pscli can't stop peerswapd when LND is down

[grubles]

If LND is shut down while peerswapd is running, pscli seems to be unable to stop peerswapd:

$ ps aux | grep peerswapd
user     2605252  0.2  0.0 1843616 22016 pts/6   Sl   15:11   0:00 ./builds/peerswap/out/peerswapd
user     2605296  0.0  0.0   6024  2008 pts/6    S+   15:11   0:00 grep --color=auto peerswapd

$ ./builds/peerswap/out/pscli stop
2023/01/12 15:12:01 rpc error: code = Unavailable desc = connection error: desc = "transport: Error while dialing dial tcp 127.0.0.1:42069: connect: connection refused"

This might be related to the bug where many (zombie?) peerswap processes are running simultaneously somehow?

[nepet]

Have the zombie processes not been peerswap plugin (cln) related?

[grubles]

There was one report from statespacedev on Discord about zombie peerswap-plugin processes. I experienced zombie plugin processes locally a month or so ago too. In this instance, there was only one running peerswapd but it was unresponsive.