chore(deploy): rework k8s pipeline, monitoring and postgres 17

Deploy workflow uses git-tracked manifests, DISABLED flag and kustomize overlays; add webinar-checker metrics with ServiceMonitor and alerts; upgrade shared postgres to 17 with statuspage DB and probes/resources.
2026-09-23 15:47:26 +02:00
1 parent 8b2cf29771
commit 46c7e99b1d
@@ -347,23 +347,3 @@ jobs:
                ;;
            esac
          done
-
-  deploy-userbot-panel:
-    needs: build
-    if: github.ref_name == 'main' && contains(needs.build.outputs.services, 'userbot')
-    runs-on: [self-hosted, linux, arch, homelab, prod]
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 # v4
-
-      - name: Apply and roll out userbot panel
-        shell: bash
-        run: |
-          kubectl apply -f userbot/k8s/base/panel.yaml
-          kubectl get secret userbot-common-secrets -n default -o json \
-            | jq 'del(.metadata.annotations,.metadata.creationTimestamp,.metadata.resourceVersion,.metadata.uid,.metadata.managedFields) | .metadata.namespace = "userbot"' \
-            | kubectl apply -f -
-          # Keep legacy deployments (forust/anna) in sync with manifests; they have no replicas field, so apply leaves scaling to the user manager only.
-          kubectl apply -f userbot/k8s/base/userbots.yaml
-          kubectl rollout restart deployment/userbot-panel -n userbot
-          kubectl rollout status deployment/userbot-panel -n userbot --timeout=180s
@@ -19,9 +19,6 @@ jobs:
          DEPLOY_USER: ${{ secrets.DEPLOY_USER }}
          DEPLOY_PATH: ${{ secrets.DEPLOY_PATH }}
          DEPLOY_KEY: ${{ secrets.DEPLOY_SSH_KEY }}
-          # Set APPLY_PRUNE=true to enable kubectl apply --prune. Requires every
-          # manifest to carry label app.kubernetes.io/managed-by=homelab-deploy,
-          # otherwise previously applied resources get deleted on the next run.
          APPLY_PRUNE: ${{ vars.APPLY_PRUNE }}
        run: |
          set -euo pipefail
@@ -57,59 +54,170 @@ jobs:
          fi

          git -C "$repo" fetch origin main
-          git -C "$repo" reset --hard origin/main

-          # Runtime selection: a service is k8s-managed when $SERVICE/k8s/active
-          # exists. Otherwise it is compose-managed, and only k8s/routing/*
-          # manifests (external Services / EndpointSlices / ServersTransport /
-          # Ingresses that route to docker backends) are applied.
-          #   migrate:   touch SERVICE/k8s/active   (+ move routing files up)
-          #   rollback:  rm SERVICE/k8s/active
+          echo "== Workstation state =="
+          echo "  local:  $(git -C "$repo" rev-parse --short HEAD)"
+          echo "  remote: $(git -C "$repo" rev-parse --short origin/main)"
+
+          if [ -n "$(git -C "$repo" status --porcelain --untracked-files=no)" ]; then
+            echo "ERROR: workstation has local tracked modifications, refusing reset:"
+            git -C "$repo" status --porcelain --untracked-files=no
+            git -C "$repo" diff --stat
+            exit 1
+          fi
+
+          git -C "$repo" reset --hard origin/main
+          cd "$repo"
+
+          is_disabled() {
+            local target="$1"
+            if [ -f "$target" ]; then
+              target="$(dirname "$target")"
+            fi
+            while true; do
+              if [ -f "$target/DISABLED" ]; then
+                return 0
+              fi
+              if [ "$target" = "$repo" ]; then
+                break
+              fi
+              target="$(dirname "$target")"
+              case "$target" in
+                "$repo"/*) ;;
+                *) break ;;
+              esac
+            done
+            return 1
+          }
+
          collect_k8s() {
-            find "$1" -type f \( -name '*.yaml' -o -name '*.yml' \) \
-              ! -path '*/routing/*' ! -path '*/overlays/*' \
-              ! -name 'kustomization.y*ml' ! -name '*.example.y*ml' \
-              ! -name '*values.y*ml' ! -name 'patch-*.y*ml' \
+            git ls-files -- "$1" \
+              | grep -E '\.ya?ml$' \
+              | grep -Ev '/routing/|/overlays/' \
+              | grep -Ev '(^|/)(kustomization\.ya?ml|.*\.example\.ya?ml|.*values\.ya?ml|patch-.*\.ya?ml)$' \
+              | grep -Ev '(^|/)[^/]*secret[^/]*\.ya?ml$' \
              | sort
          }

          collect_k8s_inactive() {
-            find "$1" -type f \( -name '*.yaml' -o -name '*.yml' \) \
-              \( -name 'namespace.y*ml' -o -path '*/routing/*' \) \
-              ! -path '*/overlays/*' ! -name '*.example.y*ml' \
-              | sort
+            collect_k8s "$1" \
+              | grep -E '(^|/)namespace\.ya?ml$|/routing/'
          }

