vh/esh-pfi-infrastructure
1
0
Fork 0
Code Issues 7 Pull Requests Actions Packages Projects Releases Wiki Activity
Files
935b35ac2e9cf1f546c8ff4d6faf1c5057a77d6f
esh-pfi-infrastructure/playbooks/homepage-regroup-ana-docker.yaml
T
vh 39da1d4a97 feat(homepage): recategorise on "do I open this?", collapse the API groups 
The board mixed tools with endpoints. A vLLM seat whose href is a /docs page
sat in the same band as ComfyUI; the MQTT broker and the RustDesk relay, which
have no page at all, sat in Apps; and `Service Networking` was thirteen members
spanning three AdGuards, five Dockges, two Traefiks and four headless agents.

Every group is now one of two kinds and they never mix. TOOLS are expanded and
sit at the top of their tab. ENDPOINTS — an API, a broker, a background agent,
an href that is /docs or /ping or nothing — carry `initiallyCollapsed: true`
and sit at the bottom. Collapsed is not hidden: the eyebrow and its rule still
render, so the tab still says the thing exists and one click expands it.

A second rule fell out of the same pass and now shapes the group boundaries: a
group's members should all carry a widget or none should. A stat strip makes a
card ~50px taller, so one widget card in a row of plain ones opens a void under
the plain ones. That is why AdGuard and Traefik get their own groups rather
than sharing one with Dockge, and it is most of why the old Service Networking
band looked broken. AdGuard (ANA) was the last short card in its row and now
carries the same query/blocked/latency strip as its two siblings — one
infra-ops AdGuard login authenticates against all three instances, verified
against each; it lives in that stack's .env on the host and is vaulted.

The sixteen GPU-backed model seats were deliberately NOT relabelled.
`homepage.group` is read at container creation, so clearer names for
`AI - Inference` and friends would have cost a recreate on six vLLM seats, four
eval seats and four TTS engines — multi-minute model reloads on endpoints peers
reach through the gateway. Order plus `initiallyCollapsed` buys the same
separation for nothing, so those names stay as they are on purpose.

28 containers that ARE cheap to bounce were relabelled, across five hosts, via
rerunnable elway playbooks. Their label steps are gated on the old value still
being present, so a second run reports skipped rather than churning. Two verify
steps were wrong on first contact and are fixed with the reason recorded: the
traefik check raced its own recreate, and asserting a model seat is "running"
cannot answer "did I bounce it" when a seat may be legitimately stopped —
container age can, and now does.

The canonical stacks/ tree was synced to the deployed labels afterwards, so
intent and reality agree again on all fourteen tracked stacks.

Also documents the real nature of the post-recreate blank dashboard, which cost
~25 minutes here and an hour on 2026-08-19. `initialSettings":{}` in the served
HTML is the catch branch of the page's data loader, not a warm-up and not a
cache — and the error can vanish entirely, because the logger is assigned inside
the same try and the catch only logs if the logger exists. Ruled out by
measurement this time: all four API routes return 200 with correct content while
the page serves {}, and the previous known-good settings.yaml reproduces it
identically. The README now carries the one-command test and the next lead.

Before/after, all four tabs: http://10.100.10.50:8090/b/homepage-relayout/
2026-08-24 08:54:06 -07:00

