Will a dead/inaccessible RHCS/Ceph monitor node mentioned in /etc/ceph/ceph.conf affect ceph client performance ?

Issue

• Will a dead/inaccessible RHCS/Ceph monitor node mentioned in /etc/ceph/ceph.conf affect ceph client performance?

• If a monitor node dies or is taken out of service but its IP address is still used by the client, perhaps in a Qemu KVM-RBD connection string, will it have any impact on the connection?

• This is possible in Openstack virtual machines, since it is not practical to quickly update the existing VM-RBD connection string since it is statically stored inside a MySQL DB.