Zaps Failing, Fragility, and Bet Update \ stacker news ~meta

pull down to refresh

7361 sats \ 55 comments \ @denlillaapan 15 Jan meta

I've been pretty negative on the whole cowboy circus (i.e., here #829853; and there's more to come). What most troubles me is the fragility: I have no real oversight into wth is going on so when zaps fail randomly—which they do, suddenly and without warning, for all manner of reasons I can't figure out, (#839353), I get swamped by CCs and then have to troubleshoot wth is up.

Dudes, I'm a simple writer and editor. All I do is fuck around in google docs and read books and occasionally dig out bank balance statements from dusty, old archive (#847595). Not fixing up nodes or connecting wallets or figuring out why they malfunction.

Today, two things apparently happened. First, I had hit my weekly zap budget so my zaps started failing because I had zapped so lavishly before. uh-hu, good problem. Second, on the computer the NWC connection just decided to stop working. (After having reset the zap limit, I could freely zap on the phone, but the computer still said no.)

Detaching and re-attaching seems to have worked (congrats, @Shugard, for receiving a bunch of test zaps! #850630).

Now, this sort of fragility is precisely what I most object to. When I drive a car I want the car to work; if there is an error I don't want to have to deal with it. The way we do things in a world of specialization is to make things idiot proof such that idiots like me can still manage, and then have extremely simple procedures in place when things go wrong—in the car example, a light on the dashboard that tells me to get the car to a professional (i.e., mechanic).

Anyway, rant for another time.

Now, we also have a nice little predyx bet on whether SN hits #1,400,000 at the end of 2025, of which I'm a proud (foolish?!) owner of a "no" contracts—at least I think so, but who tha hell knows.

The post count at the beginning of 2025 was #831750 or so, meaning that we have advanced roughly 19,000 items in 15 days—a rate of 1,263 a day. For the bet to become "yes," SN needs about 1,550 new items daily; so we're structurally falling behind. Put differently, we're only at about 3.33% of the way to the goal while 4.1% of the year has already passed.

A little bit more with this trend and it's going to run away from us/you/the suckers on the other side of the bet.

Also, I made a cute lil excel sheet graph that we can use to keep track of this whenever I remember. Happy posting, everyone!

view all related items

754 sats \ 21 replies \ @k00b 15 Jan

I'm sorry this is hard to use and hard to keep it working. We know it's difficult to use and have a lot of work left to do. We will make it better.

For context, it's quite a bit more complex than say making a button do something when you press it. We are coordinating a payment between multiple computers we don't control. Our goal is to get it close to button-press-like easiness in terms of your experience, but it's going to take us time, lots of fucking work, and experimentation to hide and communicate that complexity.

This thing is a Model T right now - you crank it, press 2 buttons, 3 switches, blow in a tube, pray, then maybe it gets you where you're going. It'll get better faster than a model T did because it's software, but a lot of what we're doing is new.

Also, your complaining helps a lot so please keep it up. I struggle with how angry you sound sometimes but that's on me. Most people in your shoes would've given up. Props to you!

200 sats \ 19 replies \ @elvismercury 15 Jan

My zaps fail only rarely -- when they do, the logs report something about routing fees are too high. When I retry, it has always worked. I don't actually change anything about my setup -- I wouldn't know how to do more than add another channel, and the channel I have goes to a pretty fat LN pipe.

I know ya'll have no shortage of stuff to do, but it would really be illustrative to know what's happening on the backend, how you're iterating and evolving things, etc. I feel like this pragmatic guide to using LN in real life is totally dark territory. Could you an @ek do a podcast every couple weeks or something?

If I was in Austin I'd do a documentary on ya'll.

278 sats \ 18 replies \ @SimpleStacker 15 Jan

The only time my zaps fail is when I have a problem on my end. Usually lost internet connection or my computer that runs my wallet restarted.

I highly recommend everyone who wants to use Stacker News with an attached wallet to use a self-custodial lightning wallet with a direct channel. You will not pay any routing fees and you will have a consistent, reliable experience.

153 sats \ 6 replies \ @SimpleStacker 15 Jan

I highly recommend everyone who wants to use Stacker News with an attached wallet to use a self-custodial lightning wallet with a direct channel. You will not pay any routing fees and you will have a consistent, reliable experience.

I may have spoken too soon, lol.

Last night my computer restarted and when I reopened Alby, I reopened the wallet, but the NWC connection to SN couldn't get re-established. I ultimately had to delete the old NWC connection and start a new one.

I'm curious as to why that might be. The wallet logs didn't show any routing errors or anything, just timeouts on sending/receiving payments.

100 sats \ 5 replies \ @k00b 15 Jan

I've noticed after prior updates to Alby Hub my NWC strings are all invalidated. I think it might be a symptom of Alby Hub being kind of new and them correcting mistakes they made with our older NWC strings when we update. I hope this won't always be the case.

