A socket bound to 127.0.0.1 refuses container traffic that arrives via
the podman bridge: the first live webhook failed with 'dial tcp
169.254.1.2:18093: connection refused'. The relay now binds the
host-gateway endpoint — link-local to the podman bridge, off the LAN —
and the deployer's health probe follows it there.
Context: 262144 was measured before it was deployed -- 24.72 GB on GPU 0 and
29.14 GB on GPU 1 of 34.21 GB each, ~5 GB of headroom on the busier card, and a
140,069-token prompt prefilled through the single slot in 134 s to prove the
window is real. The 114688 wall was the vLLM predecessor's arithmetic, not this
runtime's.
Throughput, measured through the service: 1.0-1.4k tok/s prefill from 4k to
140k, 43.8 tok/s decode on a 512-token generation, 41-71 % DFlash2 acceptance,
and the prefix cache re-prefilling 4 of 998 tokens on a repeated prompt.
The signals move with the runtime, so the fleet's rules and dashboard are
re-cut rather than renamed: vllm: series do not exist any more, and two of the
old rules watched things llama.cpp does not export at all (KV-pool fill, p95
latency). Those are retired with the reason written down, and LlmSpecDecodeOff
takes one of the slots -- a lost drafter is the failure no client can see. The
observability seam now fails if any rule names a series the endpoint does not
export, which is what would have caught the stale vllm: expressions.
omp models.yml (outside the repo) is updated in the same pass: contextWindow
262144, and the full reasoning-effort map, because Qwen3.8's template answers
500 on 'high' -- verified against the live endpoint, low/medium/xhigh answer
200. Subagents run at high.
The runbook for the new deployment, plus the sweep of everything that still
described the vllm-radiance replica pair: the vault index and its ACTIVE
marker, the superseded banner on the radiance page, the pointer from the
ROCmFPX page, the homepage card, the Grafana dashboard description and the
observability page's scrape table.
The new page carries the two facts a future agent needs and the flags do not
say: llama.cpp's layer split is what lets BOTH cards serve ONE model without
any collective (vLLM's PP handoff is an RCCL send/recv, which is the primitive
this host cannot dispatch), and the rocBLAS Tensile-path trap that makes a
healthy-looking build fail on the first request.
frank-large has no /metrics to scrape any more, and scraping it would be red by
construction: the halogen engine publishes /health, /cache and /v1/* only. That
job is removed with the reason written down (the remedy is the blackbox
exporter this fleet already wants for the comfyui/services probes), and
frank-small's llama.cpp endpoint takes the slot at 192.168.0.143:7771.
Homepage: frank-large's card still described the ik runtime, and the new
endpoint had no card at all. Both fixed, with a loopback probe -- the checker
runs on Dankclaw, which is also what makes the card green-able.
Dashboard: the endpoint picker was sourced from a vLLM-only metric, so a
llama.cpp job could never be selected, and the prose named frank-large as the
llama.cpp endpoint. The picker now matches both runtimes' series and the
llama.cpp targets carry the job filter every other panel already used.
The seams assert the new contract including the ABSENCE of a 7770 target, so a
well-meaning edit cannot re-add a permanently-red probe.
The board showed mariadb, valkey and Postgres as "down". None of them is down:
the checker runs on Dankclaw and can reach that host's loopback, its container
names, and published LAN addresses — while all three cards probe
container-internal database ports (192.168.0.235:3306, 127.0.0.1:6379/:5432).
They were red by construction, which is how a status board stops being read.
Publishing a database port on the LAN to make a badge green is not a trade this
fleet makes, so they are wip with the reason in their description, and the apps
that front them carry the status. Gluetun (WireGuard, UDP) was the pre-existing
example of the same call.
The rule is now written down in the Homepage doc: a card must be probeable from
the monitor's host or be wip. That is the same lesson as the Prometheus jobs
that scrape HTML pages and can never report up — an always-red signal teaches
the operator to ignore the surface it lives on.
