Skip to content

P9 isolated Windows trial: exact ARSAS+ARIEC SHA, nonrelease four-RCB validation - #767

Draft
masarray wants to merge 1 commit into
fix/p8-siebcu-first-connect-bits-display-20261105from
integration/p9-multircb-exact-windows-trial-20261105
Draft

masarray wants to merge 1 commit into
fix/p8-siebcu-first-connect-bits-display-20261105from
integration/p9-multircb-exact-windows-trial-20261105

Conversation

@masarray

@masarray masarray commented Nov 5, 2026

Copy link
Copy Markdown
Owner

P9 exact-SHA Windows integration candidate (nonrelease)

Immutable provenance

Scope

Use the existing label-gated, explicitly nonpromotable ci-p1-shadow reporting integration pipeline; update only the trial manifest and add backward-compatible P9-specific artifact filename. Fail closed if the manifest's engine SHA disagrees with exact checked-out engine; preserve legacy P8 naming for a P8 manifest. The test workflow includes Windows solution build, focused reporting regression, native bridge CTest, single EXE build, executable smoke, and exact SHA256 manifest.

No premature acceptance

CI success will prove build/smoke and synthetic correctness, not SG7 physical 77/77 or 4/4, first-click reliability on SIEBCU, or lossless SOE. Keep this PR Draft, test artifact nonrelease, and SG7 issue #707 Open. No hidden cyclic MMS polling, blanket RCB ownership mutation, or automatic control is authorized.

Please keep the separate P8 and P9 source branches intact to avoid cross-thread collisions.

Reuse vetted nonpromotion FAST DEV workflow, preserve P8 fallback, stage P9-specific artifact name and engine/manifest verification without changing mainline engine lock.

masarray commented Nov 5, 2026

Copy link
Copy Markdown
Owner Author

P9 exact-SHA software milestone — CI verified, NOT physical acceptance (2026-11-05).

What this DOES NOT establish: SG7 physical recovery to 77/77 and 4/4; original SIEBCU first-attempt TCP/ACSE failure cause; lossless SOE continuity or physical endurance. CI is a synthetic/integration build gate, not IEC 61850 interoperability FAT. Keep PRs Draft and issue #707 OPEN pending actual field evidence. No cyclic MMS polling or spontaneous control/RCB ownership changes were introduced.

masarray commented Nov 5, 2026

Copy link
Copy Markdown
Owner Author

Final dual-path validation (exact ARSAS source, 2026-11-05)

  • P9 nonrelease isolated engine 58d22be43d93d1d439df04a99ba935a7ec45cd4c: CI-P1 FAST reporting #41981252170 SUCCESS: 951/951 focused reporting PASS, native bridge/CTest PASS, EXE smoke PASS, SHA-256 23b5742166605b16c1e015d0f584e1eaf936dbb2a5dac62d2d988b22c4ea2f76.
  • Unchanged production engine baseline f75e323e9f8443aa20e0a3818a1d5d77255b2c0f: canonical Windows Build #41981246131 SUCCESS: 3,027/3,027 ARSAS tests PASS plus 6/6 packaged smoke checks. This is a separate pinned-engine build, not proof that its binary embeds P9.
  • CI-P0 Workflow Integrity #41981246074 SUCCESS; production pin and release unchanged.

Code and nonrelease Windows trial packaging gates are fully green at source 063cf6e26c401e0586dc440adb5e646d8e74b0cd. PR remains Draft as physical SG7 77/77 + 4/4 reporting recovery is still unverified.

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant