Merge pull request #66464 from wongma7/round-overflow

Automatic merge from submit-queue (batch tested with PRs 66464, 66488). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.

Avoid overflowing int64 in RoundUpSize and return error if overflow int

**What this PR does / why we need it**:
There are many places in plugins (some I may have missed) that we naively convert a resource.Quantity.Value() which is an int64, to an int, which may be only 32 bits long.

Background, optional to read :): Kubernetes canonicalizes resource.Quantities, and from what I have seen testing creating PVCs, decimalSI is the default. If a quantity is in `decimalSI` format and its value in bytes would overflow an int64, e.g. `10E`, nothing happens. If it is in binarySI and its value in bytes would overflow an int64, e.g. `10Ei`, it is set down to 2^63-1 and there's no overflow of the field value. But there may be overflow later in the code which is what this PR is addressing.

* Change `RoundUpSize` implementation to avoid overflowing `int64`
* Add `RoundUp*Int` functions for use when an `int` is expected instead of an `int64`, because `int` may be 32bits and naively doing `int($INT64_VALUE)` can lead to silent overflow. These functions return an error if overflow has occurred.
* Rename `*GB` variables to `*GiB` where appropriate for maximum clarity
* Use `RoundUpToGiB` instead of `RoundUpSize` where possible

**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Fixes #

**Special notes for your reviewer**: please review carefully as we don't have e2e tests for most plugins!

**Release note**:

```release-note
NONE
```
edit: remove 'we do not need to worry about...'. yes we do, i worded that badly :))

pkg/cloudprovider/providers/aws/aws.go

@@ -2510,9 +2510,8 @@ func (c *Cloud) ResizeDisk(
 		descErr := fmt.Errorf("AWS.ResizeDisk Error describing volume %s with %v", diskName, err)
 		return oldSize, descErr
 	}
-	requestBytes := newSize.Value()
 	// AWS resizes in chunks of GiB (not GB)
-	requestGiB := volumeutil.RoundUpSize(requestBytes, 1024*1024*1024)
+	requestGiB := volumeutil.RoundUpToGiB(newSize)
 	newSizeQuant := resource.MustParse(fmt.Sprintf("%dGi", requestGiB))
 
 	// If disk already if of greater or equal size than requested we return

pkg/cloudprovider/providers/gce/gce_disks.go

@@ -746,9 +746,8 @@ func (gce *GCECloud) ResizeDisk(diskToResize string, oldSize resource.Quantity,
 		return oldSize, err
 	}
 
-	requestBytes := newSize.Value()
 	// GCE resizes in chunks of GiBs
-	requestGIB := volumeutil.RoundUpSize(requestBytes, volumeutil.GIB)
+	requestGIB := volumeutil.RoundUpToGiB(newSize)
 	newSizeQuant := resource.MustParse(fmt.Sprintf("%dGi", requestGIB))
 
 	// If disk is already of size equal or greater than requested size, we simply return

pkg/cloudprovider/providers/openstack/openstack_volumes.go

@@ -411,13 +411,15 @@ func (os *OpenStack) ExpandVolume(volumeID string, oldSize resource.Quantity, ne
 		return oldSize, fmt.Errorf("volume status is not available")
 	}
 
-	volSizeBytes := newSize.Value()
 	// Cinder works with gigabytes, convert to GiB with rounding up
-	volSizeGB := int(volumeutil.RoundUpSize(volSizeBytes, 1024*1024*1024))
-	newSizeQuant := resource.MustParse(fmt.Sprintf("%dGi", volSizeGB))
+	volSizeGiB, err := volumeutil.RoundUpToGiBInt(newSize)
+	if err != nil {
+		return oldSize, err
+	}
+	newSizeQuant := resource.MustParse(fmt.Sprintf("%dGi", volSizeGiB))
 
 	// if volume size equals to or greater than the newSize, return nil
-	if volume.Size >= volSizeGB {
+	if volume.Size >= volSizeGiB {
 		return newSizeQuant, nil
 	}
 
