Volume replica set doesn't update on node taint / destruction

After a scale down and scale back up (a complete node cycle)

The volume replica sets now have the wrong IP, and cannot bring up the volume. Anyone know

kubectl exec -it portworx-k6dpr  -n kube-system /opt/pwx/bin/pxctl volume inspect pvc-13524216-7ea4-11e7-98d9-025dab342ba7
Volume  :  56796491782559141
        Name                     :  pvc-13524216-7ea4-11e7-98d9-025dab342ba7
        Group                    :  rmq_vg
        Size                     :  8.0 GiB
        Format                   :  ext4
        HA                       :  2
        IO Priority              :  HIGH
        Creation time            :  Apr 15 06:51:45 UTC 2020
        Shared                   :  yes
        Status                   :  up
        State                    :  Attached: 26b197b0-1846-43a6-996d-2f41d085e254 (IP HIDDEN FOR FOURM POST) ***.***.***.*** 
        Device Path              :  /dev/pxd/pxd56796491782559141
        Labels                   :  app=rabbitmq-ha,group=rmq_vg,io_priority=high,namespace=customer-10042931-production,pvc=data-rmq-rabbitmq-ha-0,repl=2,shared=true
        Reads                    :  0
        Reads MS                 :  0
        Bytes Read               :  0
        Writes                   :  0
        Writes MS                :  0
        Bytes Written            :  0
        IOs in progress          :  0
        Bytes used               :  0 B
        Error in stats           :  Node(s) in replication sets are not online
        Replica sets on nodes:
                Set 0
                  Node           : (IP HIDDEN FOR FOURM POST) ***.***.***.*** (Pool bc8590fa-5aac-4a2c-9ee8-7ab725a035d4 )
                  Node           : (IP HIDDEN FOR FOURM POST) ***.***.***.*** (Pool 0959b11v-dea1-41cd-825f-c69307dc0194 )*
        Replication Status       :  Not in quorum
        Volume consumers         :
                - Name           : rmq-rabbitmq-ha-0 (cb4161b7-82c6-11ea-95d9-025dab342c28) (Pod)
                  Namespace      : customer-10042931-production
                  Running on     : ***-***-***-***.**-******-*
                  Controlled by  : rmq-rabbitmq-ha (StatefulSet)

Can anyone shed light on how this is meant to work under Auto Scale Groups?