Validator Nodes
As validators on the Cardano, Solana, and Flow blockchains, we actively contribute to decentralisation of the ecosystem.
Operating Blockchain Validators: A Behind-the-Scenes Look
We are proud to serve as active validators on the Cardano, Solana, and Flow blockchains. By participating in consensus, we not only support decentralization but also stay closely involved in the critical events happening on these networks. Running a validator node, however, comes with its own set of challenges. In this article, we offer a sneak peek into how we manage our validator operations.
Cardano
Our largest operation as validators is on the Cardano blockchain, where we manage 20 Block Producer (BP) nodes and 8 Relay nodes to maintain a robust topology. For simplicity and efficiency, we deploy these nodes using a Docker image of cardano-node.
Our infrastructure is provisioned through Google Cloud Platform (GCP) using Terraform. Once the environment is up, we ensure the proper configuration of nodes via custom Ansible playbooks. For metric collection, we rely on Prometheus endpoints exposed by cardano-node by default.
To track our block production, we utilize the cncli tool, which generates a block production schedule at the beginning of each epoch. We then cross-check this data with Blockfrost to verify if our blocks were successfully produced.
Solana
Operating a validator on the Solana blockchain is perhaps the most challenging of our validator operations. The pace of activity is rapid, and even minor mistakes can lead to downtime or delinquency. To ensure high availability, we maintain two nodes (primary and secondary), though only one actively votes at a time
Our Solana nodes run on custom hardware and are configured using Ansible playbooks. For security, metrics are collected from the validator’s RPC endpoint, which is firewall-protected and accessible only to localhost.
Flow
Our Flow verification node operates in GCP Cloud and is also managed with Ansible playbooks. The Flow team simplifies the process by distributing the software as Docker images, making setup relatively straightforward. With a few configuration tweaks, logs collected by Promtail, and metrics gathered by Prometheus, we ensure smooth operation of our node.
Monitoring
A key aspect of running any validator is monitoring. Without it, we wouldn’t be able to visualize metrics, receive timely alerts, or react promptly to incidents. All of our monitoring data is integrated into Grafana Cloud, which enables us to stay on top of any issues as they arise.