-          mapfile -t compose_stacks < <(
-            find "$repo" -type f \( -name 'compose.yaml' -o -name 'compose.yml' \) | sort
+          kustomize_overlay() {
+            if [ -f "$1/overlays/prod/kustomization.yaml" ]; then
+              echo "$1/overlays/prod"
+            elif [ -f "$1/base/kustomization.yaml" ]; then
+              echo "$1/base"
+            fi
+          }
+
+          mapfile -t k8s_dirs < <(
+            git ls-files '*.yaml' '*.yml' \
+              | grep -E '(^|/)k8s/' \
+              | sed -E 's#((^|.*/)k8s)/.*#\1#' \
+              | sort -u
          )

-          mapfile -t k8s_manifests < <(
-            for kd in $(find "$repo" -type d -name k8s ! -path '*/.git/*' | sort); do
+          k8s_manifests=()
+          kustomize_apps=()
+          for kd_rel in "${k8s_dirs[@]}"; do
+            kd="$repo/$kd_rel"
+            if is_disabled "$kd"; then
+              echo "skip (DISABLED): $kd_rel"
+              continue
+            fi
            if [ -f "$kd/active" ]; then
-                collect_k8s "$kd"
+              overlay="$(kustomize_overlay "$kd" || true)"
+              if [ -n "${overlay:-}" ]; then
+                echo "kustomize app: ${overlay#$repo/}"
+                kustomize_apps+=("$overlay")
              else
-                collect_k8s_inactive "$kd"
+                while IFS= read -r f; do
+                  [ -n "$f" ] && k8s_manifests+=("$repo/$f")
+                done < <(collect_k8s "$kd_rel" || true)
+              fi
+            else
+              while IFS= read -r f; do
+                [ -n "$f" ] && k8s_manifests+=("$repo/$f")
+              done < <(collect_k8s_inactive "$kd_rel" || true)
            fi
          done
+
+          mapfile -t compose_rel < <(
+            git ls-files '*/compose.yaml' '*/compose.yml' compose.yaml compose.yml | sort
          )

+          compose_stacks=()
+          for cf_rel in "${compose_rel[@]}"; do
+            cf="$repo/$cf_rel"
+            if is_disabled "$cf"; then
+              echo "skip (DISABLED): $cf_rel"
+              continue
+            fi
+            if [ -f "$(dirname "$cf")/k8s/active" ]; then
+              echo "skip (k8s-managed): $cf_rel"
+              continue
+            fi
+            compose_stacks+=("$cf")
+          done
+
          echo "== Validate compose stacks =="
          for cf in "${compose_stacks[@]}"; do
-            dir=$(dirname "$cf")
-            if [ -f "$dir/k8s/active" ]; then
-              echo "  skip (k8s-managed): $dir"
-              continue
-            fi
            echo "  config: $cf"
            docker compose -f "$cf" config --quiet
          done

-          echo "== Validate k8s manifests (kubectl dry-run) =="
+          echo "== Validate k8s manifests (kubectl dry-run=client) =="
          for m in "${k8s_manifests[@]}"; do
            echo "  apply --dry-run=client $m"
            kubectl apply --dry-run=client -f "$m" >/dev/null
          done
+          for k in "${kustomize_apps[@]}"; do
+            echo "  apply -k --dry-run=client $k"
+            kubectl apply -k "$k" --dry-run=client >/dev/null
+          done
+
+          echo "== Validate k8s manifests (kubectl dry-run=server) =="
+          for m in "${k8s_manifests[@]}"; do
+            echo "  apply --dry-run=server $m"
+            kubectl apply --dry-run=server -f "$m" >/dev/null
+          done
+          for k in "${kustomize_apps[@]}"; do
+            echo "  apply -k --dry-run=server $k"
+            kubectl apply -k "$k" --dry-run=server >/dev/null
+          done
+
+          echo "== Checking referenced Secrets exist =="
+          echo "  (deploy never applies *secret*.yaml; create missing ones from the laptop)"
+          ref_secrets=()
+          if [ "${#k8s_manifests[@]}" -gt 0 ]; then
+            while IFS= read -r s; do
+              [ -n "$s" ] && ref_secrets+=("$s")
+            done < <(
+              {
+                grep -h -A1 -E 'secretRef:|secretKeyRef:' "${k8s_manifests[@]}" 2>/dev/null || true
+                grep -h -E 'secretName:' "${k8s_manifests[@]}" 2>/dev/null || true
+              } | grep -E 'name:' | sed -E 's/.*name:[[:space:]]*//' | tr -d '"'"'"' "'"'" | sed -E 's/[[:space:]]*#.*//' | awk 'NF' | sort -u || true
+            )
+          fi
+          missing_secrets=()
+          all_secrets="$(kubectl get secrets -A --no-headers -o custom-columns=:metadata.name 2>/dev/null || true)"
+          for s in "${ref_secrets[@]}"; do
+            if printf '%s\n' "$all_secrets" | grep -qx "$s"; then
+              echo "  ok: $s"
+            else
+              echo "  MISSING: $s"
+              missing_secrets+=("$s")
+            fi
+          done
+          if [ "${#missing_secrets[@]}" -gt 0 ]; then
+            echo "ERROR: ${#missing_secrets[@]} referenced Secret(s) not found in the cluster:"
+            printf '  - %s\n' "${missing_secrets[@]}"
+            echo "Create them manually from the laptop, e.g.:"
+            echo "  kubectl apply -f SERVICE/k8s/secrets.yaml  # see SERVICE/k8s/secrets.yaml.example"
+            exit 1
+          fi

