Akka cluster sharding hosted entities histogram errors when nodes go away

[polymorphic]

Greetings --

I have an Akka cluster service using cluster sharding and instrumented with Kamon. There are instances when the service code passivates cluster entities as described in the Akka documentation. Doing so has the effect of decrementing the number of sharded entities, something that causes Kamon to log the following warning:

2021-02-15 10:39:26,969 WARN  kamon.metric.Histogram traceId=undefined spanId=undefined  - Failed to record value [-1] on [akka.cluster.sharding.shard.hosted-entities,{type=I-typed,system=svc-system}] because the value is outside of the configured range. The recorded value was adjusted to the highest trackable value [3600000000000]. You might need to change your dynamic range configuration for this metric
java.lang.ArrayIndexOutOfBoundsException: Histogram recorded value cannot be negative.

I read it as the Kamon code assuming that the number of shared entities is monotonically increasing; that is not the case when passivation occurs. Have others ran into this problem? Perhaps there's a configuration parameter I missed...