Watching blockchain nodes from the outside in
Custom exporters, Prometheus and Grafana over a node management platform, covering endpoint health and resource use on every node.
01 — The challenge
Where it started.
The platform’s whole service is the condition of its customers’ blockchain nodes: endpoint health, status visible in real time, and CPU, memory and disk read early enough to act on.
Off-the-shelf monitoring does not know what a healthy node looks like, so the signals that matter were the ones not being collected.
02 — What we built
The work itself.
Decide the metrics first. The key metrics identified and a collection and visualisation strategy agreed before anything was built.
Custom exporters. Exporters written for the signals that general-purpose monitoring does not expose, feeding Prometheus.
Across every node. The exporters deployed fleet-wide with the Prometheus and Grafana infrastructure behind them.
Tuned on what it showed. Performance watched after rollout and the solution revised against observed behaviour rather than left at version one.
03 — Results
What the client has now.
Node status and performance visible in real time, to the platform’s team and to its customers. Downtime reduced by alerting on conditions early rather than on failure. Resource use measured well enough to manage capacity against it. Support answered with evidence, which is what the customers were paying for.
Looking for a similar solution? Contact us today!