          echo "== Applying Kubernetes manifests =="
          ns_files=()
@@ -130,15 +238,19 @@ jobs:
            echo "  namespaces first: ${ns_files[*]}"
            kubectl apply -f "${ns_files[@]}"
          fi
-          if [ -f "$repo/prometheus-stack/k8s/active" ]; then
+          if [ -f "$repo/prometheus-stack/k8s/active" ] && ! is_disabled "$repo/prometheus-stack/k8s"; then
+            if [ ! -f "$repo/prometheus-stack/k8s/grafana-values.yaml" ]; then
+              echo "ERROR: prometheus-stack/k8s/grafana-values.yaml (gitignored) missing on workstation, restore it first."
+              exit 1
+            fi
            echo "== Upgrading kube-prometheus-stack =="
            helm upgrade --install prometheus-stack prometheus-community/kube-prometheus-stack \
              --namespace prometheus \
              --version 86.2.3 \
              --values "$repo/prometheus-stack/k8s/grafana-values.yaml" \
-              --wait
+              --wait --timeout 10m
          fi
-          if [ -f "$repo/loki/k8s/active" ]; then
+          if [ -f "$repo/loki/k8s/active" ] && ! is_disabled "$repo/loki/k8s"; then
            echo "== Upgrading loki/alloy =="
            helm repo add grafana https://grafana.github.io/helm-charts >/dev/null 2>&1 || true
            helm repo update grafana >/dev/null 2>&1 || true
@@ -146,12 +258,12 @@ jobs:
              --version 7.3.0 \
              --namespace prometheus \
              --values "$repo/loki/k8s/loki-values.yaml" \
-              --wait
+              --wait --timeout 10m
            helm upgrade --install alloy grafana/alloy \
              --version 1.12.1 \
              --namespace prometheus \
              --values "$repo/loki/k8s/alloy-values.yaml" \
-              --wait
+              --wait --timeout 10m
          fi

          if [ "${#other_files[@]}" -gt 0 ]; then
@@ -159,14 +271,30 @@ jobs:
            kubectl apply "${prune_opts[@]}" -f "${other_files[@]}"
          fi

+          for k in "${kustomize_apps[@]}"; do
+            echo "== Applying kustomize app: ${k#$repo/} =="
+            kubectl apply -k "$k"
+          done
+
+          if [ -f "$repo/userbot/k8s/active" ] && ! is_disabled "$repo/userbot"; then
+            echo "== userbot panel hook =="
+            if kubectl get secret userbot-common-secrets -n userbot >/dev/null 2>&1; then
+              echo "  userbot-common-secrets already present in userbot ns, not touching"
+            elif kubectl get secret userbot-common-secrets -n default >/dev/null 2>&1; then
+              echo "  bootstrapping userbot-common-secrets into userbot ns"
+              kubectl get secret userbot-common-secrets -n default -o json \
+                | jq 'del(.metadata.annotations,.metadata.creationTimestamp,.metadata.resourceVersion,.metadata.uid,.metadata.managedFields) | .metadata.namespace = "userbot"' \
+                | kubectl apply -f -
+            else
+              echo "  WARNING: userbot-common-secrets missing in both default and userbot ns; create it manually from the laptop"
+            fi
+            kubectl rollout restart deployment/userbot-panel -n userbot
+            kubectl rollout status deployment/userbot-panel -n userbot --timeout=180s
+          fi
+
          echo "== Redeploying docker compose stacks =="
          for cf in "${compose_stacks[@]}"; do
-            dir=$(dirname "$cf")
-            if [ -f "$dir/k8s/active" ]; then
-              echo "  skip (k8s-managed): $dir"
-              continue
-            fi
-            echo "  compose: $dir"
+            echo "  compose: $cf"
            if grep -Eq '^\s+pull_policy:\s*build\b' "$cf"; then
              docker compose -f "$cf" build
              docker compose -f "$cf" push
@@ -0,0 +1 @@
+1.56.0
@@ -11,7 +11,7 @@ services:
      retries: 5

  playwright-service:
-    image: mcr.microsoft.com/playwright:v1.63.0-jammy
+    image: mcr.microsoft.com/playwright:v1.56.0-jammy
    restart: unless-stopped
    command: npx -y playwright@1.56.0 run-server --port 3000 --path /ws

