Open grubles opened 7 months ago
Weird! IO level logging might help here, but we would still be guessing what LND gets upset by...
Here is an IO level log for the first 1 minute or so of starting CLN:
I noticed this happening again only after I had restarted the CLN node. This time with LND 0.17.4-beta and CLN v24.02.2.
Turns out it is a gossip_query_reply
in which the sorting order was expected by lnd
but potentially CLN did not sort them?
A change in the past week or two has caused my signet test nodes to be unable to connect to each other.
One is an LND 0.17.0-beta.rc2 and the other is latest CLN master. The only thing that I've changed recently is the CLN node, which I try to track master with. Currently it's running at commit 3f4306eea99c7a61af2fa0132de2d191cf609cec.
LND logs when the CLN node is disconnected:
CLN logs while this happens: