website dashboard screenshots: stopped at the baseline — scratch 9202 runs controller 0.301.0, the fleet 0.303.0; one question for the operator
gates / gates (push) Failing after 12m10s

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012qRErfCoiTkvDK9N5XHbzb
This commit is contained in:
2026-10-08 11:05:00 +02:00
parent 2fd8ff79c5
commit 4b5f74e46c
+36
View File
@@ -0,0 +1,36 @@
# REPORT — dashboard screenshots for the website (2026-10-08): STOPPED at the baseline
> `REPORT-website-dashboard.md`, not `REPORT.md`: another session works in this repo today.
| Part | What | State |
|---|---|---|
| Baseline | a scratch guest that runs the fleet's controller | **Not met — stopped here, as the brief says** |
| A | the scene + screenshots | not started |
| B | the pages + the hu/en picture gate | not started |
| C | rules, CHANGELOG, STATUS | not started (nothing changed on the site) |
## What was measured (read only)
- **Fleet controller: 0.303.0** — `STATUS.md` header („demo-hp, demo-felhom and Tester 1 run … controller 0.303.0"),
newest entry in `felhom-controller/CHANGELOG.md`: `## v0.303.0` (2026-10-07).
- **Scratch guest 9202 runs 0.301.0** — `pct exec 9202 -- cat /etc/felhom-controller-image` →
`gitea.dooplex.hu/admin/felhom-controller:0.301.0`; `docker ps` shows `felhom-controller … 0.301.0 Up 55 minutes`.
- **No other scratch guest exists** — `pct list` on the demo-hp host: 9201 (demo-hp, the demo household — no contact),
9202 (scratch), 9401 (`upgrade-harness`, the catalog bench, stopped; it runs no controller). demo-felhom and the
Testers are fenced.
- **The site has no dashboard picture today** — `website/assets/` holds no `dashboard*` file.
- 9202's controller restarted 55 minutes before this check, and it carries paperless-ngx: **another session may be
using 9202 today** (session 1 reported „R-762 closed (9202 proven)" in d9147b02).
Nothing was installed, changed or removed on 9202. The only host command was `pct exec 9202 -- …` (read) and `pct list`.
## Decision for you
**May I move scratch guest 9202 to controller 0.303.0 — the fleet's version, not past it — and then take the
screenshots?** (On 9202 the image is set by hand in `/etc/felhom-controller-image` + a restart of its bootstrap
service; `operations/nodes.md` — allowed only there.)
- **Pick: yes**, at a time when the other session is not using 9202. Cost: about two minutes of 9202 restart. The
pictures then show what households really run.
- Or: take the pictures on 0.301.0 now. Cost: the pictures show a dashboard two releases old (0.302.0 changed the
format list and 22 messages).
- **If you do nothing:** no pictures; the site stays as it is.