I recently upgraded our HTPC because it couldn’t be upgraded to win11. I now have the 6 yr old, i5 NUC to spare and I want to have it as a standby node in case of a hardware failure of my node. I dismantled it, cleaned the fan and heat sink and upgraded the HDD to a 2TB SSD so it should be more than adequate to run S9. Can I do a basic install of S9 with the bare minimum services so I can have a fairly up to date blockchain downloaded and be as ready as possible to install my backup if and when I do have a hardware failure? I realise I can’t have LND and related services running on a duplicate node but are there also issues I need to be aware of regarding the IP of the node, passwods or anything else?. I make regular backups of all services on a network folder in a win11 PC and a portable HHD but I have never restored anything to know if it all works. If I keep the spare blank and install everything at the time it is needed then my LND node will be out of action for the time it takes to download the blockchain and restore channels. What is the best way for me to proceed? I haven’t found anything in the documentation or this forum that covers this scenario. Any advice would be appreciated.
The built in backup system does most of this. Having another fully synced bitcoin node might come in handy to restore blockchain data from, so-as you don’t have to wait for the Initial Block Download, and validation. You don’t need this standby node to be on full time. You can just turn it on once every month or two, let it sync, and turn it back off again. In the case of drive failure on your primary server, you would flash your server using the “restore from backup” recovery option, after installing a new SSD. Then use the blockchain migration method to migrate the blockchain data from the standby server.
After that, you can start the services. You’ll find that as a precaution, all of your lightning channels have been closed. The sats on your side of the channels will eventually be sent back to your on-chain wallet, but may not appear immediately. You’ll then need to fund your node, and open new channels. After that, everything should “just work” as all the addresses will be the same as the old server.
If the motherboard on your primary server died, but the drive was still good, the above method is not applicable. In most cases, you can simply physically transfer the SSD to a new server, set your boot options appropriately, and you back up and running. If for whatever reason that doesn’t work, you can reflash your server, using the “Use existing drive” recovery option, to restore your server back to the state is was in before the failure, channels intact.