@@ -426,7 +428,7 @@ func (os *OpenStack) ExpandVolume(volumeID string, oldSize resource.Quantity, ne
 		return oldSize, err
 	}
 
-	err = volumes.expandVolume(volumeID, volSizeGB)
+	err = volumes.expandVolume(volumeID, volSizeGiB)
 	if err != nil {
 		return oldSize, err
 	}

pkg/volume/aws_ebs/aws_util.go

@@ -83,11 +83,13 @@ func (util *AWSDiskUtil) CreateVolume(c *awsElasticBlockStoreProvisioner) (aws.K
 	tags["Name"] = volumeutil.GenerateVolumeName(c.options.ClusterName, c.options.PVName, 255) // AWS tags can have 255 characters
 
 	capacity := c.options.PVC.Spec.Resources.Requests[v1.ResourceName(v1.ResourceStorage)]
-	requestBytes := capacity.Value()
 	// AWS works with gigabytes, convert to GiB with rounding up
-	requestGB := int(volumeutil.RoundUpSize(requestBytes, 1024*1024*1024))
+	requestGiB, err := volumeutil.RoundUpToGiBInt(capacity)
+	if err != nil {
+		return "", 0, nil, "", err
+	}
 	volumeOptions := &aws.VolumeOptions{
-		CapacityGB: requestGB,
+		CapacityGB: requestGiB,
 		Tags:       tags,
 		PVCName:    c.options.PVC.Name,
 	}
@@ -147,7 +149,7 @@ func (util *AWSDiskUtil) CreateVolume(c *awsElasticBlockStoreProvisioner) (aws.K
 		glog.Errorf("error building labels for new EBS volume %q: %v", name, err)
 	}
 
-	return name, int(requestGB), labels, fstype, nil
+	return name, requestGiB, labels, fstype, nil
 }
 
 // Returns the first path that exists, or empty string if none exist.

pkg/volume/azure_dd/azure_provision.go

@@ -100,8 +100,10 @@ func (p *azureDiskProvisioner) Provision(selectedNode *v1.Node, allowedTopologie
 	// maxLength = 79 - (4 for ".vhd") = 75
 	name := util.GenerateVolumeName(p.options.ClusterName, p.options.PVName, 75)
 	capacity := p.options.PVC.Spec.Resources.Requests[v1.ResourceName(v1.ResourceStorage)]
-	requestBytes := capacity.Value()
-	requestGB := int(util.RoundUpSize(requestBytes, 1024*1024*1024))
+	requestGiB, err := util.RoundUpToGiBInt(capacity)
+	if err != nil {
+		return nil, err
+	}
 
 	for k, v := range p.options.Parameters {
 		switch strings.ToLower(k) {
@@ -157,18 +159,18 @@ func (p *azureDiskProvisioner) Provision(selectedNode *v1.Node, allowedTopologie
 		if p.options.CloudTags != nil {
 			tags = *(p.options.CloudTags)
 		}
-		diskURI, err = diskController.CreateManagedDisk(name, skuName, resourceGroup, requestGB, tags)
+		diskURI, err = diskController.CreateManagedDisk(name, skuName, resourceGroup, requestGiB, tags)
 		if err != nil {
 			return nil, err
 		}
 	} else {
 		if kind == v1.AzureDedicatedBlobDisk {
-			_, diskURI, _, err = diskController.CreateVolume(name, account, storageAccountType, location, requestGB)
+			_, diskURI, _, err = diskController.CreateVolume(name, account, storageAccountType, location, requestGiB)
 			if err != nil {
 				return nil, err
 			}
 		} else {
-			diskURI, err = diskController.CreateBlobDisk(name, skuName, requestGB)
+			diskURI, err = diskController.CreateBlobDisk(name, skuName, requestGiB)
 			if err != nil {
 				return nil, err
 			}
@@ -196,7 +198,7 @@ func (p *azureDiskProvisioner) Provision(selectedNode *v1.Node, allowedTopologie
 			PersistentVolumeReclaimPolicy: p.options.PersistentVolumeReclaimPolicy,
 			AccessModes:                   supportedModes,
 			Capacity: v1.ResourceList{
-				v1.ResourceName(v1.ResourceStorage): resource.MustParse(fmt.Sprintf("%dGi", requestGB)),
+				v1.ResourceName(v1.ResourceStorage): resource.MustParse(fmt.Sprintf("%dGi", requestGiB)),
 			},
 			VolumeMode: volumeMode,
 			PersistentVolumeSource: v1.PersistentVolumeSource{