The card probing 192.168.0.235:7770 — frank-medium's endpoint since the
radiance cutover — was still called "frank-small", a service retired on
2026-09-15. The probe target was right; only the identity was wrong, which is
the worse half: a green card named after a service that no longer exists teaches
the wrong thing about the fleet. It is now frank-medium, described as what it
actually is (Qwen3.8-27B Uncensored MXFP4, vllm-radiance, two R9700s behind one
endpoint), and frank-large's description stops crediting the retired halogen
runtime for work ik_llama.cpp does.
The seam added to homepage_config.bats pins the cross-file contract that let
this drift: a card's probe target and its name must agree with the fleet, and a
retired service must not keep a card.
The scraped metrics have no rules over them, so every graph was read after the
fact by a human. This adds shared/prometheus/rules/llm-fleet.yml, loaded via
rule_files and installed by the same copier that ships prometheus.yml (a rule
file nothing loads evaluates nothing, and Prometheus reports no error for it):
LlmKvPoolSaturated kv pool > 90% for 10m — the 122,636-token wall
LlmRequestsQueueing waiting > 0 for 10m — MAXSEQS=4 reached
LlmPrefixCacheCollapse reuse < 20% for 15m with traffic — the thrash signature
LlmEndpointDown scrape target down 2m — both replicas gone
LlmLatencyHigh p95 end-to-end > 60s for 10m — what an operator feels
job:llm_prefix_cache_reuse:rate5m — recording rule shared by the alert and the
dashboard, so the two cannot drift
Every threshold is tied to something measured on this fleet, not to a guess,
and the file says plainly at the top that **nothing is notified**: there is no
Alertmanager here, so a firing alert reaches Prometheus' /alerts page, the
ALERTS metric and the dashboard panel added alongside it — and nowhere else.
Choosing how Frank actually gets paged is his call, not a missing config line.
Seam covers the silent failure mode: rules exist, the config references the
directory, the deployer installs it, and the dashboard graphs ALERTS.
Grafana has been running with zero datasources and zero dashboards, so the
metrics scraped last commit were only reachable through the Prometheus API.
This wires the plumbing and one dashboard, both declared in
shared/grafana/provisioning/ and installed to /opt/grafana/provisioning by the
deployer (the unit already mounted that directory read-only, so nothing about
the container changes):
- datasources/prometheus.yml — Prometheus at http://prometheus:9090, the
container name, not the loopback publish, with a fixed uid so dashboards
reference it without a lookup.
- dashboards/llm-fleet.json — KV pool fill, queue depth, prefix-cache reuse
rate, e2e latency p95, and decode/prefill throughput, with an endpoint
variable covering both frank-medium and frank-large. Every series name was
checked against the live /metrics endpoint: this fleet already shipped one
comment naming a metric that does not exist, and a dashboard full of those
would be worse.
Dashboards stay editable in the UI (allowUiUpdates) because the panels are a
starting point, but the file is the source of truth and the deployer
overwrites Grafana's copy on every deploy.
Planner and seams follow the existing pattern: shared/grafana/** selects the
grafana service, and the tests pin the container-name address plus the
provisioning contract so a rename cannot silently empty the dashboard.
vllm:prefix_cache_hit_rate does not exist in this build; the exported pair is
vllm:prefix_cache_hits_total / vllm:prefix_cache_queries_total, whose rates give
the hit rate. A comment naming a series nobody can query is worse than no
comment.
Neither LLM endpoint was scraped, and frank-medium exports 418 vllm: series —
including the ones this fleet argues about under load: kv_cache_usage_perc (how
full the per-card pool is), prefix_cache_hit_rate (whether pinned conversations
are being reused), num_requests_waiting (queue depth) and the request-latency
histograms. Everything diagnosed tonight by hand — 122,636 KV tokens per card,
thrash when three large contexts overlap, cold-vs-warm prefix reuse — is a
graph now, on the LAN addresses rather than the guarded vhosts.
frank-large comes along in the same change: llama.cpp exposes throughput
counters under llamacpp:, and watching one LLM endpoint without the other was
arbitrary.
The seam added to observability_quadlet.bats pins the cross-file contract that
breaks silently: if a scrape target drifts from the :7770 the units publish,
the graphs vanish with no error anywhere.
