recipe-maintainers/cc-ci

Fork 0

Commit Graph

Author	SHA1	Message	Date
autonomic-bot	c6e94af766	fix(2): F2-5 — dep teardown verify=True, errors propagate to run-fail (Adversary cold) Per REVIEW-2 ## Q2 FAIL: runner/harness/deps.py::teardown_deps suppressed ALL exceptions via contextlib.suppress(Exception), silently swallowing teardown failures. The 'DEPS teardown' print fired even when undeploy actually raised — leaving leftover swarm services/volumes/secrets that broke the NEXT run targeting the same deterministic dep domain (this is what caused the Q3.1 dep flake I saw immediately after the Q2.4 acceptance run). Fix: - runner/harness/deps.py: teardown_deps now uses lifecycle.teardown_app(..., verify=True) so residuals raise TeardownError. Errors are LOGGED LOUDLY per-dep but we continue to other deps so one failure doesn't strand the rest. After all attempts: raise a combined TeardownError if any dep failed. - runner/run_recipe_ci.py: orchestrator catches the dep TeardownError in finally, prints it, captures into dep_teardown_error; the run summary surfaces it and the exit code is non-zero. The run STILL prints the diagnosable summary so a leak doesn't hide other failures. Per §9 teardown sacred / DG7: a green run that leaks state is not 'green'. F2-5 now correctly fails the run instead of silently passing. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-05-28 09:00:37 +01:00
autonomic-bot	4d6b040ba7	feat(2): Q2.3 — dep resolver + SSO-setup harness primitives - runner/harness/deps.py: dep resolver primitive (Phase 2 §4.2 / Q2.3). - declared_deps(recipe) reads DEPS list from tests/<recipe>/recipe_meta.py - dep_domain(parent, pr, ref, dep) — per-run domain per (parent, dep) pair so two recipes' deps of the same kind don't collide on a host - deploy_deps / teardown_deps — sequential deploy + reverse-order teardown - read/write of run-scoped $CCCI_DEPS_FILE - runner/harness/sso.py: SSO-setup / OIDC-flow primitive (Phase 2 §4.2 / Q2.3). - setup_keycloak_realm: idempotent realm + confidential OIDC client + test user with generated 25-char alphanumeric password (class-B per §4.4-B); returns SsoCreds dict with discovery_url, token_url, all identifiers. - oidc_password_grant: exercises the password-grant OIDC flow; returns access_token (a JWT) or raises. - assert_discovery_endpoint: GET /.well-known/openid-configuration; asserts issuer matches the per-run provider domain+realm. - runner/run_recipe_ci.py: wired in dep deploy BEFORE recipe-under-test, dep teardown LAST in finally (reverse order). DG4.1 deploy-count guard now expects 1 + len(deps_state) — accommodates declared deps without breaking the no-extra-deploys invariant. - tests/conftest.py: deps_apps fixture reads $CCCI_DEPS_FILE -> dict mapping dep_recipe -> dep_domain. - tests/unit/test_deps.py: 7 unit tests covering declared_deps parsing, per-(parent,dep) domain distinctness, run-state JSON write/load, env-var no-op semantics. 28/28 unit tests PASS on cc-ci. Smoke test confirmed deploy_count == expected (1) when no deps declared (custom-html install run, log /root/ccci-q2-deps-smoke.log). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-05-28 07:41:56 +01:00

Author

SHA1

Message

Date

autonomic-bot

c6e94af766

fix(2): F2-5 — dep teardown verify=True, errors propagate to run-fail (Adversary cold)

Per REVIEW-2 ## Q2 FAIL: runner/harness/deps.py::teardown_deps suppressed ALL exceptions via
contextlib.suppress(Exception), silently swallowing teardown failures. The 'DEPS teardown' print
fired even when undeploy actually raised — leaving leftover swarm services/volumes/secrets that
broke the NEXT run targeting the same deterministic dep domain (this is what caused the Q3.1 dep
flake I saw immediately after the Q2.4 acceptance run).

Fix:
- runner/harness/deps.py: teardown_deps now uses lifecycle.teardown_app(..., verify=True) so
  residuals raise TeardownError. Errors are LOGGED LOUDLY per-dep but we continue to other deps
  so one failure doesn't strand the rest. After all attempts: raise a combined TeardownError if
  any dep failed.
- runner/run_recipe_ci.py: orchestrator catches the dep TeardownError in finally, prints it,
  captures into dep_teardown_error; the run summary surfaces it and the exit code is non-zero.
  The run STILL prints the diagnosable summary so a leak doesn't hide other failures.

Per §9 teardown sacred / DG7: a green run that leaks state is not 'green'. F2-5 now correctly
fails the run instead of silently passing.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

2026-05-28 09:00:37 +01:00

autonomic-bot

4d6b040ba7

feat(2): Q2.3 — dep resolver + SSO-setup harness primitives

- runner/harness/deps.py: dep resolver primitive (Phase 2 §4.2 / Q2.3).
  - declared_deps(recipe) reads DEPS list from tests/<recipe>/recipe_meta.py
  - dep_domain(parent, pr, ref, dep) — per-run domain per (parent, dep) pair
    so two recipes' deps of the same kind don't collide on a host
  - deploy_deps / teardown_deps — sequential deploy + reverse-order teardown
  - read/write of run-scoped $CCCI_DEPS_FILE
- runner/harness/sso.py: SSO-setup / OIDC-flow primitive (Phase 2 §4.2 / Q2.3).
  - setup_keycloak_realm: idempotent realm + confidential OIDC client +
    test user with generated 25-char alphanumeric password (class-B per §4.4-B);
    returns SsoCreds dict with discovery_url, token_url, all identifiers.
  - oidc_password_grant: exercises the password-grant OIDC flow; returns
    access_token (a JWT) or raises.
  - assert_discovery_endpoint: GET /.well-known/openid-configuration; asserts
    issuer matches the per-run provider domain+realm.
- runner/run_recipe_ci.py: wired in dep deploy BEFORE recipe-under-test, dep
  teardown LAST in finally (reverse order). DG4.1 deploy-count guard now
  expects 1 + len(deps_state) — accommodates declared deps without breaking
  the no-extra-deploys invariant.
- tests/conftest.py: deps_apps fixture reads $CCCI_DEPS_FILE -> dict mapping
  dep_recipe -> dep_domain.
- tests/unit/test_deps.py: 7 unit tests covering declared_deps parsing,
  per-(parent,dep) domain distinctness, run-state JSON write/load, env-var
  no-op semantics. 28/28 unit tests PASS on cc-ci.

Smoke test confirmed deploy_count == expected (1) when no deps declared
(custom-html install run, log /root/ccci-q2-deps-smoke.log).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

2026-05-28 07:41:56 +01:00

2 Commits