Can / Should One Run Core Lightning and LND on the Same Node? \ stacker news ~lightning

Can / Should One Run Core Lightning and LND on the Same Node?

4920 sats \ 20 comments \ @kepford 13 Feb lightning

I could have sworn that @DarthCoin recommended this for Umbrel users but either I imagined it or I just can't search. Regardless.

Is this possible?
Is it a bad idea?

My gut tells me it is possible but is a bad idea. Feel free to tell me how dumb I am.

view all related items

7 replies \ @DarthCoin 13 Feb

Yes, I was running both on an Umbrel software, some time ago. But just for testing. I will NOT recommend doing it with Umbrel or any other software like that.

In theory works fine, but if you are not enough skilled with CLI and all that stuff, you will have serious problems doing it and none of your nodes will work properly.

I can enter in more details, but is quite long.

Why would you want to run such a scenario? Is better to try to compartmentalize things...

Fun fact: did you know that you can run a BTC Core + CLN, activate neutrino on that Core node and then from other machine, just install LND and point it to your local Core node neutrino to sync? So in this way you do not need a 2nd core node with LND...

Is it a bad idea?

I can say yes... because you can kill easily your disk drive. LN nodes require a lot of I/O and with just one LN node is enough to stress it at maximum... with 2 nodes is killing totally.

0 sats \ 3 replies \ @Kathleen_Croteau 14 Feb

Would it also prematurely wear out an NVM SSD?

10 sats \ 1 reply \ @DarthCoin 14 Feb

Is not about speed

0 sats \ 0 replies \ @Kathleen_Croteau 14 Feb

Just the constant wear and tear?

0 sats \ 0 replies \ @kepford OP 14 Feb

Why would you use a NVM SSD. I've used SSDs for years with no issue though. I wouldn't waste the sats on an NVMe drive. SSD should be fine. As darth said. Not about speed. No need for the speed.

0 sats \ 2 replies \ @kepford OP 13 Feb

Don't waste your time @DarthCoin. Thanks for confirming.

119 sats \ 1 reply \ @DarthCoin 13 Feb

Here I wrote some guides in regards of running multiple nodes. You can run also public / private combination...

0 sats \ 0 replies \ @kepford OP 13 Feb

Thanks. I think it was one of these guides that was in the back of my mind.

729 sats \ 8 replies \ @RocketNode 13 Feb

Possible? Yes Bad Idea? Depends.. but probably yes. Lightning (as a routing node) takes capital and splitting it into two nodes is probably not the best use of funds. Also, most tools don't overlap so you will have to learn different things for each implementation. But if you want a 2nd node for privacy or something, then sure, go for it.

11 sats \ 0 replies \ @DarthCoin 13 Feb

Exactly, all depends of what do you want to achieve, to do... are many solutions and scenarios.

0 sats \ 6 replies \ @kepford OP 13 Feb

Yeah, putting it out of my mind.

I was thinking more along the lines of running a second node with no announced (public) channels but I could do that on a separate VM/box. No need to complicate and possibly cause issues with my routing node.

10 sats \ 5 replies \ @RocketNode 13 Feb

I was thinking more along the lines of running a second node with no announced (public) channels but I could do that on a separate VM/box. No need to complicate and possibly cause issues with my routing node.

You can run Blixt on mobile with a private (even better as zero-conf) channel to your node.

0 sats \ 4 replies \ @kepford OP 13 Feb

I like Blixt but I don't like turning a phone into a critical piece of infra. To my knowledge you can't move Blixt nodes to a new device. I prefer to treat phones as disposable. Its a nice option if you don't have computer hardware but I think if you do it would be better to run a second node on a computer.

0 sats \ 1 reply \ @RocketNode 13 Feb

Zero conf channels.. Essentially disposable channels. Actual sats are on your node, so the mobile can spend/receive without having to worry about the phone. If the phone is lost/compromised, you just kill the channel from your node. I know @DarthCoin has Blixt guides, but not sure if it goes into private/trusted zero conf channels.

10 sats \ 0 replies \ @DarthCoin 13 Feb

Yes I wrote a guide about that scenario here and indeed I didn't talk too much about 0-conf channels, just briefly mention them. https://darthcoin.substack.com/p/dunder-lsp-and-lightning-box-provider and here https://darthcoin.substack.com/p/bitcoin-private-banks-over-lightning

But I don't know what OP want to do, so not sure if these scenarios are good for him.

0 sats \ 1 reply \ @DarthCoin 13 Feb

You must make the difference between public and private nodes ! Blixt, Zeus, Breez, Electrum, Phoenix, Mutiny are built-in PRIVATE nodes. In a small shop like a grocery maybe you can use them to take payments but NOT for other stuff, like routing.

Again, depends on what do you want to achieve use one or another.

0 sats \ 0 replies \ @kepford OP 13 Feb

Yeah, I know what you're saying.

0 sats \ 1 reply \ @kristapsk 14 Feb

When I wanted to test LND in addition to CLN, I just started running Bitcoin node on another device at home.

21 sats \ 0 replies \ @kepford OP 14 Feb

Yeah. The more I think about it I'd rather have a separate machine/VM vs adding complexity to my routing node. My reason for posting this question was that I had thought Darth had written about doing this and he did. Having a private node running on the same machine with unannounced channels. But that can be done on another system, phone, etc.

0 sats \ 0 replies \ @OT 13 Feb

I was wondering this also. You see core on umbrel and start9. Umbrel is set up as an LND node so not sure how this works