Naprawa kontraktu CLAUDE.md §Service Structure (opcja b). Migracja do KB
zabrala README z katalogow serwisow i hostow, przez co 0/26 katalogow
services/ spelnialo wymagany layout. Wskazniki przywracaja nawigacje,
nie duplikujac tresci.
31 wskaznikow, jednolity format, dokladnie 5 linii:
# <nazwa>
<jedno zdanie opisu>
Dokumentacja: [kb/...](../../kb/...)
Opis nie jest pisany od zera — wyciagany z kb-doca: pierwsze pelne zdanie
pierwszego akapitu (sklejane z zawinietych linii, ciete tylko tam, gdzie
backticki i nawiasy sa zbilansowane), a dla node'ow czlon tytulu H1 po
myslniku. Dla ha-mcp opis z H1, bo pierwszy akapit zaczyna sie od markera
statusu. Wiodace markery "**Status: ...**" sa zdejmowane.
26 x services/<svc>/README.md, 5 x hosts/<node>/README.md.
WYJATEK services/home-assistant/config/ken-legacy/README.md: pelne
ostrzezenie "historical archive, do not deploy" przywrocone doslownie
z historii (odzyskane z drzewa sprzed migracji) + link do kb-doca.
Ostrzezenie musi stac tam, gdzie chroni — w katalogu archiwum, nie tylko
w KB. Odwolanie do services/home-assistant/DESIGN.md przepiete na
kb/decisions/ha-configs-as-code.md + kb/incidents/2026-07-22-ha-dwie-instancje.md.
check_okf.py: POINTER_GLOBS + is_pointer() wykluczaja wskazniki ze scope'u
lintu. Wskazniki celowo NIE maja frontmattera OKF — to nawigacja, nie
dokumenty KB. Wykluczenie zapisane wprost, zeby poszerzenie SCOPE nie
zaczelo ich nagle walidowac.
Bez wskaznikow: hosts/chelsty-ha/ i hosts/lustro/ — nie maja dokumentow
w kb/nodes/ (luka odnotowana juz w reconie etapu 1). Utworzenie ich
wymagaloby napisania nowej dokumentacji, czyli wyjscia poza konwersje.
Lint: 190/190 ZGODNE. Weryfikacja 822 plikow: 0 martwych linkow.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
126 plikow (md, yaml, sh, py) odwolywalo sie do sciezek sprzed migracji.
15 markdown-linkow [..](..) -> policzona sciezka WZGLEDNA wobec pliku
odsylajacego (wczesniej czesc z nich byla repo-root-relative i nie
rozwiazywala sie z katalogu, w ktorym lezala)
200 odwolan tekstowych (backticki, proza, yaml, importy w kodzie)
-> nowa sciezka repo-root-relative, zgodnie z konwencja repo
5 linkow rodzenstwa (gole nazwy plikow, np. "](DEPLOY.md)") — dzialaly
tylko w starym katalogu; przeliczone recznie
Objete m.in.: CLAUDE.md (scripts/onboard/README.md -> kb/runbooks/
node-onboarding-tool.md, docs/backlog.md -> kb/phases/backlog.md),
README.md, .claude/skills/, 20 session logow, kod jobow.
Ostatnie 5 odwolan pochodzi z tresci wciagnietej rebasem z origin/master
(session log 2026-07-31, override node-agenta na SOLARII, dwie pozycje
backlogu) — wskazywaly na docs/incidents/, docs/kb/modules/ i
services/narty27/README.md sprzed migracji.
Dodany wzajemny link miedzy kb/services/control-plane.md (stub kodu)
a kb/subsystems/control-plane.md (opis, deprecated) — dwa dokumenty o tym
samym systemie, latwe do pomylenia.
Weryfikacja na 790 plikach: 0 odwolan do starych sciezek,
0 martwych linkow markdown. Lint OKF: 190/190 plikow ZGODNE.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
R1 (critical): _prune_stopped_containers no longer calls containers.prune().