@@ -0,0 +1,77 @@
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  name: edu-master-webinar
+  namespace: edu-master
+  labels:
+    release: prometheus-stack
+spec:
+  groups:
+    - name: edu_master.webinar
+      rules:
+        # No successful webinar check for 5m (~2-3 missed 2-min checks).
+        # Catches: playwright hangs/timeouts, version skew, site changes, hung job.
+        - alert: WebinarCheckerNoSuccessfulCheck
+          expr: |
+            (time() - webinar_check_last_success_timestamp_seconds > 300)
+              and (webinar_check_last_run_timestamp_seconds > 0)
+          for: 2m
+          labels:
+            severity: critical
+          annotations:
+            summary: "Webinar checker has no successful check for 5m"
+            description: "edu-master/webinar-checker: last successful webinar check was {{ $value | humanizeDuration }} ago. Checks are failing or hanging (see consecutive failures alert). Notifications about new webinars are NOT being sent."
+
+        # Fast path: 3 consecutive failures (~6+ min at 2-min interval).
+        - alert: WebinarCheckerConsecutiveFailures
+          expr: |
+            webinar_check_consecutive_failures >= 3
+          for: 5m
+          labels:
+            severity: critical
+          annotations:
+            summary: "Webinar checker failing consecutively"
+            description: "edu-master/webinar-checker: {{ $value }} consecutive webinar check failures (timeout / playwright error / page error). Check pod logs (Loki: {namespace=\"edu-master\", container=\"webinar-checker\"})."
+
+        # Metrics endpoint not scraped for 10m: pod down, metrics server dead, or ServiceMonitor broken.
+        - alert: WebinarCheckerScrapeDown
+          expr: |
+            absent(webinar_check_last_run_timestamp_seconds) == 1
+          for: 10m
+          labels:
+            severity: critical
+          annotations:
+            summary: "Webinar checker metrics missing"
+            description: "edu-master/webinar-checker: no metrics series for 10m. Pod may be down, metrics server dead, or ServiceMonitor/Service broken. Webinar checks are unobserved."
+
+        # EDU session lost: session-keeper down or credentials expired. Without PHPSESSID every check is skipped.
+        - alert: EduPhpsessidMissing
+          expr: |
+            edu_phpsessid_present == 0
+          for: 10m
+          labels:
+            severity: critical
+          annotations:
+            summary: "EDU_PHPSESSID missing"
+            description: "edu-master: EDU_PHPSESSID absent from redis for 10m. Webinar/diari/schedule checks are all skipped. Check session-keeper logs and EDU credentials."
+
+        # Hard deps: checker and playwright deployments unavailable.
+        - alert: WebinarCheckerDeploymentDown
+          expr: |
+            kube_deployment_status_replicas_unavailable{deployment="webinar-checker", namespace="edu-master"} > 0
+          for: 10m
+          labels:
+            severity: critical
+          annotations:
+            summary: "Webinar checker deployment unavailable"
+            description: "edu-master/webinar-checker deployment has {{ $value }} unavailable replica(s) for 10m."
+
+        - alert: PlaywrightServiceDown
+          expr: |
+            kube_deployment_status_replicas_unavailable{deployment="playwright-service", namespace="edu-master"} > 0
+          for: 10m
+          labels:
+            severity: critical
+          annotations:
+            summary: "Playwright service unavailable"
+            description: "edu-master/playwright-service deployment has {{ $value }} unavailable replica(s) for 10m. All webinar/diari/schedule checks fail without it."
@@ -17,7 +17,8 @@ spec:
    spec:
      containers:
        - name: playwright
-          image: mcr.microsoft.com/playwright:v1.63.0-jammy
+          # renovate: datasource=docker depName=mcr.microsoft.com/playwright versioning=docker
+          image: mcr.microsoft.com/playwright:v1.56.0-jammy
          imagePullPolicy: IfNotPresent
          command:
            - npx
@@ -21,6 +21,8 @@ stringData:
  WEBINAR_TELEGRAM_TOKEN: ""
  WEBINAR_ADMIN_ID: ""
  WEBINAR_CHECK_INTERVAL: "60"
+  # Prometheus metrics endpoint (scraped via ServiceMonitor, alerts in k8s/alerts.yaml)
+  METRICS_PORT: "8000"
  # Database
  REDIS_HOST: "redis"
  REDIS_PORT: "6379"
@@ -0,0 +1,15 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: webinar-checker
+  namespace: edu-master
+  labels:
+    app: edu-master-webinar-checker
+spec:
+  selector:
+    app: edu-master-webinar-checker
+  ports:
+    - name: metrics
+      port: 8000
+      targetPort: metrics
+      protocol: TCP
@@ -0,0 +1,16 @@
+apiVersion: monitoring.coreos.com/v1
+kind: ServiceMonitor
+metadata:
+  name: webinar-checker
+  namespace: edu-master
+  labels:
+    release: prometheus-stack
+spec:
+  selector:
+    matchLabels:
+      app: edu-master-webinar-checker
+  endpoints:
+    - port: metrics
+      path: /metrics
+      interval: 30s
+      scrapeTimeout: 10s
@@ -47,6 +47,10 @@ spec:
        - name: webinar-checker
          image: gcr.forust.xyz/forust/webinar-checker:latest
          imagePullPolicy: Always
+          ports:
+            - name: metrics
+              containerPort: 8000
+              protocol: TCP
          envFrom:
            - secretRef:
                name: edu-master-secrets
@@ -2,8 +2,11 @@ FROM python:3.11-slim

