Control-service (actors) become stuck reported as ERROR on network blip
Closed this issue · 2 comments
vlofgren commented
Probably a try-catch that's not quite right somewhere. Attaching a large number of stacktraces to sift through.
Need to at this point restart control-service to unstick them.
vlofgren commented
I think it should be possible to replicate by starting in docker-compose and shutting down mariadb. Maybe we need an emphemeral state for this event?
vlofgren commented
Can not reproduce this error. Seems to have been incidentally fixed by some of the work in the mqapi.