initial(main): Add files for basic gridscale deployment of 2 hosts with config

templates/Caddyfile.j2

@@ -0,0 +1,13 @@
+{
+  metrics /metrics {
+    per_host
+    observe_catchall_hosts
+  }
+  admin :2019 {
+    origins http://{{ private_ip_caddy }}:2019
+  }
+}
+:80 {
+  root * /usr/share/caddy
+  file_server
+}

templates/alert_rules.yml.j2

@@ -0,0 +1,20 @@
+{% raw %}
+groups:
+  - name: demo
+    rules:
+      - alert: Caddy Down
+        expr: up{job="caddy"} == 0
+        for: 1m
+        labels:
+          severity: critical
+      
+      # 5% 4xx rate is a rough default. Client-error rates vary widely by application (bots, API misuse, rate limits, short-lived token expiry, malformed clients) — adjust based on your baseline.
+      - alert: CaddyHighHTTP4xxErrorRateService
+        expr: sum(rate(caddy_http_request_duration_seconds_count{code=~"4.."}[3m])) by (instance) / sum(rate(caddy_http_request_duration_seconds_count[3m])) by (instance) * 100 > 5 and sum(rate(caddy_http_request_duration_seconds_count[3m])) by (instance) > 0
+        for: 1m
+        labels:
+          severity: critical
+        annotations:
+          summary: Caddy high HTTP 4xx error rate service (instance {{ $labels.instance }})
+          description: "Caddy service 4xx error rate is above 5%\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
+{% endraw %}

templates/prometheus.yml.j2

@@ -0,0 +1,10 @@
+global:
+  scrape_interval: 15s # default is 1 minute
+
+rule_files:
+  - /etc/prometheus/alert_rules.yml
+
+scrape_configs:
+  - job_name: caddy
+    static_configs:
+      - targets: ['{{ private_ip_caddy }}:2019']