The unfiltered API call removes EVERY non-running container, ignoring
RestartPolicy and compose labels — that is what deleted ollama@solaria 19 s
after an operator `docker stop`. No filter argument can fix it (`until` filters
on creation time, not stop time). Replaced with explicit enumeration over
containers.list(all=True, filters={"status": "exited"}), skipping anything with
restart policy unless-stopped/always/on-failure or a com.docker.compose.project
label. A stopped managed service is recorded operator intent and now belongs to
the module's NEVER TOUCHED list; only one-off leftovers are removed. Dangling
image and build cache prune are unchanged.
R2 (high): _sd_card_rate_ok → _cleanup_rate_ok, applied to every cleanup-
eligible node type. ai_node and standard pruned every 60 s (1440/day, "0 MB
reclaimed" in practically every cycle) with no rate limit at all; they now share
sd_card's CLEANUP_INTERVAL_SECS (24 h) and mark the cleanup timestamp.
R3 (medium): removals are named in the log — WARNING with the container names
when the list is non-empty, INFO otherwise. The old line reported megabytes
only, which is why ollama's deletion looked identical to every no-op cycle.
Tests: services/node-agent/tests/test_safe_cleanup.py (22 cases) — the ollama
regression (exited + unless-stopped survives, prune() never called), all three
restart policies, compose-label protection, disposable-leftover removal, sweep
continues past a failed remove, WARNING-level naming, and the rate limit for
ai_node/standard/sd_card plus lte_node still doing nothing.
Refs docs/incidents/2026-07-30-ollama-solaria-vanish.md §7 R1-R3.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Executor nie ma klienta ssh ani klucza do floty (uid 1000 homelab, brak
~/.ssh, brak resolucji nazw wezlow) — container_restart przez subprocess ssh
failowal w 6ms na kazdej probie. Zamiast dodawac SSH do executora, kierunek
jest odwrocony: executor zapisuje zlecenie do
/opt/homelab/actions/dispatch/<node>/<action_id>.json, a node-agent na
docelowym wezle (ktory ma dzialajacy docker.sock i juz ma klucz SSH do VPS
uzywany do shippingu eventow) sam je odbiera i wykonuje lokalnie.
- executor: _dispatch_container_restart pisze zlecenie zamiast ssh;
_reconcile_running_actions konsumuje zwrotne action_result eventy i
timeoutuje akcje bez odpowiedzi (ACTION_TIMEOUT_SECS, domyslnie 300s).
redeploy/disk_cleanup/alert_only bez zmian.
- node-agent: nowy krok w petli — rsync-pull wlasnej podkatalogu dispatch z
VPS (ten sam klucz co _ship_events_to_vps, w przeciwnym kierunku; no-op na
VPS, gdzie katalog jest lokalny), walidacja (node_name, whitelist tylko
container_restart, odmowa restartu wlasnego kontenera), wykonanie przez
docker SDK, raport jako event action_result (istniejacy kanal shippingu).
Idempotencja przez znacznik w /opt/homelab/state/processed-actions/.
- 26 nowych testow (10 executor, 16 node-agent), pelny suite obu serwisow
183/183 zielony.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- node-agent: service_healthy emitowany tylko przy przejsciu w stan zdrowy
(per-service in-memory state), nie co cykl dla kazdego zdrowego serwisu.
To samo dla control-plane HTTP probe. healthcheck_failed/containers_not_running/
incydenty pozostaja emitowane bez zmian (realne sygnaly).
- node-agent: naprawiono _cleanup_control_plane_fs — istniejaca retencja
eventow byla martwa od migracji observer_checkpoint.json na epoch-int
(str(path) <= int rzucal TypeError, lapane cicho przez broad except).
Teraz porownuje epoch-do-epoch i czysci wylacznie service_healthy/node_health
starsze niz checkpoint + 3-dniowy bufor; healthcheck_failed/incydenty/ha_*
zachowane bezterminowo.
