I've built a CoreOS Kubernetes (1.0.7) Cluster in AWS and everything seems to work swimmingly, except Volumes. I manually create a volume in the same zone as my nodes prior to bringing up an RC with the volume configured. As below:

 apiVersion: v1
 kind: ReplicationController
 metadata:
   name: jenkins
 spec:
   replicas: 1
   selector:
     app: jenkins
   template:
     metadata:
       labels:
         app: jenkins
     spec:
       containers:
       - name: jenkins
         image: jenkins
         ports:
         - containerPort: 8080
         - containerPort: 50000
         volumeMounts:
         - name: "jenkins-data"
           mountPath: "/var/jenkins_home"
         livenessProbe:
           httpGet:
             path: /
             port: 8080
           initialDelaySeconds: 60
           timeoutSeconds: 180
       volumes:
       - name: "jenkins-data"
         awsElasticBlockStore:
           volumeID: aws://us-east-1c/vol-abcd123
           fsType: ext4

The pod begins to come up, but waits for the volume to mount. The volume properly attaches to the instance (I see it in fdisk) but is unable to be mounted properly when I turn up logging on the kublet:

 safe_format_and_mount[1504]: Running: fsck.ext4 -a /dev/xvdf
 safe_format_and_mount[1508]: /dev/xvdf: clean, 11/655360 files, 79663/2621440 blocks
 safe_format_and_mount[1513]: Running: mount -o discard,defaults /dev/xvdf /var/lib/kubelet/plugins/kubernetes.io/aws-ebs/mounts/aws/us-east-1c/vol-abcd123
 safe_format_and_mount[1516]: mount: mount point /var/lib/kubelet/plugins/kubernetes.io/aws-ebs/mounts/aws/us-east-1c/vol-abcd123 does not exist
 safe_format_and_mount[1519]: Disk /dev/xvdf looks formatted but won't mount.  Giving up.

Now that error is correct, the /var/lib/kubelet/plugins dir exists but is empty from there out. Seems like I'm missing something here that should be creating this directory? I've looked around previous issues and seen many regarding safe_format_and_mount but haven't seen any regarding that directory. Any ideas?