feat(docker): add Prometheus + Grafana monitoring compose overlay

[amir-deris]

Summary

• Adds docker/docker-compose.monitoring.yaml — a compose overlay that attaches Prometheus and Grafana to the existing localnet network alongside the 4-node local cluster
• Adds docker/docker_compose_monitoring/prometheus.yaml — scrape config targeting all four nodes (sei-node-{0-3}:1317) at 5 s intervals
• Adds docker/docker_compose_monitoring/grafana-datasource.yaml — auto-provisions Prometheus as the default Grafana datasource on startup
• Adds make docker-cluster-start-monitoring / make docker-cluster-stop-monitoring Makefile targets and updates docker/README.md to document the combined workflow alongside the existing script-based alternative

---

Note

Low Risk
Low risk: adds optional local Docker Compose monitoring services and make targets without changing chain/runtime logic or production configuration.

Overview
Adds a docker-compose.monitoring.yml overlay that runs Prometheus and Grafana on the existing localnet network, including a Prometheus scrape config for all four local nodes and a provisioned Grafana Prometheus datasource.

Introduces make docker-cluster-start-monitoring / make docker-cluster-stop-monitoring to bring the cluster + monitoring stack up/down in one command, and updates docker/README.md to document the combined workflow alongside the existing script-based approach.

^{Reviewed by Cursor Bugbot for commit a314fb0. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

The latest Buf updates on your PR. Results from workflow Buf / buf (pull_request).

Build	Format	Lint	Breaking	Updated (UTC)
✅ passed	✅ passed	✅ passed	✅ passed	May 15, 2026, 8:02 PM

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 59.31%. Comparing base (823a78d) to head (c924ea1).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3436      +/-   ##
==========================================
+ Coverage   59.29%   59.31%   +0.02%     
==========================================
  Files        2125     2118       -7     
  Lines      175629   174715     -914     
==========================================
- Hits       104144   103639     -505     
+ Misses      62404    62039     -365     
+ Partials     9081     9037      -44     

Flag	Coverage Δ
sei-db	70.41% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.
see 138 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[cursor]

PR Summary

Low Risk
Low risk: changes are limited to local Docker compose/Makefile workflows and documentation, with no impact on chain runtime or production deployments unless developers opt into the new targets.

Overview
Adds a docker-compose.monitoring.yml overlay that brings up Prometheus and Grafana on the existing localnet network, with Prometheus scraping sei-node-{0..3}:26660 and Grafana auto-provisioned to use it as the default datasource.

Introduces make docker-cluster-start-monitoring / make docker-cluster-stop-monitoring to start/tear down the cluster plus monitoring stack in one command, and updates docker/README.md to document the combined flow alongside the existing script-based approach.

^{Reviewed by Cursor Bugbot for commit c924ea1. Bugbot is set up for automated code reviews on this repo. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

There are 2 total unresolved issues (including 1 from previous review).

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 1f70ad0. Configure here.}

[cursor]

Monitoring target duplicates entire cluster-start recipe body

Low Severity

The docker-cluster-start-monitoring target copy-pastes the entire body of docker-cluster-start — cleanup steps, directory creation, detach-flag logic, and a long inline list of ~10 environment variables (INVARIANT_CHECK_INTERVAL, GIGA_EXECUTOR, AUTOBAHN, etc.). The only difference is the compose file arguments and --no-attach flags. When a new feature-flag env var is added to docker-cluster-start, it can easily be forgotten in the monitoring target, causing silent behavioral differences between the two cluster modes.

Additional Locations (1)

• Makefile#L289-L300

 

^{Reviewed by Cursor Bugbot for commit 1f70ad0. Configure here.}