 WORKDIR /app

-# Install dependencies
-RUN pip install --no-cache-dir pip==25.0.1 && pip install --no-cache-dir playwright==1.56.0 redis==5.2.1 requests==2.32.3 "python-telegram-bot[job-queue]==21.10"
+# renovate: datasource=pypi depName=playwright versioning=pep440
+ARG PLAYWRIGHT_VERSION=1.56.0
+
+# Install dependencies - PLAYWRIGHT_VERSION is single-source, renovate updates ARG above and all other places via regexManagers
+RUN pip install --no-cache-dir pip==25.0.1 && pip install --no-cache-dir playwright==${PLAYWRIGHT_VERSION} redis==5.2.1 requests==2.32.3 "python-telegram-bot[job-queue]==21.10"

 COPY checker.py .

@@ -1,12 +1,15 @@
+import asyncio
 import contextlib
 import json
 import logging
 import os
 import re
 import tempfile
+import threading
 import time
 from datetime import datetime, timedelta
 from html import escape
+from http.server import BaseHTTPRequestHandler, HTTPServer

 import redis
 from playwright.async_api import async_playwright
@@ -48,6 +51,106 @@ USER_AGENT = _env(
 )
 WEBINAR_TELEGRAM_TOKEN = _env('WEBINAR_TELEGRAM_TOKEN')
 ADMIN_ID = int(_env('WEBINAR_ADMIN_ID', '0'))
+METRICS_PORT = int(_env('METRICS_PORT', '8000'))
+
+# --- Prometheus metrics (stdlib only, no extra deps) ---
+# Scraped by prometheus-stack via ServiceMonitor (edu_master/k8s/servicemonitor.yaml).
+# Critical alerts in edu_master/k8s/alerts.yaml fire to Telegram via Alertmanager.
+_METRICS_LOCK = threading.Lock()
+_METRICS = {
+    'last_run': 0.0,  # Unix ts of last check start
+    'last_success': 0.0,  # Unix ts of last successful check
+    'last_duration': 0.0,  # Duration of last check in seconds
+    'success_total': 0,
+    'failure_total': 0,
+    'consecutive_failures': 0,
+    'phpsessid_present': 1,  # 1 if EDU_PHPSESSID found in redis, else 0
+}
+
+
+def _metric_check_start():
+    with _METRICS_LOCK:
+        _METRICS['last_run'] = time.time()
+
+
+def _metric_check_ok(duration: float):
+    now = time.time()
+    with _METRICS_LOCK:
+        _METRICS['last_success'] = now
+        _METRICS['last_duration'] = duration
+        _METRICS['success_total'] += 1
+        _METRICS['consecutive_failures'] = 0
+        _METRICS['phpsessid_present'] = 1
+
+
+def _metric_check_fail(duration: float, phpsessid_missing: bool = False):
+    with _METRICS_LOCK:
+        _METRICS['last_duration'] = duration
+        _METRICS['failure_total'] += 1
+        _METRICS['consecutive_failures'] += 1
+        _METRICS['phpsessid_present'] = 0 if phpsessid_missing else 1
+
+
+def _metrics_render() -> bytes:
+    with _METRICS_LOCK:
+        m = dict(_METRICS)
+    lines = [
+        '# HELP webinar_check_last_run_timestamp_seconds Unix timestamp of last webinar check start.',
+        '# TYPE webinar_check_last_run_timestamp_seconds gauge',
+        f'webinar_check_last_run_timestamp_seconds {m["last_run"]}',
+        '# HELP webinar_check_last_success_timestamp_seconds Unix timestamp of last successful webinar check.',
+        '# TYPE webinar_check_last_success_timestamp_seconds gauge',
+        f'webinar_check_last_success_timestamp_seconds {m["last_success"]}',
+        '# HELP webinar_check_last_duration_seconds Duration of last webinar check in seconds.',
+        '# TYPE webinar_check_last_duration_seconds gauge',
+        f'webinar_check_last_duration_seconds {m["last_duration"]}',
+        '# HELP webinar_check_success_total Total successful webinar checks.',
+        '# TYPE webinar_check_success_total counter',
+        f'webinar_check_success_total {m["success_total"]}',
+        '# HELP webinar_check_failure_total Total failed webinar checks (timeout, playwright error, page error).',
+        '# TYPE webinar_check_failure_total counter',
+        f'webinar_check_failure_total {m["failure_total"]}',
+        '# HELP webinar_check_consecutive_failures Consecutive failed webinar checks (reset on success).',
+        '# TYPE webinar_check_consecutive_failures gauge',
+        f'webinar_check_consecutive_failures {m["consecutive_failures"]}',
+        '# HELP edu_phpsessid_present 1 if EDU_PHPSESSID exists in redis, 0 otherwise.',
+        '# TYPE edu_phpsessid_present gauge',
+        f'edu_phpsessid_present {m["phpsessid_present"]}',
+    ]
+    return ('\n'.join(lines) + '\n').encode()
+
+
+class _MetricsHandler(BaseHTTPRequestHandler):
+    def do_GET(self):
+        if self.path == '/metrics':
+            body = _metrics_render()
+            self.send_response(200)
+            self.send_header('Content-Type', 'text/plain; version=0.0.4')
+            self.send_header('Content-Length', str(len(body)))
+            self.end_headers()
+            self.wfile.write(body)
+        elif self.path in ('/healthz', '/health'):
+            body = b'ok\n'
+            self.send_response(200)
+            self.send_header('Content-Type', 'text/plain')
+            self.send_header('Content-Length', str(len(body)))
+            self.end_headers()
+            self.wfile.write(body)
+        else:
+            self.send_response(404)
+            self.end_headers()
+
+    def log_message(self, *args):
+        pass  # keep bot logs clean
+
+
+def start_metrics_server(port: int = METRICS_PORT):
+    server = HTTPServer(('0.0.0.0', port), _MetricsHandler)  # noqa: S104 - k8s ServiceMonitor scrapes pod IP
+    thread = threading.Thread(target=server.serve_forever, name='metrics-server', daemon=True)
+    thread.start()
+    logger.info(f'Metrics server listening on :{port}/metrics')
+    return server
+

