The memory-yield LOW (start-inhibit) watermark forced pending=0, which the
idle-debounce path then read as 'no work' and used to idle-stop a RUNNING,
actively-analyzing vision once the box floated past LOW (idle stack + LLM
≈ 80% on this Pi) — so vision was killed mid-analysis at ~80%, not the 92%
HIGH watermark. Split the two levers: LOW now sets a mem_block_start flag
that only holds off STARTING a stopped vision; only the HIGH active stop (or
a genuinely drained queue) stops a running one. Observed live: vision now
holds through the 79–81% band and completes pages.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Backend:
- Log on readiness state transitions (ready<->not-ready) so a misconfigured
ANALYSIS_VISION_HEALTH_URL, which otherwise parks the worker silently
forever, is diagnosable.
- Add unit tests for HttpVisionReadiness: 2xx->ready, non-2xx->not-ready,
connection error->not-ready (the production status mapping had no test).
vision-manager:
- Guard the backlog count against non-numeric output before `-gt`, so a stray
value can't exit-2 and kill the loop under `set -e`.
- Throttle the crawl-mutex "deferring start" log to once per episode.
- Only reset the idle/uptime timers when `docker stop` actually succeeds, so a
failed stop retries next tick instead of waiting a full debounce window.
- Decouple the per-probe curl timeout (HEALTH_TIMEOUT) from the warm-up poll
cadence.
Docs:
- Correct the docker-socket-proxy comment: CONTAINERS+POST permits the full
container lifecycle (not just start/stop); state the real trust boundary.
- Document that the externally-defined mangalord-vision container must share
the compose network for name resolution.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add a tiny privileged sidecar (vision-manager/) that polls the analysis
backlog in Postgres and starts/stops the mangalord-vision container by name:
start when analyze_page jobs are pending, stop after STOP_DEBOUNCE idle.
It is the single owner of the vision lifecycle and the only component with
Docker access — scoped through tecnativa/docker-socket-proxy (CONTAINERS+POST
only) on an internal-only network, so the internet-facing backend never
touches the socket.
Both helpers sit behind `profiles: [ai]` (vanilla `compose up` is
unaffected). The manager debounces the stop, gates the first request on
GET /health==200 after a cold start, honours a crawl RAM-mutex on the 8 GiB
box, and owns its own idle timer (leak-safe if the backend dies). Backlog
query uses the real schema (state + payload->>'kind'); a read-only DB role
(readonly-role.sql) keeps it off the backend creds.
Bump 0.80.0 -> 0.81.0.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
