For people who have/are running nodes, are there any strategies that have been most effective for you for earning routing fees?
From my very limited routing experience so far it seems there are a few main questions:
- Who should I open a channel to? And how much capacity should it be?
- How much should I set the fees of that channel given the tradeoff of routing traffic and profitability?
- How do I manage the ongoing operations of my node - ie circular rebalancing - to keep liquidity on both ends and routes flowing?
Just thinking about things generally, it seems liquidity would flow in some sort of network loop:
- Fiat -> Exchanges -> Wallet and Point of Sale Providers -> Businesses/Apps that accept LN -> Wallet and Point of Sale Providers -> Exchanges -> Fiat
Am I thinking about this correctly?
Connecting with the "big nodes" I can see that I route payments when my fees are low enough (but this seems like a race to the bottom) - for example why would Kraken route many hops through my node when it can go direct to WalletOfSatoshi (other than lower fees)
And connecting with "small nodes" it seems we are at best an "outer hop alternative route" of sorts for any of Exchanges, Wallets, Businesses?
Has anyone who has had routing success be willing to share their experiences and strategies?
Or are there other challenges/advice I am not yet thinking of?
I run the node c-otto.de and left a few notes on https://c-otto.de. Aside from that, I think that having lots of capital is crucial. Your node needs to be well connected, but it also needs to have enough inbound and outbound liquidity for a lucrative large forward to happen (I love those 10M+ forwards, even though they are rather rare). Most implementations take the channel capacity into account when finding routes, which also means that you should try to have large channels.
I think that LnRouter (https://lnrouter.app/) is really helpful. Before you open a (large) channel to a node, make sure that it can actually put those sats to good use. If the peer lacks outbound liquidity, it doesn't help to open yet another channel TO them.
Some helpful stuff from Alex Bosworth:
How I set my fees (and why):
https://yalls.org/articles/97d67df1-d721-417d-a6c0-11d793739be9:0965AC5E-56CD-4870-9041-E69616660E6F/a81e96f9-b1a7-4763-9f7a-adf226ebde59
Regarding rebalances:
Don't rebalance if it's not worth it. If you use rebalance-lnd, only routes are used that (if routing happens) put you in a better position.
Another advice: weed out bad peers, keep the good peers, use available sats to improve your position and/or try out new peers. With lnd-manageJ (https://github.com/C-Otto/lnd-manageJ) you can get a rating for each peer based on routing activity, and use that number to guide your decisions (which is what I do).
Thank you so much for sharing all of this, your site and all of these links are incredibly helpful. I'm excited to dive into all of these resources and your products in more detail.
I forward on average a little over 1 Btc / day profitably (different node than this SN ID haha) . Mix of manual and automated
I think you're spot on with that liquidity flow you described, and many have posted some great router docs. Ill share a bit of my thoughts
As a pleb, I think many would be better off identifying and focusing on a SPECIFIC flow rather than trying to act as a generalized router. Plebs with 1 BTC are never going to compete against the big 40BTC+ router hubs in terms of real routes - its a liquidity game, and they have more than you
What I mean very specifically by this is, focus on facilitating a specific flow. Breez -> WOS is one example that I know is quite profitable ;) . I think many would do much better for themselves and expend less effort to focus on providing a highly reliable route for such flows, though there can be challenge in identifying such
cheers
Here's a paper I always recommend
This is great, thanks for sharing 🙏
See also
Another one
Check out my post history for synopses of Lightning Papers. I think I did one on the first one I linked you.
Thanks, these are great
The user @darthcoin has way to many guides from wich you can learn a lot IMO, he is active here to most of the time, though I dont know if I've @ him correctly.
I hope he's okay and is vacationing. He hasn't been active for almost a month.
Sorry man, I am building my own citadel with my own hands. Seriously there's a lot of work to be done so during summer I am not online so much time (my citadel is deep into the mountains and will take some time until I would have signal there).
That's okay! Just wanted to make sure nothing weird happened.
When will be ready and the satellite internet connection installed, maybe I will put a webcam to broadcast 24/7 from my citadel and start a video Q&A help channel for noobs, to ask me anything they want about Bitcoin and LN...
If I'm not so busy with the vegetables garden.
This comment was featured on This Day in Stacker News.
Obligatory not a node routing expert, but this guy is a good follow: https://mobile.twitter.com/cold_sats
He's been documenting his journey, pitfalls and all.
Definitely! I really enjoy all of Steven's posts, tools, and transparency
This comment was featured on This Day in Stacker News.
