This is not a channel size issue, I'm willing to open a big channel, but I don't even get to that step. This happens when I try to connect to their node.

Yes, this is a tor node. That might be it then. I'm running it on Voltage though, so I'm not sure how I'd go about restarting tor specifically.

Well as of October 2024 sideshift has removed LN support :(

Ok I did figure out the #2, which is the sweep transaction. It's explained here.

This seems super innefficient though... :\

Yeah my question was more like why did this happen? it's not like my node or this peer were out of communication. Like I said we were communicating fine, even forwarding several other HTLCs. I have like 13 routing events with this peer since Jan 13th. And I saw this stuck HTLC for several days before the force close, I was afraid it was going to happen. It seems more like a bug.

Also by analyzing the structure of the transaction, I can see that my output is the to_remote (unencumbered one). So why the extra transaction, is labeled as a "sweep" by RTL but I was under the impression that only the to_local needed a sweep. So this other tx makes no sense to me.