Hysen Labsnode
How to run an Ink Node
The Ink node runbook ships with an expiration date
A Docker Compose setup for running an Ink node on the current op-geth stack, complete with healthcheck and monitoring services, and a README that tells you plainly to plan your exit.
What the Compose stack gives you
The repository explains how to run an Ink node, and the current answer is a Docker Compose setup on the op-geth based stack, with healthcheck and monitoring services alongside. It is Shell scripting under the MIT license, around 36,500 stars, with the documentation living at docs.inkonchain.com.
May 31, 2026
That date matters. Per Optimism, op-geth support ends on May 31, 2026, and nodes still running it at the L1 Glamsterdam hardfork will not be able to follow the canonical chain. The README is explicit that these instructions are the current op-geth runbook for the Compose stack, not a long term recommendation, which is a refreshing level of candor for infrastructure docs.
The migration it prescribes
Operators of production or long lived nodes are told to start planning an op-reth migration: run the new client in parallel, validate it over a meaningful window, and prepare a fresh snapshot before the hardfork window arrives. The framing is firm on one point, that the migration is operator owned and not something to delay until a sequencer side client switch happens.
What is still missing
The repository does not yet ship an op-reth Compose path, and the README lists what it still needs before one can exist. Until then this is a transitional runbook: fully useful today, honest about being temporary, and with the next step written down rather than implied.
Editorial conclusion
Infrastructure repos rarely volunteer their own obsolescence. This one does, and node operators on any chain benefit from the habit.
Community notes