MetroCluster vserver config degraded during upgrade to ONTAP 9.3

One of the nodes has been upgraded to 9.3, but after the first switchback one or more of the vservers is in a degraded state.

 

 

ClusterA::*> metrocluster vserver show

...
                              Cluster  : ClusterA
                              Vserver  : SVM_B
                      Partner Vserver  : SVM_B-mc
                  Configuration State  : degraded
                        Stream Status  : operation-failed
                    Corrective Action  : Resynchronize the configuration using "metrocluster vserver resync" command.
                                         Contact technical support for assistance if configuration-state does not change to                                          healthy.
                            Failed Row : Table Name: vserver_web_services, Operation: get, Fields: SVM_B-mc rest_api data                                          1.0.0 Remote Administrative REST API Support This service supports a RESTful
                                         Interface that can be used to remotely manage all elements of the cluster
                                         infrastructure. - admin readonly vsadmin vsadmin-protocol vsadmin-readonly
                                         vsadmin-volume false false
                         Failed Reason : Apply failed for Object: vserver_web_services Method: baseline. Reason: entry
                                         doesn't exist
                 Last apply fail time  : 10/10/2020 15:14:30

ClusterA::*> version

NetApp Release 9.1P20: Mon Oct 28 14:37:28 UTC 2019

ClusterB::*> version

NetApp Release 9.3P20: Wed Sep 23 21:44:43 UTC 2020