pkg/volume/cinder/cinder_util.go

@@ -169,9 +169,12 @@ func (util *DiskUtil) CreateVolume(c *cinderVolumeProvisioner) (volumeID string,
 	}
 
 	capacity := c.options.PVC.Spec.Resources.Requests[v1.ResourceName(v1.ResourceStorage)]
-	volSizeBytes := capacity.Value()
 	// Cinder works with gigabytes, convert to GiB with rounding up
-	volSizeGB := int(volutil.RoundUpSize(volSizeBytes, 1024*1024*1024))
+	volSizeGiB, err := volutil.RoundUpToGiBInt(capacity)
+	if err != nil {
+		return "", 0, nil, "", err
+	}
+
 	name := volutil.GenerateVolumeName(c.options.ClusterName, c.options.PVName, 255) // Cinder volume name can have up to 255 characters
 	vtype := ""
 	availability := ""
@@ -208,7 +211,7 @@ func (util *DiskUtil) CreateVolume(c *cinderVolumeProvisioner) (volumeID string,
 		}
 	}
 
-	volumeID, volumeAZ, volumeRegion, IgnoreVolumeAZ, err := cloud.CreateVolume(name, volSizeGB, vtype, availability, c.options.CloudTags)
+	volumeID, volumeAZ, volumeRegion, IgnoreVolumeAZ, err := cloud.CreateVolume(name, volSizeGiB, vtype, availability, c.options.CloudTags)
 	if err != nil {
 		glog.V(2).Infof("Error creating cinder volume: %v", err)
 		return "", 0, nil, "", err
@@ -221,7 +224,7 @@ func (util *DiskUtil) CreateVolume(c *cinderVolumeProvisioner) (volumeID string,
 		volumeLabels[kubeletapis.LabelZoneFailureDomain] = volumeAZ
 		volumeLabels[kubeletapis.LabelZoneRegion] = volumeRegion
 	}
-	return volumeID, volSizeGB, volumeLabels, fstype, nil
+	return volumeID, volSizeGiB, volumeLabels, fstype, nil
 }
 
 func probeAttachedVolume() error {

pkg/volume/flocker/flocker_util.go

@@ -49,7 +49,7 @@ func (util *FlockerUtil) DeleteVolume(d *flockerVolumeDeleter) error {
 	return d.flockerClient.DeleteDataset(datasetUUID)
 }
 
-func (util *FlockerUtil) CreateVolume(c *flockerVolumeProvisioner) (datasetUUID string, volumeSizeGB int, labels map[string]string, err error) {
+func (util *FlockerUtil) CreateVolume(c *flockerVolumeProvisioner) (datasetUUID string, volumeSizeGiB int, labels map[string]string, err error) {
 
 	if c.flockerClient == nil {
 		c.flockerClient, err = c.plugin.newFlockerClient("")
@@ -74,7 +74,10 @@ func (util *FlockerUtil) CreateVolume(c *flockerVolumeProvisioner) (datasetUUID
 
 	capacity := c.options.PVC.Spec.Resources.Requests[v1.ResourceName(v1.ResourceStorage)]
 	requestBytes := capacity.Value()
-	volumeSizeGB = int(volutil.RoundUpSize(requestBytes, 1024*1024*1024))
+	volumeSizeGiB, err = volutil.RoundUpToGiBInt(capacity)
+	if err != nil {
+		return
+	}
 
 	createOptions := &flockerapi.CreateDatasetOptions{
 		MaximumSize: requestBytes,

pkg/volume/glusterfs/glusterfs.go

@@ -744,7 +744,10 @@ func (p *glusterfsVolumeProvisioner) CreateVolume(gid int) (r *v1.GlusterfsVolum
 	capacity := p.options.PVC.Spec.Resources.Requests[v1.ResourceName(v1.ResourceStorage)]
 
