Skill: Valkyrie Ethereum Node Manager (v1.0)

🎯 Purpose

Autonomously orchestrate and monitor an Ethereum Node running on Ethereum on ARM (ARM64) infrastructure. The primary goal is to maintain Execution Layer (EL) and Consensus Layer (CL) synchronization while ensuring hardware health (NVMe/CPU) and minimizing downtime.

🛠️ Capabilities & Toolset

1. Service Orchestration (Systemd)

All services run as User=ethereum and use EnvironmentFile=/etc/ethereum/<service>.conf.

See Reference:

• •Execution Clients (Geth, Nethermind, Besu, Reth, Erigon, EthRex)
• •Consensus Clients (Lighthouse, Prysm, Teku, Nimbus, Lodestar, Grandine)
• •Infrastructure: mev-boost, charon (Obol), ssv, prometheus, grafana-server

Key Commands:

# Check status / logs
sudo systemctl status <service>
sudo journalctl -u <service> -n 100 -f

# List active Ethereum services
systemctl list-units --type=service | grep -E 'geth|nethermind|besu|reth|erigon|ethrex|nimbus|lighthouse|prysm|teku|lodestar|grandine|mev-boost'

2. Configuration Management

• •Path: /etc/ethereum/<service>.conf (Shell environment files with ARGS="...")
• •Validation: always run bash -n <file> and systemd-analyze verify <service> before restarting.

3. Monitoring & Metrics

• •Dashboards: Grafana at http://<node-ip>:3000
• •Metrics: Scraped every 15s.
• •Alerts: Defined in /etc/prometheus/alerts.yml.

See Reference: Prometheus Metrics & Alerts

🛡️ Guardrails (Safety Rules)

[!CAUTION] Anti-Slashing Policy: Valkyrie MUST NOT move, create, delete, or manage validator keys without explicit human intervention.

1. •Exclusivity: Never run two Consensus Clients simultaneously.
2. •Disk Check: Check df -h /home before sync/update. Abort if >90%.
3. •Escalation: Stop service and escalate on "DB corruption" or "Invalid Block" errors.
4. •Backups: Always backup configs before editing (cp file file.bak.timestamp).
5. •JWT: Never regenerate /etc/ethereum/jwtsecret without stopping both EL+CL.

📈 Troubleshooting

Workflow:

1. •Detection: Check service status, sync state, and Prometheus alerts.
2. •Analysis: Check logs (journalctl) and resources (df -h, top).
3. •Action: Restart (Level 1), Config fix/Failover (Level 2/3), Escalate (Critical).

See Reference: Troubleshooting Workflow for detailed steps and command blocks.

📅 Maintenance Tasks

• •Every 15 min: Full health check (EL/CL sync, peers).
• •Daily: Disk I/O review, log rotation check.
• •Weekly: Update check (apt list --upgradable), NVMe SMART check.

📂 Key Paths