We usually have to recover funds and close all channels due to data corruption, because even with a no break system, sometimes we have a crash/surge/restart on the machine and unfortunatelly, our channel.db gets corrupted and cant be read by LND properly, this will probably be addressed soon, but for now that's the way it is.
Recently, it has been discovered a bug where whoever runs LND tries to recover from a crash and tries the SCB Recovery, CLN ignores the SCB recovery signal and keep the channel upo and active on the remote side.
The problem is that your funds will be automatically frozen forever, since remote CLN node won't be force closing it on their side, because to them, your channel is working fine.
A Famous node with CLN with this BUG is BCASH_IS_TRASH, apart from using the bugged CLN, they WON'T BE FOUND ANYWHERE you could contact them and ask to force close the channel on their side, if your channel is locked in that "waiting close" state, they are nowhere to be found and you are left forever with your funds locked.
Here's the sollution: you need to use a specific version of chantools and execute a command specific for this problem, and for those who aren't a tech savvy, this could be a problem and you risk losing access to your funds for years.
For those with the problem, the solution is on this link:
seems like a disagreement about spec rather than "bug": https://github.com/ElementsProject/lightning/issues/5876
not that that's helpful to the user :)
reply