Suppose we are using viewstamp replication, allowing for up to F process failures.

$1.$ What is the minimum number of replicas that we require for viewstamp replication?

$2.$ When does the primary server acknowledge the update back to the client?

$3.$ Suppose the primary and several replicas crash. $$What guarantees that at least one of the surviving replicas will know of the update?

$4.$ Suppose the new primary was not one of the replicas contacted by the old primary to perform the update. $$How will the new primary learn of the update?

$5.$ What failure model does viewstamp replication assume?$$