Adds the Marreta content-reader tile (http://marreta.franksdank.cloud,
probe 127.0.0.1:8082 via gluetun's publish) to the user grid so regular
authenticated users get a link to it. Registry-driven e2e picks it up
automatically.
frank-large moves from the kingjones30/ROCmFPX llama.cpp build to
peonist-ai/halogen-flash-server 0.5.8 serving Qwen3.8-Flash-Next (w4b
.hgn 115.55 GiB + quality overlay). Upstream measures this engine at
roughly 4x the prefill throughput of the llama.cpp family on the same
gfx1151 silicon, and prefill dominates any prompt with real context.
Shape: a rootless Quadlet container (dankclaw/frank-large/quadlet/),
engine on container loopback :8730 (never published), OpenAI front-end
:8731 published as host 7770, HALOGEN_MODEL_ID=frank-large so no client
renamed anything. Context 262144, KV pool 262144 positions (~7.2 GiB,
conservative for a shared 121 GiB host), 4 slots, host reserve 24,
vision tower on, weights pinned, all other flags at vendor defaults.
Weights are fetched by HALOGEN_DOWNLOAD on first start into a writable
/home/dank/models/halogen.
The llama.cpp era (unit, env, serve script) is staged in .AGENTTRASH/
rather than deleted; the STRIX payload and the ROCmFPX build stay on
disk as the rollback target. deploy_llm_large becomes a user-quadlet
handler: it retires the old ROOT unit (moving its checkout symlink into
the home trash, never rm) and gates the first start on the download
rather than reporting a failure for a fetch that is working exactly as
designed.
frank-npu is retired in the same change. The kernel cmdline moves from
iommu=pt to amd_iommu=off for the 13-16% of prefill upstream measured,
and that takes the XDNA2 NPU with it: without PASID/SVA the driver
cannot open /dev/accel/accel0 (err=-19). Its five dependent routes are
repointed (Hermes vision -> frank-large, Hermes title generation ->
frank-small, harness vision -> frank-large, tiny -> frank-small, task ->
frank-large:medium) and its provider blocks are removed from both the
harness config and Hermes's live config. Audio input leaves the fleet
with no replacement; image input survives on halogen's vision tower.
Docs: new halogen-flash-server runbook, new IOMMU decision record, and
frank-npu rewritten as a retirement record. AGENTS.md topology, runtime
sections, benchmarking policy (frank-large is a user unit now) and the
harness model-config section are corrected.
Flip the homepage Sonarr card from the user category to admin and repoint
its link at the LAN-direct port (192.168.0.112:8989). Retire the unguarded
public vhost sonarr.franksdank.cloud (direct reverse_proxy, no auth) and its
post-auth @sonarr handler from the Caddyfile; drop the Pi-hole addn-hosts A
record with it. No public Porkbun record ever existed (NXDOMAIN at the zone),
so after this there is no WAN path to Sonarr: users request TV via Seerr
(guarded), admins reach Sonarr's own UI from the LAN only.
Fixes the seerr/sonarr mix-up from 19b2589/e0bade6.
FrankBook-Pro becomes the only *arr/qBittorrent deployment. Every fleet
container (qB + five arrs) now shares gluetun's netns, so all egress —
torrent, indexer APIs, metadata lookups — exits the WireGuard tunnel by
construction (AllowedIPs=0.0.0.0/0); the arrs' own ports publish on the
gluetun unit, none on the arr units themselves.
- dankserver/compose.yaml: remove the six arr services (the Aug 20 compose
copy — never healthy, superseded; config dirs trashed on the host)
- dankclaw/Caddyfile + shared/homepage/services.json: arr vhosts and
probes repoint 192.168.0.235 -> 192.168.0.112 (FrankBook-Pro)
- frankbook-pro units: five arrs join gluetun netns (Wants/After/Network);
gluetun publishes 9696/8989/7878/8686/8787 alongside 8080/6881
- tests/frankbook_quadlet.bats: new VPN egress seam — every arr+qB unit
must share gluetun's netns and publish no port; gluetun must publish the
whole fleet's ports
- deploy-plan.py: gluetun changes carry the whole netns family (a gluetun
recreation kills all six joiners); dankserver compose set loses the arrs
- AGENTS.md: deployment is complete only when unit tests AND the Playwright
E2E suite both pass
Mechanical identity rename across the knowledge base, AGENTS.md, README, the
hermes config example, and homepage services.json. No behavior change; served
aliases on the LLM endpoints change with the deploy of commit 69f87c8.