 	// GlusterFS/heketi creates volumes in units of GiB.
-	sz := int(volutil.RoundUpToGiB(capacity))
+	sz, err := volutil.RoundUpToGiBInt(capacity)
+	if err != nil {
+		return nil, 0, "", err
+	}
 	glog.V(2).Infof("create volume of size %dGiB", sz)
 
 	if p.url == "" {

pkg/volume/photon_pd/photon_util.go

@@ -88,9 +88,11 @@ func (util *PhotonDiskUtil) CreateVolume(p *photonPersistentDiskProvisioner) (pd
 	}
 
 	capacity := p.options.PVC.Spec.Resources.Requests[v1.ResourceName(v1.ResourceStorage)]
-	volSizeBytes := capacity.Value()
-	// PhotonController works with GB, convert to GB with rounding up
-	volSizeGB := int(volumeutil.RoundUpSize(volSizeBytes, 1024*1024*1024))
+	// PhotonController works with GiB, convert to GiB with rounding up
+	volSizeGB, err := volumeutil.RoundUpToGiBInt(capacity)
+	if err != nil {
+		return "", 0, "", err
+	}
 	name := volumeutil.GenerateVolumeName(p.options.ClusterName, p.options.PVName, 255)
 	volumeOptions := &photon.VolumeOptions{
 		CapacityGB: volSizeGB,

pkg/volume/quobyte/quobyte_util.go

@@ -35,7 +35,10 @@ type quobyteVolumeManager struct {
 
 func (manager *quobyteVolumeManager) createVolume(provisioner *quobyteVolumeProvisioner, createQuota bool) (quobyte *v1.QuobyteVolumeSource, size int, err error) {
 	capacity := provisioner.options.PVC.Spec.Resources.Requests[v1.ResourceName(v1.ResourceStorage)]
-	volumeSize := int(util.RoundUpSize(capacity.Value(), 1024*1024*1024))
+	volumeSize, err := util.RoundUpToGiBInt(capacity)
+	if err != nil {
+		return nil, 0, err
+	}
 	// Quobyte has the concept of Volumes which doen't have a specific size (they can grow unlimited)
 	// to simulate a size constraint we set here a Quota for logical space
 	volumeRequest := &quobyteapi.CreateVolumeRequest{

pkg/volume/rbd/rbd_util.go

@@ -579,7 +579,10 @@ func (util *RBDUtil) CreateImage(p *rbdVolumeProvisioner) (r *v1.RBDPersistentVo
 	capacity := p.options.PVC.Spec.Resources.Requests[v1.ResourceName(v1.ResourceStorage)]
 	volSizeBytes := capacity.Value()
 	// Convert to MB that rbd defaults on.
-	sz := int(volutil.RoundUpSize(volSizeBytes, 1024*1024))
+	sz, err := volutil.RoundUpSizeInt(volSizeBytes, 1024*1024)
+	if err != nil {
+		return nil, 0, err
+	}
 	volSz := fmt.Sprintf("%d", sz)
 	mon := util.kernelRBDMonitorsOpt(p.Mon)
 	if p.rbdMounter.imageFormat == rbdImageFormat2 {

pkg/volume/storageos/storageos.go

@@ -602,7 +602,11 @@ func (c *storageosProvisioner) Provision(selectedNode *v1.Node, allowedTopologie
 		c.labels[k] = v
 	}
 	capacity := c.options.PVC.Spec.Resources.Requests[v1.ResourceName(v1.ResourceStorage)]
-	c.sizeGB = int(util.RoundUpSize(capacity.Value(), 1024*1024*1024))
+	var err error
+	c.sizeGB, err = util.RoundUpToGiBInt(capacity)
+	if err != nil {
+		return nil, err
+	}
 
 	apiCfg, err := parsePVSecret(adminSecretNamespace, adminSecretName, c.plugin.host.GetKubeClient())
 	if err != nil {

