The operational issues were due to an underlying configuration of nodes in our San Franciso cluster. We informed our cloud provider and we have been working closely with them regarding the issue. Our cloud provider found an issue with their cluster deployment and subsequently have rolled out an update on their end which required a rolling replacement of all the VMs in the cluster for changes to take place. All our nodes are healthy again with 100% liveliness checks. After thorough checks, we are confident that the fix is working and hence have opened up the creation of new dynamo nodes in San Francisco region.