I'm curious as to why that might be. The wallet logs didn't show any routing errors or anything, just timeouts on sending/receiving payments.

NWC strings are doing two things:

giving us permission to do stuff on your node
telling us how to communicate with your node to do that permitted stuff

So, when an NWC string goes bad, it either has to do with those permissions going bad, ie being revoked or expiring, or the way we've been instructed to communicate with your node changing or not working.

Routing and fees are in another layer, let's call it The Lightning Layer, and The NWC Layer sits on top of it doing the things I mentioned above. If The NWC Layer goes bad, it means we can't communicate with The Lightning Layer, and the only indication we have that that's happening is The NWC Layer ignores us and we timeout.

176 sats \ 4 replies \ @SimpleStacker 15 Jan

Ok, then I'm reasonably sure the error was related to NWC strings and not Lightning.

What I don't quite understand is that the wallet was working right up til last night. My computer restarted after I went to bed, and it wasn't working after restarting the wallet in the morning. It started working again after I downloaded the new Alby update, deleting my old NWC ones, and making new ones.

30 sats \ 3 replies \ @k00b 15 Jan

Restarting must have invalidate the NWC strings on Alby Hub's end.

The NWC connection isn't persistent or anything, so it wouldn't affect our end. On our end, if you give us an NWC string, whether good or bad, all we can do is sign messages with it and publish them to the relay specified then wait for responses.

Timeouts usually happen when responses don't come back, and responses don't come back when Alby considers the NWC string invalid afaik.

0 sats \ 2 replies \ @SimpleStacker 15 Jan

So I guess I'm wondering if/where/when an Alby server sits between my computer and Stacker News?

Because I was running the same version of the Alby software on my computer as the night before, with the same NWC strings that failed in the morning. That shouldn't happen unless restarting somehow changed something on an Alby server somewhere, right?

Edit: Sorry, I know this is probably more a question for Alby. But I'm not really sure how NWC works, so this is all educational to me.

view replies

30 sats \ 10 replies \ @elvismercury 15 Jan

I have a channel to SN and some inbound I bought from @Rizful (link) which I recommend for learning tons about using / managing an LN node, either their cloud one or their awesome tutorial on setting up your own LND w/ Docker.

27 sats \ 9 replies \ @SimpleStacker 15 Jan

Curious as to why you'd get routing fee errors then, if you have a direct channel.

0 sats \ 8 replies \ @elvismercury 15 Jan

My guess is that since the whole point of this is that SN is not the custodian, and since other people don't necessarily have a channel open to SN, that it's something to do with that leg of things. If it were just me > SN > them via direct channels it would probably be fine.

108 sats \ 7 replies \ @k00b 15 Jan

Ah, by the sounds of it your direct channel probably gets depleted of inbound sometimes, so we go through the channel you bought (which has fees unlike your direct channel).

146 sats \ 4 replies \ @elvismercury 15 Jan

I just checked logs -- here's an example.

I think this is somebody trying to zap me, right? My channel to SN has 29k inbound; my other channel has 2.4m inbound. My midwit understanding is that

a) they tried to zap me b) they don't have a channel to SN c) the route to me someplace has fees that fail

Would love to be corrected on this one.

view replies

30 sats \ 1 reply \ @elvismercury 15 Jan

Except the ones I notice are when my zaps to others don't go through; and then I retry later, and they do. Wouldn't that mean it's the other person's channels that are the culprit?

view replies

137 sats \ 0 replies \ @denlillaapan OP 15 Jan

Appreciate it, sir. You keep on doing your thing and I keep doing mine (complaining, venting) and we'll both end up someplace good!

Also love this

This thing is a Model T right now - you crank it, press 2 buttons, 3 switches, blow in a tube, pray, then maybe it gets you where you're going

59 sats \ 2 replies \ @nym 15 Jan

SN and lightning are pushing the envelope of what is possible, so there will be bugs. Stick to onchain if you want stability.

0 sats \ 1 reply \ @sangekrypto 15 Jan

did you succeed in zaping? I don't understand what goes beyond limits, isn't everything systemized?

0 sats \ 0 replies \ @nym 15 Jan

No problems here

93 sats \ 4 replies \ @Undisciplined 15 Jan

I also took the "no" side of that bet.

You also wrote about a bet against $40T in US National Debt. Did you pull the trigger on that?

41 sats \ 0 replies \ @denlillaapan OP 15 Jan

Yes, but that was with the other betting site and someone said they were shifting their system early this year—so I don't know what's gonna happen about that

41 sats \ 1 reply \ @SimpleStacker 15 Jan

Are you at all worried that people can manipulate the outcome? Like by posting with bots?

0 sats \ 0 replies \ @Undisciplined 15 Jan

No, I'm not particularly worried. Bots can alter the price, but that doesn't change the eventual outcome or my decisions to buy/sell at particular prices.

0 sats \ 0 replies \ @Coinsreporter 15 Jan

This is bearish!

