StreamStorage architecture question

[HeikoMueller]

I have a general question regarding StreamStorage for stateful functions. I have a working Pulsar cluster of 3 nodes (Broker, Bookie, Zookeeper, Function Worker). I want to enable StreamStorage for stateful functions, and I added this to the Bookie config:

advertisedAddress={IP-ADDRESS}
bookieId={NODE_ID}

storageserver.grpc.port=4181
dlog.bkcEnsembleSize=3
dlog.bkcWriteQuorumSize=2
dlog.bkcAckQuorumSize=2
storage.range.store.dirs=data/bookkeeper/ranges
storage.serve.readonly.tables=false
storage.cluster.controller.schedule.interval.ms=30000
extraServerComponents=org.apache.bookkeeper.stream.server.StreamStorageLifecycleComponent

And I can verify that it is working with

telnet localhost 4181

The startup of the cluster works as long as I have this setting ONLY ON ONE bookie. As soon as I add this on each of the three bookies, the Bookkeeper cluster is crashing at start up.

Is this expected behavior? Would I have to adjust the config on the other two Bookies to enable StreamStorage on them as well or is it intended to only have one StreamStorage Bookie?

Thank you for clarification

....or maybe someone can confirm that they are running more than one Bookie in parallel with these settings?