 # Redis Keys
 KEY_WHITELIST = 'bot:whitelist'
@@ -597,8 +700,9 @@ async def _collect_event_times(page) -> dict:
 async def fetch_diary_data(phpsessid: str) -> dict | None:
    logger.info('Fetching diary data via Playwright...')
    try:
+        async with asyncio.timeout(60):
            async with async_playwright() as p:
-            browser = await p.chromium.connect(PLAYWRIGHT_WS)
+                browser = await asyncio.wait_for(p.chromium.connect(PLAYWRIGHT_WS), timeout=15)
                try:
                    context_browser = await browser.new_context(user_agent=USER_AGENT)
                    await context_browser.add_cookies(
@@ -607,7 +711,7 @@ async def fetch_diary_data(phpsessid: str) -> dict | None:
                    page = await context_browser.new_page()

                    try:
-                    await page.goto(DIARY_URL, wait_until='domcontentloaded')
+                        await asyncio.wait_for(page.goto(DIARY_URL, wait_until='domcontentloaded'), timeout=30)
                        await page.wait_for_selector('table.calendar', timeout=10000)
                        await page.wait_for_timeout(1500)

@@ -646,10 +750,16 @@ async def fetch_diary_data(phpsessid: str) -> dict | None:
                        logger.error(f'Error parsing diary: {e}')
                        return None
                    finally:
-                    await page.close()
-                    await context_browser.close()
+                        with contextlib.suppress(Exception):
+                            await asyncio.wait_for(page.close(), timeout=5)
+                        with contextlib.suppress(Exception):
+                            await asyncio.wait_for(context_browser.close(), timeout=5)
                finally:
-                await browser.close()
+                    with contextlib.suppress(Exception):
+                        await asyncio.wait_for(browser.close(), timeout=5)
+    except TimeoutError:
+        logger.error('Diary fetch timed out (60s)')
+        return None
    except Exception as e:
        logger.error(f'Playwright error in diary fetch: {e}')
        return None
@@ -931,8 +1041,9 @@ def _parse_schedule_html(table_html: str) -> dict:
 async def fetch_schedule_data(phpsessid: str) -> dict | None:
    logger.info('Fetching schedule data via Playwright...')
    try:
+        async with asyncio.timeout(60):
            async with async_playwright() as p:
-            browser = await p.chromium.connect(PLAYWRIGHT_WS)
+                browser = await asyncio.wait_for(p.chromium.connect(PLAYWRIGHT_WS), timeout=15)
                try:
                    context_browser = await browser.new_context(user_agent=USER_AGENT)
                    await context_browser.add_cookies(
@@ -941,7 +1052,7 @@ async def fetch_schedule_data(phpsessid: str) -> dict | None:
                    page = await context_browser.new_page()

                    try:
-                    await page.goto(SCHEDULE_URL, wait_until='domcontentloaded')
+                        await asyncio.wait_for(page.goto(SCHEDULE_URL, wait_until='domcontentloaded'), timeout=30)
                        await page.wait_for_selector('table.schedule-table', timeout=10000)
                        await page.wait_for_timeout(1500)

@@ -969,10 +1080,16 @@ async def fetch_schedule_data(phpsessid: str) -> dict | None:
                        logger.error(f'Error parsing schedule: {e}')
                        return None
                    finally:
-                    await page.close()
-                    await context_browser.close()
+                        with contextlib.suppress(Exception):
+                            await asyncio.wait_for(page.close(), timeout=5)
+                        with contextlib.suppress(Exception):
+                            await asyncio.wait_for(context_browser.close(), timeout=5)
                finally:
-                await browser.close()
+                    with contextlib.suppress(Exception):
+                        await asyncio.wait_for(browser.close(), timeout=5)
+    except TimeoutError:
+        logger.error('Schedule fetch timed out (60s)')
+        return None
    except Exception as e:
        logger.error(f'Playwright error in schedule fetch: {e}')
        return None
@@ -1485,10 +1602,13 @@ async def check_webinars_job(context: ContextTypes.DEFAULT_TYPE):
        int: Number of webinars found, or None if check failed
    """
    logger.info('Running webinar check...')
+    _t0 = time.time()
+    _metric_check_start()