Both are gonna be. Yes, yes, yes.

Perth and Renegades are out btw, but we can still make a few sats there.

61 sats \ 0 replies \ @grayruby 15 Jan

Things have slowed down on SN since the wallet stuff. Probably got rid a lot of useless posts and comments. May not hit our goal but improved the quality of content. Fair trade.

Let's see how the year plays out. SN does tend to go through lulls at certain points in the year.

29 sats \ 0 replies \ @DarthCoin 15 Jan

Another post for my "special collection"...

87 sats \ 0 replies \ @TresDMan 15 Jan

I feel you. I also just read through the comments and I think I'm actually a bit more confused.

Just based off of @koob's first comment, is this the way self-custodial LN payments work everywhere else, or is this clunkiness occurring because it's a new implementation for SN?

If it's like this everywhere else, I'd imagine it will be hard to gain widespread adoption. It takes a lot of research to get everything working correctly. That's not such a foreign concept for me, but it can feel very daunting sometimes.

87 sats \ 1 reply \ @charonnakamoto 15 Jan

I also failed a similar issue yesterday, and it wasn't enjoyable to debug #850124 I would also like a car that works. However, I feel we need to be more self-reliant and less reliant on the mechanics in areas like self-custody and sovereignty. In the context of SN, this might mean getting your hands dirty in logs, using LLMs to guide you if you're not a coder, or simply asking for help like I did. I'm not asking you to be a full-time coder, but I am asking you to do some basic Level 1 debugging and fixing(turning it off and on, haha) instead of taking it to the mechanic, which you did; it looks like. PS: Apologies for the preachy tone.

9 sats \ 0 replies \ @denlillaapan OP 15 Jan

No trouble at all, appreciate the remark!

24 sats \ 1 reply \ @cleaningup12 15 Jan

I have to admit, as soon as I get a noti saying cc's, I'm straight to my wallet log to look for the dreaded...

I've changed (raised) the parameters for max fee for etc, which I never had to do before

Not complaining obviously, koob and Co, work tirelessly to bring us this amazing platform

It's all darths fault, has to be

31 sats \ 0 replies \ @denlillaapan OP 15 Jan

agreed. Everything is on Darth

22 sats \ 3 replies \ @sasasuina 15 Jan

The basic principle of economics says that division of labor boosts efficiency. So, just go custodial and let the experts handle the tough stuff—everyone just needs to focus on their own area of genius! 😄

0 sats \ 2 replies \ @denlillaapan OP 15 Jan

Preeeecisely why, economically speaking, the CC nonsense troubles me

0 sats \ 1 reply \ @sasasuina 15 Jan

I get why CC might seem like nonsense. But I see it as a necessary fix during the transition to noncustodial wallets.

Stacker News used to be a custodial wallet, but I heard there were legal risks, so they switched to a noncustodial model. In my opinion, CC is more like a workaround to address the issues that came up during this transition. Without CC, the failure rate for zaps would be much higher. After all, for a zap to go through, both the sender and the receiver need to have their wallets properly set up. You can imagine that when they first switched, the percentage of users who had set up wallets was probably pretty low. That might’ve caused zap activity on Stacker News to drop significantly.

With CC, at least in some situations—like when a wallet isn’t set up—users can still receive zaps as CC.

When it comes to spending bitcoin on SN, CC works as a substitute in most cases, like posting, commenting, or tipping. But CC can’t be withdrawn as bitcoin. My guess is that this was a deliberate choice to avoid legal risks. Otherwise, if CC == bitcoin, Stacker News might still be classified as a custodial wallet.

When we spend CC, part of it goes into a reward pool. During the daily reward distribution, the CC in the pool is converted back into bitcoin and shared with qualified users. This is the only way for CC to turn back into bitcoin.

0 sats \ 0 replies \ @denlillaapan OP 15 Jan

Yeah I get this.

Pretty decent summary as to how I've understood the communication from @k00b etc

17 sats \ 3 replies \ @WeAreAllSatoshi 15 Jan

Maybe I’m just lower maintenance, but I haven’t had a ton of issues with this switch. I get the occasional zap failure, so I retry it from the notifications page and then it works. I have an auto withdraw set up and my SN balance is always low. Seems fine to me?

0 sats \ 2 replies \ @denlillaapan OP 15 Jan

Well, good for you.

24 sats \ 1 reply \ @WeAreAllSatoshi 15 Jan

Sorry, I didn’t mean to come across as snarky. Just sharing my experience :) it’s clear that there are some improvements to be made and some rough edges to smooth out

100 sats \ 0 replies \ @denlillaapan OP 15 Jan

is all good :)

17 sats \ 8 replies \ @realBitcoinDog 15 Jan

Yea I’m up to 521 cc’s. Idk if there is a way to trade them for sats. Can I pay territory fees with them? I think so? Can I use them when I pay to post?