I personally found this blog insightful:
https://darthcoin.substack.com/p/lightning-routing-fees-experiment
Thanks for sharing this 🙏
https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com%2Fpublic%2Fimages%2F6839aeff-4088-4daf-9b01-5366f5b4e2db_677x921.png
This meme image summary and the fact that you can play with minimum and maximum HTLC to promote or restrict flow was an eye opener for me.
Also I really think a 0 Base fee practice is healthy for the LN ecosystem. Play with ppm HTLC and when On-Chain fees are low open or close channels to nodes to test out which give better results.
This exact issue is why I am excited for FediMints. I am a pretty technical person, but I just can't be bothered to run a node at this point. I use Phoenix now, and will use FediMints once it comes out. I'd be more serious about running one as part of a federation, but as it sits there just is not much an incentive to do it for me.
Thanks, I keep hearing about FediMints but have not taken the time to do any research yet - I'll dig into this more
Here is their website: https://fedimint.org/
Also check out this recent podcast: https://optoutpod.com/episodes/s3e01-fedimints/
I want to find out if anyone successfully balanced channels with *only changing fees"
it should be possible specially for a high traffic node with many channels.
And I'd be curious in the details - I suppose that many of the sophisticated nodes have constant rebalancing algorithms for optimizing this traffic flow/maximizing fees
I would be very interested in this too
I am in the process of launching a new service based on lightning micropayments. (My second initiative). I've committed 3 bitcoin to this node, and am slowly adding channels based on my past experience. So far: lnbig, zerofeerouting, wallet of satoshi, breez, and stacker.news.
I am no expert but, feel free to open a channel with me.
Here's my 1ml.com link:
https://1ml.com/node/0316bf2be68ad2725bcbc14733b426f3132045677e591963b5d94a9c237991e1ae
Here's my node id:
0316bf2be68ad2725bcbc14733b426f3132045677e591963b5d94a9c237991e1ae@yuo4eyabzvx6ymdsne277tb5uddl3gqx7ipvrf5vxhivmilzr7zmw4qd.onion:9735
I’m no routing node expert, but one thing I consistently hear from great routing node operators is that there is no formula for success, Lightning is a fluid system that routing nodes must always adapt to.
It’s the same concept as “what got you here won’t take you to where you’re going next”.
At all times, routers are competing directly against all other routers, and winning strategies are quickly exploited until they are no longer winning strategies.
Not to say that the routing landscape is perfectly efficient, just that any strategies you hear from other node operators should be taken with a grain of salt.
The incentives for routing are set up in a way that great operators never share their best strategies, because those make them money… and the strategies people do share are typically being shared because they’re either common knowledge or aren’t super profitable.
It’s kind of like trading in this sense. If you had a trading algorithm that you could rely on for consistent profits, why would you share it with other traders, thereby lowering your profit opportunity?
Just my 2 sats, looking forward to hearing other views on this topic too.
Thanks, that makes sense - I figure this question may be a shot in the dark because if I were crushing it with my routing I certainly wouldn't be posting all my secrets :)
First 2 questions may be addressed by this post. It outlines some practically steps suggested by several papers.
https://medium.com/@notgeld/lightning-network-computation-layer-27c7ba81a214
@zerofeerouting happily shares his recipe. You may find some info in his twitter I think.
Thanks for sharing this. And definitely - I learned a lot from his latest on the Stephan Livera podcast: https://stephanlivera.com/episode/384/
I’m still getting the ins and outs after about a year of running a node, I think a follow of @alexbosworth on Twitter will help a lot
And in the spirit of @kr 's response and the idea that successful routing strategies is likely proprietary , perhaps anyone who has previously routed can share anything that did not work when you ran your node?
Was there anything you would have avoided when you started? Or something you did that never panned out?
I think the most important thing is to have a lot of capital.
That's what I got from SN so far. Balancing and strategic channels etc etc is important and all. But if you don't have the raw mass of capital it won't be successfull. Or at least not profitable.
Personally, I don't generate any revenue or Sats from my node. I set all channels to 0/0 fees. I'd rather help facilitate payment on the lightning network than have my BTC sitting there and doing nothing. Maybe, some day, some kind soul will see that my node routed a lot of payments for them and keysend me some Sats, but if that never happens, I'm fine with that too. I just want to Lightning to succeed and I'm super content to just do my part and help.
This post was featured on This Day in Stacker News.
Also see:
Adventures in Running a Bitcoin Lightning Node
#73901
https://medium.com/coinmonks/adventures-in-running-a-bitcoin-lightning-node-part-1-the-awakening-61345585acc3
Adventures in Running a Bitcoin Lightning Node — Part 2: Growing up…
#73937
https://medium.com/coinmonks/adventures-in-running-a-bitcoin-lightning-node-part-2-growing-up-430c6efcd40