Parameters with “0” values are passing to nailgun. If feature is required this system parameters there must be sanity check for it.

[Sergii]

Detailed bug description:
In my case I deploy cluster with dpdk feature enabled. And in "Node Attributes -> CPU pinning -> DPDK CPU pinning" values are "0". And Fuel allows deploy, but fails on openstack deploying stage with "All nodes are finished. Failed tasks: Task[netconfig/1] Stopping the deployment process!"

And in ovs-vswitchd.log we have:
2017-03-23T13:37:09.714Z|00009|dpdk|INFO|EAL ARGS: ovs-vswitchd -c 0x1 --socket-mem 256 -n 2 --vhost-owner libvirt-qemu:kvm --vhost-perm 0664
2017-03-23T13:37:09.726Z|00002|daemon_unix|ERR|fork child died before signaling startup (killed (Aborted), core dumped)
2017-03-23T13:37:09.726Z|00003|daemon_unix|EMER|could not initiate process monitoring

Because we don't have cpu for it

Steps to reproduce:
 1. Create cluster
 2. Enable DPDK
 3. Set DPDK CPU pinning to "0"
 4. Deploy cluster

Expected results:
 Error message with reasonable description.

Actual result:
 Cluster starts to deploy

Reproducibility:
 Checked on Fuel 10, snapshot 1499, 1507