Concourse not removing retired workers from the DB

Hello,

we have the following setup running on AWS (Concourse v6.7.1):
one Concourse web instance and a few concourse workers managed by an autoscaling group. Usually once a day our worker instances (spot instances) are being recreated and get connected to the Master/web instance. The issue is our retired workers stay in the worker list (fly workers -t live) indefinitelly until one of us manually prunes them:
name containers platform tags team state version age
ip-X-X-X-X.eu-west-1.compute.internal 84 linux none none retiring 2.2 2d
ip-X-X-X-X.eu-west-1.compute.internal 43 linux none none running 2.2 3h25m
ip-X-X-X-X.eu-west-1.compute.internal 52 linux none none retiring 2.2 1d
ip-X-X-X-X.eu-west-1.compute.internal 92 linux none none running 2.2 17h22m
ip-X-X-X-X.eu-west-1.compute.internal 51 linux none none retiring 2.2 1d
ip-X-X-X-X.eu-west-1.compute.internal 59 linux none none running 2.2 9h11m
ip-X-X-X-X.eu-west-1.compute.internal 38 linux none none retiring 2.2 22h49m
ip-X-X-X-X.eu-west-1.compute.internal 48 linux none none running 2.2 3h26m
ip-X-X-X-X.eu-west-1.compute.internal 67 linux none none retiring 2.2 1d
ip-X-X-X-X.eu-west-1.compute.internal 41 linux none none retiring 2.2 19h41m
ip-X-X-X-X.eu-west-1.compute.internal 55 linux none none running 2.2 14h52m

Does anybody have any clue on how we can fix this or troubleshoot this further?