During diego_database update, NCP job fails at pre-start phase with error in log "vmware_nsxlib.v3.exceptions.ServiceClusterUnavailable: Service cluster: 'https://<NSX-T-MANAGER>' is unavailable. Please, check NS X setup and/or configuration".
During NCP pre-start, it will create pre-configured ipblocks and ippools if they do not exist on NSX-T yet. "ServiceClusterUnavailable" error is returned by NSX-T Manager. Even though NSX-T Manager could respond some other APIs, the create ipblocks or ippools calls may fail, for example in the case of NSX-T transport node is unhealthy. To troubleshoot this issue, please review NSX-T Manager and resolve any alarms.