Allow 39 atached EBS devices.

AWS has soft support limit for 40 attached EBS devices. Assuming there is just
one root device, use the rest for persistent volumes.

The devices will have name /dev/xvdba - /dev/xvdcm, leaving /dev/sda - /dev/sdz
to the system.

Also, add better error handling and propagate error
"Too many EBS volumes attached to node XYZ" to a pod.

plugin/pkg/scheduler/algorithmprovider/defaults/defaults.go

@@ -21,6 +21,7 @@ import (
 	"os"
 	"strconv"
 
+	"k8s.io/kubernetes/pkg/cloudprovider/providers/aws"
 	"k8s.io/kubernetes/pkg/util/sets"
 	"k8s.io/kubernetes/plugin/pkg/scheduler"
 	"k8s.io/kubernetes/plugin/pkg/scheduler/algorithm"
@@ -31,10 +32,6 @@ import (
 	"github.com/golang/glog"
 )
 
-// Amazon recommends having no more that 40 volumes attached to an instance,
-// and at least one of those is for the system root volume.
-const DefaultMaxEBSVolumes = 39
-
 // GCE instances can have up to 16 PD volumes attached.
 const DefaultMaxGCEPDVolumes = 16
 
@@ -117,7 +114,7 @@ func defaultPredicates() sets.String {
 			"MaxEBSVolumeCount",
 			func(args factory.PluginFactoryArgs) algorithm.FitPredicate {
 				// TODO: allow for generically parameterized scheduler predicates, because this is a bit ugly
-				maxVols := getMaxVols(DefaultMaxEBSVolumes)
+				maxVols := getMaxVols(aws.DefaultMaxEBSVolumes)
 				return predicates.NewMaxPDVolumeCountPredicate(predicates.EBSVolumeFilter, maxVols, args.PVInfo, args.PVCInfo)
 			},
 		),