Files

T

Frank c99eb8e11b docs(fleet): live-probe corrections across dankclaw, caddy, gitea, searxng, vpn docs

Live read-only probes on Dankclaw (BatchMode ssh, 2026-10-19):

- system/dankclaw.md: full 32-container census added (authentik IdP,
  prometheus/grafana/loki/alloy/alertmanager monitoring, pihole, ntfy,
  seerr, hermes stack, anubis, frank-tts :7772 all undocumented);
  dragoon pool corrected 401,408 @ ctx 200,704 (was conflated);
  zealot corrected to PrismML build 10777, Ternary-Bonsai-2-9B
  PQ2_0-MTP, 2x262,144 over one 524,288 pool, f16 K+V (doc claimed
  build 10754, 27B, 2x200,704)
- caddy/index.md: v2.11.7 (in-container ), deduped
  caddyfile listing
- gitea: version 1.27.3 probed via API; systemd-gitea network + rootless
  100999 mapping confirmed live (gitea-server attaches systemd-gitea;
  dank subuid 100000:65536); index See-Also de-duplicated
- searxng-tuning.md: live instance section (ghcr image, valkey:9-alpine,
  marreta netns-sharing explains duplicate port maps); POLICY DRIFT FLAG:
  tuned engine policy (google/bing disabled, verified 2026-08-16) is NOT
  live — /config now shows all 262 engines enabled; upstream settings.yml
  count corrected 388@master; limiter table already matched source
- vpn-routing.md: verify snippet fixed — gluetun image ships no curl;
  health server is 127.0.0.1:9999/health ('OK'), :8000 is the CONTROL
  server (404s on health paths); correct probes provided

Refs #29

2026-10-19 04:12:04 -07:00

4.5 KiB

Raw Permalink Blame History

type, title, description, tags, generated, verified

type

title

description

Dankclaw

Strix Halo mini-PC carrying the fleet's secondary models and the bulk of the service stack. Two GPU-class devices: the Radeon 8060S iGPU with 119 GiB unified memory, and the RTX 3090 moved over from Dankserver in September 2026.

Service Stack (live census, 32 containers, 2026-10-19)

User-scope Podman Quadlet units (~/.config/containers/systemd/) unless noted. Full names; detail pages where they exist.

Group	Containers
Inference (root systemd, not quadlet)	`frank-dragoon` (:7770), `frank-zealot` (:7771)
TTS	`frank-tts` (:7772), `frank-tts-engine` (:8880)
Git forge	`gitea-server` (:3000/:2222), `gitea-postgres` — see /tools/gitea/
Web/search	`caddy` (:80/:443, Porkbun build), `searxng` (:8081/:8082 via `gluetun`), `searxng-valkey`, `marreta`, `marreta_browser`, `anubis` (bot shield), `chrome-devtools-protocol` (headless-shell)
Identity	`authentik-server` (:9000), `authentik-worker`, `authentik-outpost`, `authentik-outpost-ldap` (:389/:636), `authentik-postgres`
Monitoring	`prometheus` (:9090), `grafana` (:3002), `loki`, `alloy`, `alertmanager` (:9093), `node-exporter` — loopback-only ports
Media/apps	`open-webui` (:3005), `comfyui` (:8188, ROCm `gfx1151`), `seerr` (:5055), `hermes`/`hermes-webui`/`hermes-dashboard` (:9119), `ntfy` (pubsub), `pihole` (:53 DNS, :8083 admin)

Connection Details

Property	Value
Hostname	dankclaw
IP Address	192.168.0.143 (LAN)
Auth	none required on :7770 (verified 2026-10-18)

Inference Services

Model ID	Unit	Endpoint	Engine / weights
`frank-dragoon`	`frank-dragoon.service` (root)	`:7770` (container `frank-dragoon-halogen`, 7770→8731)	halogen-flash-server container (`ghcr.io/peonist-ai/halogen-flash-server`, digest-pinned) — Quat3rnion NATIVE v2 abliterated Qwen3.8-Flash-Next, MTP spec decode, 2 slots over ONE 401,408-position pool at ctx 200,704, vision ON, Strix Halo iGPU (unified memory)
`frank-zealot`	`frank-zealot.service` (root)	`:7771`	llama.cpp (PrismML fork, build 10777) — Ternary-Bonsai-2-9B PQ2_0-MTP GGUF (qwen35 hybrid), 2×262,144 slots over one 524,288 pool, unquantized f16 K+V, MTP off, mmproj Q8_0 vision, RTX 3090

Harness-side context windows (agent/models.yml): dragoon 155,648, zealot 200,704 per slot. The models.yml display labels "(dankclaw-gpu)" / "(dankclaw-3090)" predate naming cleanup; the hardware column above is current. Subagent roles scout, sonic, designer, librarian, security-reviewer route to frank-zealot; reviewer/task run on frank-archon (Dankserver).

curl http://192.168.0.143:7770/v1/models     # dragoon
curl http://192.168.0.143:7771/v1/models     # zealot

Hardware (verified live 2026-10-18)

Property	Value
CPU/APU	AMD Ryzen AI MAX+ 395 (Strix Halo, Radeon 8060S, gfx1151)
dGPU	NVIDIA GeForce RTX 3090 (moved from Dankserver 2026-09)
RAM	121 GiB (mostly usable as GPU unified memory)
Swap	64 GiB
OS	Ubuntu 26.04 LTS, kernel 7.0.0-38-generic

The pre-rename story (frank-default-large on ROCmFPX llama.cpp/CIRU, 262,144-token context) was superseded by the 2026-10-08/09 model-id rename and the dragoon/zealot engine swaps; the 119→121 GiB RAM figure follows the live free reading.

serverAppConfigs — deployment repo + vault, on this workstation at ~/Projects/serverAppConfig/ (docs vault: ~/Projects/serverAppConfig/serverAppConfigs-docs/). Live runbooks: LLM/frank-dragoon-flash-server-v2-native.md and LLM/frank-zealot-llama-prism-bonsai2-pq20.md — deploy-side source of truth for engines, flags, and restart procedures. Service stack docs under Dankclaw/, Infrastructure/.

4.5 KiB Raw Permalink Blame History Unescape Escape