if you are using the @latest like i recommended, when you run the dockerfile now it will patch to a version that has the fix for that waitgroup panic... it was an edge case because it was triggered by clients sending malformed messages repeatedly (it ignores them after 16 messages wrong) i'm not sure that i've written that code correctly but for sure the waitgroup done in that clause causes two decrements of the counter which makes it go negative when a client keeps sending garbage