- scripts/maintenance/cleanup_event_backlog.py: jednorazowy skrypt do
bezpiecznego czyszczenia backlogu na VPS (dry-run domyslnie, --apply
do usuniecia). Ten sam warunek: typ szumu + starsze niz checkpoint + 1h bufor.
check_containers() classified only exited/dead, running+unhealthy and running.
Docker's "restarting", "paused", "removing" and any future state matched no
branch, so a crash-looping container (restart policy + continuous crash) emitted
ZERO events and world-state kept showing it "healthy" (observed live:
lustro/pi-watchtower-1). Every Docker state is now handled:
- restarting + RestartCount >= CRASH_LOOP_RESTART_THRESHOLD (default 3):
reuse containers_not_running (high, crash_loop=true) — parity with exited/dead,
rides the existing supervisor-wired remediation path.
- restarting below threshold: new observational container_restarting (low),
visible but non-actionable so benign post-deploy restarts don't alarm.
- paused / unknown-or-future state: new diagnostic container_state_unexpected
(medium) — no more silent fall-through; new Docker states become visible.
- removing: conscious documented skip (ephemeral teardown).
- created: unchanged skip (compose tracking artifact).
RestartCount (top-level inspect field) distinguishes a crash-loop from a one-off
restart. Adds services/node-agent/tests/test_check_containers.py pinning the full
state table.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
node-agent had no USER instruction and no user: in compose, running
as root and writing root-owned files to /opt/homelab bind-mount.
- Dockerfile: add useradd -m -u 1000 homelab + USER homelab
- docker-compose.yml: add user: "1000:1000" and group_add: ["999"]
(GID 999 = docker group on VPS) to retain docker.sock access
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
node-agent: use com.docker.compose.service label as canonical name
- Add _canonical_container_name() method: prefers compose label,
falls back to hash-prefix-stripped c.name
- Replace bare c.name usage in check_containers()
- Skip 'created'-state containers (Docker stale-state artifacts)
observer: prune hash-prefixed ghost keys in _prune_stale_world()
- Each reconcile cycle removes service keys matching <node>/<12hex>_<name>
- Acts as safety net for entries already in services.json + future slippage
control-plane/docker-compose.yml already has explicit container_name on
all four services — no change needed there.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Multiple service_healthy (or containers_not_running) events emitted in the
same second for different containers shared the same filename pattern
evt-{node}-{ts}-{type}.json — the second write silently overwrote the first,
so the observer only ever saw the last container checked per event type per cycle.
Fix: include a sanitized service name slug in the ID so every event gets a
unique file, e.g. evt-vps-1234-service_healthy-node-agent.json.
Also adds import re (required for re.sub in the slug generation).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- node_agent: emit service_healthy for all running managed containers so
observer populates services.json (previously empty → supervisor flooded
action queue with missing_service redeploys for healthy services)
- node_agent: VPS-only _check_control_plane_health() probes the HTTP
endpoint to emit service_healthy/unhealthy for the 'control-plane' logical
service (multi-container stack, container names don't match service name)
- node_agent: fix _cleanup_control_plane_fs() to read new node_checkpoints
format from observer checkpoint (was reading old last_processed_file key,
always found nothing, never cleaned up old events)
- observer: handle service_healthy event type → sets service status healthy
without resolving incidents (unlike service_recovered which also resolves)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
When ~/.ssh is mounted from the host oskar user into a container that
runs as root, OpenSSH rejects ~/.ssh/config with 'Bad owner or
permissions' because the file UID doesn't match the running process.
Add -F /dev/null to the rsync SSH command to skip the config file
entirely. Also add UserKnownHostsFile=/dev/null so no known_hosts
write is attempted into a potentially read-only mounted .ssh dir.
The key itself (/root/.ssh/id_rsa) is still read as an implicit
default identity and is not affected by -F.
Reproduces on chelsty-infra (has ~/.ssh/config); safe for all nodes.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Oskar Kapala