    phpsessid = redis_client.get(KEY_PHPSESSID)
    if not phpsessid:
        logger.warning('PHPSESSID missing. Skipping check.')
+        _metric_check_fail(time.time() - _t0, phpsessid_missing=True)
        # --- DEBUG LOGGING ---
        try:
            with open('phpsessid_missing.log', 'a') as f:
@@ -1502,9 +1622,10 @@ async def check_webinars_job(context: ContextTypes.DEFAULT_TYPE):
    content = ''

    try:
+        async with asyncio.timeout(90):
            async with async_playwright() as p:
                # Connect to remote Playwright service
-            browser = await p.chromium.connect(PLAYWRIGHT_WS)
+                browser = await asyncio.wait_for(p.chromium.connect(PLAYWRIGHT_WS), timeout=15)

                try:
                    # Create browser context with user agent
@@ -1520,7 +1641,7 @@ async def check_webinars_job(context: ContextTypes.DEFAULT_TYPE):

                    try:
                        # Navigate to webinar page
-                    await page.goto(WEBINAR_URL, wait_until='domcontentloaded')
+                        await asyncio.wait_for(page.goto(WEBINAR_URL, wait_until='domcontentloaded'), timeout=30)

                        # Wait for the table to load
                        await page.wait_for_selector('#meetings table', timeout=10000)
@@ -1564,16 +1685,25 @@ async def check_webinars_job(context: ContextTypes.DEFAULT_TYPE):
                            if page and not content:
                                content = await page.content()

+                        _metric_check_fail(time.time() - _t0)
                        return None
                    finally:
-                    await page.close()
-                    await context_browser.close()
+                        with contextlib.suppress(Exception):
+                            await asyncio.wait_for(page.close(), timeout=5)
+                        with contextlib.suppress(Exception):
+                            await asyncio.wait_for(context_browser.close(), timeout=5)

                finally:
-                await browser.close()
+                    with contextlib.suppress(Exception):
+                        await asyncio.wait_for(browser.close(), timeout=5)

+    except TimeoutError:
+        logger.error('Webinar check timed out after 90s (playwright hang)')
+        _metric_check_fail(time.time() - _t0)
+        return None
    except Exception as e:
        logger.error(f'Playwright error: {e}')
+        _metric_check_fail(time.time() - _t0)
        return None

    # --- DEBUG LOGGING (Saving last response content) ---
@@ -1637,6 +1767,7 @@ async def check_webinars_job(context: ContextTypes.DEFAULT_TYPE):
        else:
            logger.info(f'Found {len(current_webinars)} webinar(s), but all are already known')

+    _metric_check_ok(time.time() - _t0)
    return len(current_webinars)


@@ -1680,6 +1811,8 @@ def main():
    job_queue = app.job_queue
    job_queue.run_repeating(check_webinars_job, interval=WEBINAR_CHECK_INTERVAL, first=10)

+    start_metrics_server()
+
    logger.info('Bot started polling...')
    app.run_polling()

@@ -31,7 +31,7 @@ spec:
    spec:
      containers:
        - name: gitea
-          image: docker.gitea.com/gitea:1.27.3
+          image: gitea/gitea:1.27.3
          envFrom:
            - configMapRef:
                name: gitea-config
@@ -3,3 +3,4 @@ AUTHENTIK_DB_PASSWORD=
 GITEA_DB_PASSWORD=
 NETRONOME_DB_PASSWORD=
 PENPOT_DB_PASSWORD=
+STATUSPAGE_DB_PASSWORD=
@@ -1,23 +1,22 @@
 # Shared PostgreSQL

-This directory contains a PostgreSQL 15 deployment draft for Authentik, Gitea,
-Netronome, and Penpot. It creates one database and one login role per service;
-it does not migrate existing data or change application connection settings.
+This directory contains the shared PostgreSQL 17 deployment for Authentik,
+Gitea, Netronome, and Statuspage. It creates one database and one login role
+per service. Per-service standalone databases were removed after the
+migration (Sep 2026); Penpot stays on its own compose PostgreSQL (archived,
+not part of the shared instance).

 ## Compatibility baseline

-| Service   | Current application | Current standalone PostgreSQL | Common PostgreSQL 15                                                                 |
-| --------- | ------------------- | ----------------------------: | ------------------------------------------------------------------------------------ |
-| Authentik | 2025.10.2           |                            15 | Supported (Authentik requires 14+)                                                   |
-| Gitea     | 1.27.3              |                            14 | Supported (Gitea requires 12+)                                                       |
-| Netronome | 0.14.0              |                            17 | Validate in staging; upstream's example uses 17 but no 17-only feature is documented |
-| Penpot    | 2.17.2              |                            15 | Supported by the official deployment                                                 |
+| Service   | Current application | Shared PostgreSQL 17 |
+| --------- | ------------------- | -------------------- |
+| Authentik | 2025.10.x           | Supported (Authentik requires 14+) |
+| Gitea     | 1.27.3              | Supported (Gitea requires 12+) |
+| Netronome | 0.14.0              | Supported (upstream's example uses 17) |
+| Statuspage| custom              | Supported |

-PostgreSQL 15 is the conservative common major. A major-version downgrade or
-change must use a logical dump/restore; changing only the image tag while
-keeping a data directory is not supported. Back up and migrate one application
-at a time, starting with Netronome because its current standalone deployment
-uses PostgreSQL 17.
+A major-version change must use a logical dump/restore; changing only the
+image tag while keeping a data directory is not supported.