227 lines
9.0 KiB
YAML
Raw Blame History

# Homepage recategorisation — ana-docker (10.250.50.70), 13 containers.
#
# Splits the dashboard on ONE axis: do you open this thing, or is it an
# endpoint you only want to know is alive? See the layout: block in
# stacks/homepage/conf/settings.yaml for the target shape.
#
# memos, miniflux, nevermore, searxng -> Daily (was Notes / News / Apps)
# zed-fim-proxy -> AI - Inference (no UI; href is /ping)
# adguardhome -> DNS & Filtering
# traefik -> Reverse Proxies
# dockge -> Compose Consoles
# crowdsec, mailrise, rest-server,
# gitea-runner, hbbr (RustDesk relay) -> Agents (no UI)
#
# `homepage.group` is read at container CREATION, so each edit is followed by
# `compose up -d <service>` — a restart would leave the old label in place.
# Both halves are idempotent: the sed is gated on the old value still being
# present, and `up -d` is a no-op when the container already matches its spec.
#
# Run: scripts/elway infra-ops@10.250.50.70 --playbook playbooks/homepage-regroup-ana-docker.yaml
steps:
# ---- label edits -------------------------------------------------------
- name: memos -> Daily
sudo: true
shell: >-
sed -i 's|homepage.group=Notes$|homepage.group=Daily|'
/opt/docker/compose/memos/compose.yaml
when: grep -q 'homepage.group=Notes$' /opt/docker/compose/memos/compose.yaml
- name: miniflux -> Daily
sudo: true
shell: >-
sed -i 's|homepage.group=News$|homepage.group=Daily|'
/opt/docker/compose/miniflux/compose.yaml
when: grep -q 'homepage.group=News$' /opt/docker/compose/miniflux/compose.yaml
- name: nevermore -> Daily
sudo: true
shell: >-
sed -i 's|homepage.group=News$|homepage.group=Daily|'
/opt/docker/compose/nevermore/compose.yaml
when: grep -q 'homepage.group=News$' /opt/docker/compose/nevermore/compose.yaml
- name: searxng -> Daily
sudo: true
shell: >-
sed -i 's|homepage.group=Apps$|homepage.group=Daily|'
/opt/docker/compose/searxng/compose.yaml
when: grep -q 'homepage.group=Apps$' /opt/docker/compose/searxng/compose.yaml
- name: zed-fim-proxy -> AI - Inference
sudo: true
shell: >-
sed -i 's|homepage.group=AI - Gateways . Chat$|homepage.group=AI - Inference|'
/opt/docker/compose/zed-fim-proxy/compose.yaml
when: grep -q 'homepage.group=AI - Gateways . Chat$' /opt/docker/compose/zed-fim-proxy/compose.yaml
- name: adguardhome -> DNS & Filtering
sudo: true
shell: >-
sed -i 's|homepage.group=Service Networking$|homepage.group=DNS \& Filtering|'
/opt/docker/compose/adguard-ana/compose.yaml
when: grep -q 'homepage.group=Service Networking$' /opt/docker/compose/adguard-ana/compose.yaml
- name: traefik -> Reverse Proxies
sudo: true
shell: >-
sed -i 's|homepage.group=Service Networking$|homepage.group=Reverse Proxies|'
/opt/docker/compose/traefik/compose.yaml
when: grep -q 'homepage.group=Service Networking$' /opt/docker/compose/traefik/compose.yaml
- name: dockge -> Compose Consoles
sudo: true
shell: >-
sed -i 's|homepage.group=Service Networking$|homepage.group=Compose Consoles|'
/opt/docker/compose/dockge/compose.yaml
when: grep -q 'homepage.group=Service Networking$' /opt/docker/compose/dockge/compose.yaml
- name: crowdsec -> Agents (no UI)
sudo: true
shell: >-
sed -i 's|homepage.group=Service Networking$|homepage.group=Agents (no UI)|'
/opt/docker/compose/crowdsec/compose.yaml
when: grep -q 'homepage.group=Service Networking$' /opt/docker/compose/crowdsec/compose.yaml
- name: mailrise -> Agents (no UI)
sudo: true
shell: >-
sed -i 's|homepage.group=Service Networking$|homepage.group=Agents (no UI)|'
/opt/docker/compose/mailrise/compose.yaml
when: grep -q 'homepage.group=Service Networking$' /opt/docker/compose/mailrise/compose.yaml
- name: rest-server -> Agents (no UI)
sudo: true
shell: >-
sed -i 's|homepage.group=Service Networking$|homepage.group=Agents (no UI)|'
/opt/docker/compose/rest-server-ana/compose.yaml
when: grep -q 'homepage.group=Service Networking$' /opt/docker/compose/rest-server-ana/compose.yaml
- name: gitea-runner -> Agents (no UI)
sudo: true
shell: >-
sed -i 's|homepage.group=Toolchain$|homepage.group=Agents (no UI)|'
/opt/docker/compose/gitea-runner/compose.yaml
when: grep -q 'homepage.group=Toolchain$' /opt/docker/compose/gitea-runner/compose.yaml
- name: rustdesk (hbbr) -> Agents (no UI)
