Monitoring and Alerting
Nodes are monitored throughout their lifecycle with our comprehensive tooling. Core machine-level metrics — CPU load, available disk space, memory — are monitored in real-time, with alerting thresholds set so the platform can automatically notify our tech team about any potential errors before they may occur.
However, nodes themselves present additional key measures that must be watched — peer counts, block height, transaction pools, etc. Our experience allows us to monitor, repair, and replace nodes where protocol-level errors may occur.