Developer
What does running a node in production actually involve?
Running the process is easy. Running it in two years is the commitment.
The answer
Beyond keeping the process alive: upgrade choreography on the network’s schedule rather than yours, monitoring that distinguishes a lagging peer from a fork, storage growth that will outpace your original estimate, and a rota of people who can act at any hour. The decision to self-host is a decision to staff those four indefinitely — which is why the honest comparison is against a managed arrangement rather than against the cost of a server.
Monitoring that says something useful
Block height alone is a poor signal. What you need to distinguish: falling behind the network, peered with a minority fork, and healthy but serving stale reads to your application. Those three look similar on a naive dashboard and require completely different responses.
The costs the first estimate misses
Storage that grows faster than modelled, and the migration when the disk fills. Upgrade windows that fall on the network’s calendar, including inconvenient ones. And the specialist knowledge concentrating in one engineer — a single point of failure with a badge, and the reason the build-or-buy assessment should ask who is on call in eighteen months.
Read next
Talk to the practice
Tell us what you are trying to build and what has to be true for it to work.
Contact