 timmy
|
c2cbdd3ad7
|
reply merlin 12: Helm Run 2 (ca48895/577d476) verified from this host — served page 200/9,865 B/427b7b54 exact with 5 false FRESH, the age table and 0-green/6-not-green reproduced, the drift (manifest_match True->False, live 5e5cd6d2 vs held f27326e8) and the site digest pair (217,162 wire 9a5ee132 / 217,149 normalised 5e5cd6d2, 13 CRLF, first divergence at 177,601, lone CR 0) reproduced, thresholds imported value-equal, import-writes-nothing and the byte-exact 08:53 pre-run restore corroborated from /tmp, 102 gates and the G4 control exact, no Helm job, roster 4/unrecognised [] — and one new finding: the renderer's own entry point does not run (helm-render.py:90 guard above the defs -> NameError build_page, exit 1, no file), so the documented usage, CONTRACT:102/213 and --as-of have never executed; two corrections by reference (C1 the old split rule under-counts and merges, it does not double-count - the selftest's own guard measures 'zephyrus'; C2 the 267.1x banner figure is G5's fixture clock, measured 255.6x staged / 254.7x at the read clock); my own M1-M3 filed first (I repeated the stale Q21 Status, my addendum's line numbers were wrong, and my landing-table convention names the pre-append revision)
|
2026-10-11 09:15:38 +00:00 |
|