pkg/volume/util/util.go

@@ -361,7 +361,11 @@ func CalculateTimeoutForVolume(minimumTimeout, timeoutIncrement int, pv *v1.Pers
 // RoundUpSize(1500 * 1024*1024, 1024*1024*1024) returns '2'
 // (2 GiB is the smallest allocatable volume that can hold 1500MiB)
 func RoundUpSize(volumeSizeBytes int64, allocationUnitBytes int64) int64 {
-	return (volumeSizeBytes + allocationUnitBytes - 1) / allocationUnitBytes
+	roundedUp := volumeSizeBytes / allocationUnitBytes
+	if volumeSizeBytes%allocationUnitBytes > 0 {
+		roundedUp += 1
+	}
+	return roundedUp
 }
 
 // RoundUpToGB rounds up given quantity to chunks of GB
@@ -376,6 +380,32 @@ func RoundUpToGiB(size resource.Quantity) int64 {
 	return RoundUpSize(requestBytes, GIB)
 }
 
+// RoundUpSizeInt calculates how many allocation units are needed to accommodate
+// a volume of given size. It returns an int instead of an int64 and an error if
+// there's overflow
+func RoundUpSizeInt(volumeSizeBytes int64, allocationUnitBytes int64) (int, error) {
+	roundedUp := RoundUpSize(volumeSizeBytes, allocationUnitBytes)
+	roundedUpInt := int(roundedUp)
+	if int64(roundedUpInt) != roundedUp {
+		return 0, fmt.Errorf("capacity %v is too great, casting results in integer overflow", roundedUp)
+	}
+	return roundedUpInt, nil
+}
+
+// RoundUpToGBInt rounds up given quantity to chunks of GB. It returns an
+// int instead of an int64 and an error if there's overflow
+func RoundUpToGBInt(size resource.Quantity) (int, error) {
+	requestBytes := size.Value()
+	return RoundUpSizeInt(requestBytes, GB)
+}
+
+// RoundUpToGiBInt rounds up given quantity upto chunks of GiB. It returns an
+// int instead of an int64 and an error if there's overflow
+func RoundUpToGiBInt(size resource.Quantity) (int, error) {
+	requestBytes := size.Value()
+	return RoundUpSizeInt(requestBytes, GIB)
+}
+
 // GenerateVolumeName returns a PV name with clusterName prefix. The function
 // should be used to generate a name of GCE PD or Cinder volume. It basically
 // adds "<clusterName>-dynamic-" before the PV name, making sure the resulting

pkg/volume/vsphere_volume/vsphere_volume_util.go

@@ -93,10 +93,13 @@ func (util *VsphereDiskUtil) CreateVolume(v *vsphereVolumeProvisioner) (volSpec
 	capacity := v.options.PVC.Spec.Resources.Requests[v1.ResourceName(v1.ResourceStorage)]
 	volSizeBytes := capacity.Value()
 	// vSphere works with kilobytes, convert to KiB with rounding up
-	volSizeKB := int(volumeutil.RoundUpSize(volSizeBytes, 1024))
+	volSizeKiB, err := volumeutil.RoundUpSizeInt(volSizeBytes, 1024)
+	if err != nil {
+		return nil, err
+	}
 	name := volumeutil.GenerateVolumeName(v.options.ClusterName, v.options.PVName, 255)
 	volumeOptions := &vclib.VolumeOptions{
-		CapacityKB: volSizeKB,
+		CapacityKB: volSizeKiB,
 		Tags:       *v.options.CloudTags,
 		Name:       name,
 	}
@@ -146,7 +149,7 @@ func (util *VsphereDiskUtil) CreateVolume(v *vsphereVolumeProvisioner) (volSpec
 	}
 	volSpec = &VolumeSpec{
 		Path:              vmDiskPath,
-		Size:              volSizeKB,
+		Size:              volSizeKiB,
 		Fstype:            fstype,
 		StoragePolicyName: volumeOptions.StoragePolicyName,
 		StoragePolicyID:   volumeOptions.StoragePolicyID,