I’m trying to figure out which method I want to use to start mining for BIP110 using Braiins Hashpower.
I currently have an RPi running Umbrel but I really don’t want to open ports to the outside world and I’m not seeing the benefit of using tunnels as opposed to just starting a new node (Start9) on a VPS and just pointing the hashpower to that VPS.
So I guess my question is, why are people bothering with tunnels instead of just running the node in a VPS and bypassing all the additional complexity and apps required to use a tunnel or VPS intermediary?
Technical expertise, mainly. Setting up and configuring bitcoind and Datum Gateway natively, as well as maintaining it and troubleshooting any problems that crop up over time isn’t for everyone.
I suppose if they give you the option to install a custom OS on the VPS, then that is an option. I’m not sure how common that is though. In my experience you typically must choose from a limited list of supported OS’s.
As far as using one of the standard OS’s that are commonly installed on a VPS, StartOS is by comparison a considerably easier UX for most people. In your case if you are comfortable around a Linux terminal, then just using the VPS to run Knots and Datum Gateway may be a better option.
Umbrel has a script for installing on top of Ubuntu and Mechanic’s website ( Fresh VPS node setup for DATUM ) also has a script for installing node and datum on a VPS. So it seems as though this is the most straightforward approach. My only question now is how much the VPS will cost monthly.
For just a node, you can get by with 1TB storage (but 2TB if you want to use it for other things too). And depending on the network limits, you may eat through some initial extra cost or time due to throttling during the IBD. And you’ll probably want to look at settings to reduce or limit P2P traffic long term to fit within your plan.
I’ve not set up a VPS for these particular set of constraints, so you’ll have to do some research and monitoring your stats to fit your usage to the allotted specs. Most of the popular LLMs can probably help you estimate the required specs, suggest the proper config options, and research providers that would fit the requirements.