sudo: true
shell: >-
sed -i 's|homepage.group=Apps$|homepage.group=Agents (no UI)|'
/opt/docker/compose/rustdesk/compose.yaml
when: grep -q 'homepage.group=Apps$' /opt/docker/compose/rustdesk/compose.yaml
# Two containers were both named plain "Open WebUI" and, once the ESH one
# joined this group, they landed side by side — same name, same icon family,
# only the description telling them apart. Site suffix, like Traefik/Dockge/
# AdGuard already carry.
- name: openwebui (ana) -> "Open WebUI (ana)"
sudo: true
shell: >-
sed -i 's|homepage.name=Open WebUI$|homepage.name=Open WebUI (ana)|'
/opt/docker/compose/openwebui/compose.yaml
when: grep -q 'homepage.name=Open WebUI$' /opt/docker/compose/openwebui/compose.yaml
# ---- recreates ---------------------------------------------------------
# traefik goes LAST: crowdsec is its bouncer, so bounce the bouncer first
# and let traefik come up against a settled agent.
- name: recreate memos
sudo: true
shell: cd /opt/docker/compose/memos && docker compose up -d memos
- name: recreate miniflux
sudo: true
shell: cd /opt/docker/compose/miniflux && docker compose up -d miniflux
- name: recreate nevermore-web
sudo: true
shell: cd /opt/docker/compose/nevermore && docker compose up -d nevermore-web
- name: recreate searxng
sudo: true
shell: cd /opt/docker/compose/searxng && docker compose up -d searxng
- name: recreate zed-fim-proxy
sudo: true
shell: cd /opt/docker/compose/zed-fim-proxy && docker compose up -d zed-fim-proxy
- name: recreate mailrise
sudo: true
shell: cd /opt/docker/compose/mailrise && docker compose up -d mailrise
- name: recreate rest-server
sudo: true
shell: cd /opt/docker/compose/rest-server-ana && docker compose up -d rest-server
- name: recreate gitea-runner
sudo: true
shell: cd /opt/docker/compose/gitea-runner && docker compose up -d runner
- name: recreate rustdesk relay
sudo: true
shell: cd /opt/docker/compose/rustdesk && docker compose up -d hbbr
- name: recreate openwebui (ana)
sudo: true
shell: cd /opt/docker/compose/openwebui && docker compose up -d open-webui
- name: recreate dockge
sudo: true
shell: cd /opt/docker/compose/dockge && docker compose up -d dockge
- name: recreate adguardhome
sudo: true
shell: cd /opt/docker/compose/adguard-ana && docker compose up -d adguardhome
- name: recreate crowdsec
sudo: true
shell: cd /opt/docker/compose/crowdsec && docker compose up -d crowdsec
- name: recreate traefik
sudo: true
shell: cd /opt/docker/compose/traefik && docker compose up -d traefik
verify:
- name: every relabelled container now carries its new group
sudo: true
changed_when: "false"
shell: >-
docker inspect -f '{{.Name}} {{index .Config.Labels "homepage.group"}}'
memos miniflux nevermore-web searxng zed-fim-proxy adguardhome traefik
dockge crowdsec mailrise rest-server gitea-runner hbbr
- name: no container is left in the retired Service Networking group
sudo: true
changed_when: "false"
shell: >-
test -z "$(docker ps -q --filter 'label=homepage.group=Service Networking')"
# dig is not installed everywhere in the fleet, so fall back to the AdGuard
# UI — a resolver that serves its own dashboard on :8053 has come back up.
- name: adguard is back (DNS answer, or its UI if dig is absent)
changed_when: "false"
shell: >-
if command -v dig >/dev/null 2>&1;
then dig +short +time=3 +tries=2 @10.250.50.70 gitea.phasefinal.com | grep -q .;
else curl -sf -o /dev/null -m 8 http://10.250.50.70:8053/; fi
# Retried, not one-shot: the first run of this playbook checked 0.12s after
# `Started` and got rc=7 while traefik was still binding. The container was
# fine — `:8380/` 301s to /dashboard/ and both public hostnames answered 200
# seconds later. A recreate needs a moment; assert the settled state.
- name: traefik still routes
changed_when: "false"
shell: >-
for i in 1 2 3 4 5 6 7 8 9 10; do
curl -sfL -o /dev/null -m 5 http://127.0.0.1:8380/dashboard/ && exit 0;
sleep 3; done; exit 1
- name: everything is running
sudo: true
changed_when: "false"
shell: >-
test "$(docker inspect -f '{{.State.Running}}' memos miniflux nevermore-web
searxng zed-fim-proxy adguardhome traefik dockge crowdsec mailrise
rest-server gitea-runner hbbr | sort -u)" = "true"
Reference in New Issue View Git Blame Copy Permalink