PMS 1.43 serves TLS-only on 32400: a plain-HTTP GET gets a TLS alert and
curl reports "empty reply from server" (exit 52, code 000), so the card read
down while Plex answered 200 at /web/index.html over HTTPS. TCP connect is the
honest liveness signal for this port; the public URL stays HTTP because Caddy
terminates TLS and re-keys the path.
The Hermes card probed 127.0.0.1:8090, a port retired when the hermes
stack moved from compose (8090 web dashboard) to quadlets: the dashboard
now publishes 9119 and the webui listens on 0.0.0.0:8787 in the host netns.
Nothing has listened on 8090 since, so the card reported "down" while the
gateway, dashboard, and webui were all running.
Probe 127.0.0.1:8787 instead — the webui runs HERMES_HOME=/data, the same
agent state as the gateway, and /health answers only when that server is
alive (uptime, sessions, active runs). The prober runs on Dankclaw, so the
loopback probe hits the host netns where the webui binds.
All in-repo path references (AGENTS.md, README, old scripts, report artifacts,
test code paths) rewritten in the same commit; focused-test map extended for
.gitignore, old/scripts/*, dankclaw/comfyui/*, and report/* per this repo's
own convention ('add one before committing'). Obsidian wikilinks are
title-based and unaffected.
Two real bugs found on the first production deploy:
- /admin/ 404'd: file_server looked for /srv/homepage/admin/ (a dir), but the
page is admin.html. Add '@admin rewrite * /admin.html' so the /admin/ path
serves admin.html.
- The pages used relative refs (style.css, app.js, services.json, status.json),
which from /admin/ resolved to /admin/... (404). Add <base href="/"> so all
refs resolve to the site root on both pages.
- status.json was written 0600 (root); the checker now chmods it 0644 so Caddy
can serve it regardless of its user.
Red tests first, then the pages.
- tests/homepage_pages.bats: the pages reference the data + theme; no
auto-refresh (Q41); no cursor alteration (Q28); the styling toggle strips
to bare HTML (Q27); the footer shows the last master deployment datetime
(Q43).
- index.html / admin.html: the user and admin pages. A grid of TUI panels,
a summary line, the bare toggle, the deployed-at footer.
- app.js: reads status.json + services.json once (no auto-refresh), renders
one panel per service for the page's category, the toggle, the footer.
- style.css: the terminal TUI theme (green-on-black, phosphor palette,
scanlines) + a ?bare=1 mode that strips to unstyled HTML. No custom cursor.
Functionally verified with a Node DOM shim: 9 user panels, correct summary,
correct footer datetime.
Red tests first (the old services.json was an object, not the new list schema).
- tests/homepage_config.bats: the config is a JSON list; every entry has
name/category/probe; category is user|admin; method is tcp|http; a
showLink entry has a url and a status-only entry does not.
- shared/homepage/services.json: the real fleet. User: SearXNG, ComfyUI,
OpenWebUI, Gitea, Calibre-Web, RomM, Copyparty, Plex, Jellyfin. Admin:
Hermes, Grafana, Prometheus, Loki, node-exporter, alloy, Caddy, dnsmasq,
Gluetun (wip - a tunnel has no simple local port), the two LLMs, mariadb,
valkey, Postgres. Ports confirmed from the repo config (compose/quadlet).
docs fields are empty for now (the /docs/ workstream fills them as each
service goes active).