 For Compose, copy `.env.example` to `.env`, set all passwords, and start it with
 `docker compose -f shared-compose.yaml up -d`. This file is intentionally not
@@ -5,12 +5,12 @@ set -euo pipefail
 : "${GITEA_DB_PASSWORD:?GITEA_DB_PASSWORD is required}"
 : "${NETRONOME_DB_PASSWORD:?NETRONOME_DB_PASSWORD is required}"
 : "${PENPOT_DB_PASSWORD:?PENPOT_DB_PASSWORD is required}"
+: "${STATUSPAGE_DB_PASSWORD:?STATUSPAGE_DB_PASSWORD is required}"

 create_role_and_database() {
  local role="$1"
  local database="$2"
  local password="$3"
-
  psql --username "$POSTGRES_USER" --dbname postgres \
    -v role="$role" -v database="$database" -v password="$password" \
    <<'SQL'
@@ -25,3 +25,4 @@ create_role_and_database authentik authentik "$AUTHENTIK_DB_PASSWORD"
 create_role_and_database gitea gitea "$GITEA_DB_PASSWORD"
 create_role_and_database netronome netronome "$NETRONOME_DB_PASSWORD"
 create_role_and_database penpot penpot "$PENPOT_DB_PASSWORD"
+create_role_and_database statuspage statuspage "$STATUSPAGE_DB_PASSWORD"
@@ -60,11 +60,23 @@ spec:
              command: ["pg_isready", "-U", "postgres", "-d", "postgres"]
            initialDelaySeconds: 10
            periodSeconds: 10
+          startupProbe:
+            exec:
+              command: ["pg_isready", "-U", "postgres", "-d", "postgres"]
+            failureThreshold: 30
+            periodSeconds: 10
          livenessProbe:
            exec:
              command: ["pg_isready", "-U", "postgres", "-d", "postgres"]
            initialDelaySeconds: 30
            periodSeconds: 20
+          resources:
+            requests:
+              memory: "512Mi"
+              cpu: "500m"
+            limits:
+              memory: "2Gi"
+              cpu: "2000m"
      volumes:
        - name: postgres-data
          persistentVolumeClaim:
@@ -1,6 +1,6 @@
 services:
  postgres:
-    image: postgres:15.19-alpine
+    image: postgres:17.11-alpine
    container_name: homelab-postgres
    restart: unless-stopped
    env_file:
@@ -1,14 +1,43 @@
 {
  "$schema": "https://docs.renovatebot.com/renovate-schema.json",
  "extends": ["config:recommended"],
-  "enabledManagers": ["docker-compose", "kubernetes", "helm-values"],
+  "enabledManagers": ["dockerfile", "docker-compose", "kubernetes", "helm-values", "custom.regex"],
  "helm-values": {
    "managerFilePatterns": ["/k8s/.+values\\.ya?ml$/"]
  },
  "kubernetes": {
    "managerFilePatterns": ["/k8s/.+\\.ya?ml$/"]
  },
+  "customManagers": [
+    {
+      "customType": "regex",
+      "description": "singlesource: playwright npm version pinned in npx command (k8s + compose)",
+      "fileMatch": ["^edu_master/k8s/playwright\\.yaml$", "^edu_master/compose\\.yaml$"],
+      "matchStrings": ["playwright@(?<currentValue>\\d+\\.\\d+\\.\\d+)"],
+      "datasourceTemplate": "npm",
+      "depNameTemplate": "playwright"
+    },
+    {
+      "customType": "regex",
+      "description": "singlesource: PLAYWRIGHT_VERSION file",
+      "fileMatch": ["^edu_master/PLAYWRIGHT_VERSION$"],
+      "matchStrings": ["^(?<currentValue>\\d+\\.\\d+\\.\\d+)$"],
+      "datasourceTemplate": "pypi",
+      "depNameTemplate": "playwright"
+    }
+  ],
  "packageRules": [
+    {
+      "description": "singlesource playwright - use whichever version is found, keep docker+pypi+npm in sync",
+      "matchPackageNames": ["playwright", "mcr.microsoft.com/playwright"],
+      "groupName": "playwright singlesource",
+      "groupSlug": "playwright"
+    },
+    {
+      "description": "playwright must not automerge - version skew breaks WS handshake (checker.py:1523 vs playwright.yaml:20)",
+      "matchPackageNames": ["playwright", "mcr.microsoft.com/playwright"],
+      "automerge": false
+    },
    {
      "description": "Keep private homelab images unchanged",
      "matchDatasources": ["docker"],