A stuck HTLC transaction might make a channel unresponsive and trigger a force close that cost fee on-chain and if it happens during a high fee period it can cost you much more than the fee you earn in one year on your entire node. Or just the other peer fucked up and trigger some force close automatically or manually and if you open the channel you paid for closing. Also you can get some old channel commitment protocle that predefine force close fee at opening, if you end up needing to force close the channel (other peer offline) and you have sats on your side, you will trigger the force close but the fee cannot be change and might never get mined.
In more than one year with dozens of channels, the stuck HTLC close never happened, however second case happened and cost 80$ of fee at the time, third happen with millions sats on my side and my closing transaction was stuck in meme pool for 3 months with 12sats/vB fee, the fees end up low enough for it to pass but it was stressful. Thus the last one was mostly my fault, the other peer was not really offline but still we tried everything togheter and it was impossible to close it before fee went down again.
Good response, thanks.
I've been running my node and managing my own channels for a few years and have never experienced any of these cases. But you helped me understand better what the potential threats could be.
I wonder if the benefits of using Lightning make up for these rare events that can be more costly than we wanted? Maybe we're focusing so much on these potential negatives and not enough on the positives, and it's stinting the growth of the network and scaring people away?
Just pondering here.
reply