Monitoring stack: Prometheus + Grafana dashboards

monitoring/README.md

@@ -0,0 +1,64 @@
+# Monitoring stack
+
+Ephemeral Prometheus + Grafana for local stress and soak runs
+(paritytech/web3-storage#267). Brings up dashboards for chain health,
+economics, and protocol activity against a running chain and the domain-state
+exporter.
+
+## Scrape targets
+
+| Target | Source | Port |
+| --- | --- | --- |
+| `web3-storage-chain` | `examples/papi/chain-exporter.ts` (domain state) | 9700 |
+| `parachain-node` | collator's built-in Substrate metrics | 9615 |
+| `relay-node` | relay validator's Substrate metrics (optional) | 9616 |
+
+The collator and relay Prometheus ports are pinned in `zombienet.toml`
+(zombienet otherwise assigns a random port each run).
+
+## Run it
+
+```bash
+# 1. chain + provider (from repo root)
+just start-chain
+just start-provider inmemory 3333 ./provider-data "" true
+
+# 2. domain-state exporter
+pnpm --filter web3-storage-papi-demo run chain-exporter
+
+# 3. monitoring stack
+cd monitoring && docker compose up -d
+```
+
+- Grafana: http://localhost:3000 (anonymous viewer; admin/admin to edit)
+- Prometheus: http://localhost:9090
+
+Dashboards are provisioned automatically under the "Web3 Storage" folder.
+
+Generate some load to watch, e.g. the challenge sweep:
+
+```bash
+pnpm --filter web3-storage-papi-demo run lower-timeouts -- --set ChallengeTimeout=30
+pnpm --filter web3-storage-papi-demo run challenge-saturation -- --count 800
+```
+
+## Dashboards
+
+- **Chain Health** — block height (best vs finalized), finality lag, block
+  production rate, txpool, peers, and per-class block weight (the mandatory
+  class spikes at a challenge-deadline block, PoV budget 5 MiB).
+- **Economics** — provider stake, free/reserved balance, cumulative slashes,
+  checkpoint rewards and pool.
+- **Protocol Activity** — agreements active vs expired (the lazy-cleanup
+  residual), pending challenges, challenge lifecycle (created/defended/slashed)
+  and rate, checkpoints submitted/missed, provider/bucket counts.
+
+The dashboard JSON is generated by `grafana/gen-dashboards.mjs`; edit that and
+re-run `node grafana/gen-dashboards.mjs grafana/dashboards` rather than hand-
+editing the JSON.
+
+## Pointing at a different target (e.g. PreviewNet)
+
+The containers reach the host via `host.docker.internal`. To scrape a remote
+chain or exporter instead, edit the target addresses in
+`prometheus/prometheus.yml`.

monitoring/docker-compose.yml

@@ -0,0 +1,47 @@
+# Ephemeral monitoring stack for local stress/soak runs (paritytech/web3-storage#267).
+#
+# Prometheus + Grafana that scrape three targets on the host:
+#   - the parachain collator's Substrate Prometheus endpoint (pinned to :9615
+#     in zombienet.toml)
+#   - the relay validator's endpoint (:9616, optional)
+#   - the domain-state exporter (examples/papi/chain-exporter.ts, :9700)
+#
+# The nodes and the exporter run on the host (zombienet native), so the
+# containers reach them via host.docker.internal; the extra_hosts mapping makes
+# that resolve on Linux too. To point at a different target (e.g. PreviewNet),
+# edit prometheus/prometheus.yml.
+services:
+  prometheus:
+    image: prom/prometheus:v3.1.0
+    container_name: w3s-prometheus
+    command:
+      - --config.file=/etc/prometheus/prometheus.yml
+      - --storage.tsdb.retention.time=30d
+    ports:
+      - "9090:9090"
+    volumes:
+      - ./prometheus:/etc/prometheus:ro
+      - prometheus-data:/prometheus
+    extra_hosts:
+      - "host.docker.internal:host-gateway"
+
+  grafana:
+    image: grafana/grafana:11.4.0
+    container_name: w3s-grafana
+    depends_on:
+      - prometheus
+    ports:
+      - "3000:3000"
+    environment:
+      - GF_SECURITY_ADMIN_USER=admin
+      - GF_SECURITY_ADMIN_PASSWORD=admin
+      - GF_AUTH_ANONYMOUS_ENABLED=true
+      - GF_AUTH_ANONYMOUS_ORG_ROLE=Viewer
+    volumes:
+      - ./grafana/provisioning:/etc/grafana/provisioning:ro
+      - ./grafana/dashboards:/var/lib/grafana/dashboards:ro
+      - grafana-data:/var/lib/grafana
+
+volumes:
+  prometheus-data:
+  grafana-data: