fix: metrics dev-env for linux

Fixes a hardcoded path in the compose setup for local metrics. Updates
the docs, as well, to clarify that the first-run experience via docker
compose is only relevant for Linux hosts, due to use of host-networking.
Will work on a macos-compatible dev env later, but for now focusing on
making sure what's in the repo is clear and runnable for actual testnet
node operators, which means Linux boxes.

Closes #4565.

deployments/compose/metrics.yml

@@ -13,10 +13,17 @@ services:
       dockerfile: deployments/containerfiles/Dockerfile-grafana
     network_mode: host
     user: "${UID:-1000}"
+    # Don't use ports, since ports conflicts with network_mode=host.
+    # We use network_mode=host so that metrics can talk to services bound
+    # to localhost.
+    # ports:
+    #   - "3000:3000"
 
   # The Prometheus service, for scraping metrics from Penumbra's pd metrics port.
   prom:
     image: "docker.io/prom/prometheus"
     network_mode: host
+    # ports:
+    #   - "9090:9090"
     volumes:
       - "${PWD:?}/../config/prometheus.yml:/etc/prometheus/prometheus.yml:ro"

deployments/config/prometheus.yml

@@ -1,5 +1,5 @@
 scrape_configs:
-  - job_name: 'Tendermint Node'
+  - job_name: 'CometBFT Node'
     scrape_interval: 10s
     scheme: http
     metrics_path: metrics

deployments/containerfiles/Dockerfile-grafana

@@ -1,5 +1,5 @@
 ARG GRAFANA_VERSION="10.1.5"
 FROM docker.io/grafana/grafana:${GRAFANA_VERSION}
-COPY deployments/config/grafana/provisioning /etc/grafana/provisioning
+COPY deployments/config/grafana/provisioning/ /etc/grafana/provisioning/
 COPY deployments/config/grafana/grafana.ini /etc/grafana/grafana.ini
-COPY deployments/config/grafana/dashboards /var/lib/grafana/dashboards
+COPY deployments/config/grafana/dashboards/ /var/lib/grafana/dashboards/

docs/guide/src/dev/metrics.md

@@ -62,20 +62,22 @@ After being changed, Grafana dashboards should be backed up to the repository fo
 Grafana has an [import/export](https://grafana.com/docs/grafana/latest/dashboards/export-import/) feature that
 we use for maintaining our dashboards.
 
-1. Export the dashboard as JSON with the default settings
-2. Rename the JSON file and copy into the repo (`config/grafana/dashboards/`)
-3. PR the changes into main, and confirm on preview post-deploy that it works as expected.
-
+1. View the dashboard you want to export, and click the share icon in the top bar.
+2. Choose **Export**, and enable **Export for sharing externally**, which will generalized the datasource.
+3. Download the JSON file, renaming it as necessary, and copy into the repo (`config/grafana/dashboards/`)
+4. PR the changes into main, and confirm on preview post-deploy that it works as expected.
 
 ## Editing metrics locally
 
 To facilitate working with metrics locally, first run a `pd` node on your machine with the metrics endpoint
 exposed. Then, you can spin up a metrics sidecar deployment:
 
 ```bash
-cd deployments/compose
 just metrics
 ```
 
+Note that this setup only works on Linux hosts, due to the use of host networking, so the metrics
+containers can reach network ports on the host machine.
+
 To add new Grafana visualizations, open http://localhost:3000 and edit the existing dashboards.
 When you're happy with what you've got, follow the "Backing up Grafana" instructions above to save your work.

justfile

@@ -10,6 +10,11 @@ fmt:
 proto:
     ./deployments/scripts/protobuf-codegen
 
+# Run a local prometheus/grafana setup, in containers, to scrape a local node. Linux only.
+metrics:
+    cd ./deployments/compose/ \
+        && docker-compose -f metrics.yml up --build --abort-on-container-exit --force-recreate --remove-orphans
+
 # Configures and runs a relayer instance between "preview" (latest main) and local devnet on current HEAD
 relayer-local-devnet:
     ./deployments/scripts/relayer-local-devnet