session-report · RPT-FA55C4585705

R7 validation — activation sequence and failure reports

Host activation and status routing · Research method, source claims and publication

Source identity

Record
RPT-FA55C4585705
Original file
Concresca-FULL-ROOT-r7-test-results.md
Original SHA-256
086cb988b976e7ea93b66719533c9b85379621d708b015e8c4aa523418fa903c
Source family
repair-conversation
Public copy SHA-256
086cb988b976e7ea93b66719533c9b85379621d708b015e8c4aa523418fa903c
Original custody
Exact bytes retained in the private project package under docs/_evidence/; not a public download.

Read the public source-text copy

Disposition and proof of use

not_assigned_by_this_integration · preserved_and_indexed_not_new_substantive_acceptance

Memory owners: .uai/architecture.uai, .uai/file-handoff.uai, .uai/operations.uai, .uai/report-synthesis.uai, .uai/research-review.uai, .uai/test-plan.uai. These are private repository records, not public web routes.

Jump to a report section

Report body

R7 validation — activation sequence and failure reports

Tests are local; they do not establish the live deployment's readiness.

Before the change

The unmodified r6 repair suite ran: 199 Python tests passed. Two additional regressions then demonstrated behavior absent from that suite:

  1. Under native Apache/mod_php HTTPS, an immediate activation start after a prerequisite check returned HTTP 429 ACTIVATION_RATE_LIMIT because the actions shared one timestamp. The fixture lacked a private production environment; the backend correctly reported that separate blocker.
  2. In the actual packaged JavaScript with a synthetic DOM, a rejected start left the downloadable result equal to preflight_passed. The new regression expected the failed request and failed against r6.

Reproduction records are included as R6-ACTIVATION-REPRODUCTION.json and R6-CLIENT-REPRODUCTION.txt. They do not identify the cause of the live 404.

R7 suites

Executed on the release source, then repeated against the extracted final ZIP:

Suite Passed
Packaged repair/root/host/activation suite 203 Python tests
Owned-root integration 20 Python tests
Receipt publisher 39 Python tests
Private environment loader 31 Python tests
Python total 293
PHP diagnostic, CSRF and activation functions 75 assertions
Status, diagnostic and activation JavaScript contracts 106 assertions

The Python count includes 34 native Apache/PHP HTTPS tests and seven native Apache boundary tests. Four HTTPS tests, nine PHP assertions and ten JavaScript failure-report cases were added in r7. The external validation file records final extracted-archive checks. No skipped tests are counted as passed.

New coverage checks independent check/start admission, repeat-request rejection and Retry-After, rejected cross-origin requests, redaction, replaced stale reports, network uncertainty, non-JSON errors, and saved-job refresh without starting an extra installation. All prior authentication, CSRF, receipt, workflow, access-boundary and rollback tests remain enabled.

Limitations

The native HTTPS tests run Apache/mod_php on localhost. They deliberately do not possess production environment values. An accepted start launches the real packaged backend, which must fail preflight before routing/cron/network changes. Successful installer transactions are tested with simulated host operations, not actual Passenger or LiteSpeed. MySQL production access and all 22 live workflow outcomes remain unobserved. JavaScript tests are synthetic DOM tests; no new real-browser navigation was performed. Archive hashes identify file bytes, not loaded-code attestation or source provenance. No production changes or ready/verified receipt are included.