- alpine's /bin/busybox is musl-dynamic and cannot execute in the
distroless loki image (missing loader); busybox-static links in
- alloy 1.x refuses to configure the ui service from a config block;
the default UI already listens on 127.0.0.1:12345 where the
healthcheck probes
- alloy 1.x renamed the server component to ui; the old block failed the
initial load and the container crash-looped
- grafana/loki is distroless and grafana/alloy lacks wget, so their
healthchecks could never run; local images add a static busybox from
alpine for that one command
- deploy.sh builds all three local images if missing
- tests: 3 new seams
- loki: recent releases require compactor.delete_request_store when
retention is enabled (was crash-looping on start)
- prometheus: scrape targets carried legacy dankclaw.local names with
URL schemes (invalid); point at published LAN ports and the gitea
quadlet's published port instead
- grafana: GF_SERVER_ROOT_URL to the canonical HTTPS name
- tests: 3 new seams for the deployed observability configs
Phase 1 of the local-DNS-name work. Users type a name, not a port.
- dnsmasq quadlet on dankclaw (rootless, slirp4netns, port 53 published):
serves the flat name scheme from addn-hosts, upstream 1.1.1.1/1.0.0.1.
Image built from the in-repo Containerfile (localhost/dnsmasq:latest).
- Caddyfiles (both hosts): all *.dankclaw.local / *.dankserver.local vhosts
renamed to flat <service>.franksdank.cloud. Explicit http:// site
addresses for now; TLS (wildcard LE via DNS-01/Porkbun) lands in phase 2.
- Drift reconciliation (live /etc/caddy/Caddyfile on dankserver differs
from the repo): added the live-only llms.franksdank.cloud vhost
(OpenWebUI on :3000). Dropped the legacy search vhost from dankserver
(the name belongs to dankclaw's gluetun-routed SearXNG; the legacy
instance stays reachable at 192.168.0.235:8081).
- Homepage service links updated to the new names.
- New test seams (TDD): tests/dnsmasq.bats (quadlet boot-persistence,
port 53, conf/record validity, Caddyfile<->addn-hosts cross-check on
the right host IPs); caddyfile.bats naming-convention tests updated.
LAN clients get this DNS from the SURFboard G36 (DNS Relay -> statically
configured -> 192.168.0.143); the router change is a pending user step.
Root cause of the 429s: the ip_limit rate limiter caps JSON (format!=html)
requests at API_MAX=4 per IP per hour, and that constant is hardcoded in
Python (searx/botdetection/ip_limit.py), not settable in limiter.toml. The
only bypass is pass_ip (filter_request returns before ip_limit runs).
Rootless podman strips the client IP on published-port hits, so the container
sees a slirp/bridge source (10.88.x/10.89.x), never the real client - hence
even a LAN-IP curl 429'd on /search. Expand pass_ip to loopback + the host's
private podman/slirp ranges (plus the existing LAN) so local subagents present
a trusted source and skip the rate limit entirely. All these ranges are
reachable only from the host or LAN, never the internet.
Converts the podman-compose SearXNG stack to boot-persistent quadlet units so
it survives a reboot (the compose deployment did not). searxng sets
Network=gluetun.container, which makes it reuse gluetun's network namespace so
outbound search traffic exits the WireGuard tunnel; it publishes no port of its
own (the port lives on gluetun). gluetun and valkey join a pinned searxng
network (10.89.2.0/24, DNS-enabled) so searxng reaches valkey by name.
Secrets stay in /opt/searxng/.env (mode 0600), referenced via [Service]
EnvironmentFile= so ${VAR} expands. Adds a bats seam asserting the netns
wiring, port placement, boot install, and that no secret material is tracked.
A direct hit on the published port has no X-Forwarded-For header, so the
bot-detection cannot resolve a client and returns 429. SearXNG must be
reached through Caddy (search.dankclaw.local), which sets X-Forwarded-For.
Add the podman bridge subnets to trusted_proxies so Caddy's source is
trusted for XFF extraction; note the CIDR is dynamic and should be pinned.
Verified agentic JSON search returns 26-31 relevant results via this path.
The trusted group lives in limiter.toml, not settings.yml. Add
shared/searxng/limiter.toml trusting 192.168.0.0/24 in pass_ip and
trusted_proxies, copy it into /opt/searxng/config on deploy, and correct
the SearXNG doc. New test seam: tests/searxng_limiter.bats.
Frank