PV restore时Disk 创建在错误的zone上

[chrislinan]

What steps did you take and what happened:
[A clear and concise description of what the bug is.]
velero恢复PV的时候发现阿里云操作日志里有attach disk失败的日志，因为备份的PV和velero阿里云的插件创建出来的disk不在一个zone上，例如我备份的PV在cn-shangh-l， 创建出来的disk却在cn-shanghai-g

What did you expect to happen:
创建的disk和原来备份的PV在同一个zone上

Anything else you would like to add:
[Miscellaneous information that will assist in solving the issue.]
在这行代码：https://github.com/AliyunContainerService/velero-plugin/blob/master/velero-plugin-for-alibabacloud/volume_snapshotter.go#L104

func (b *VolumeSnapshotter) CreateVolumeFromSnapshot(snapshotID, volumeType, volumeAZ string, iops *int64) (volumeID string, err error) {
	// describe the snapshot so we can apply its tags to the volume
	snapReq := ecs.CreateDescribeSnapshotsRequest()
	snapReq.SnapshotIds = getJSONArrayString(snapshotID)

	snapRes, err := b.ecs.DescribeSnapshots(snapReq)
	if err != nil {
		return "", errors.WithStack(err)
	}

	if count := len(snapRes.Snapshots.Snapshot); count != 1 {
		return "", errors.Errorf("expected 1 snapshot from DescribeSnapshots for %s, got %v", snapshotID, count)
	}

	tags := getTagsForCluster(snapRes.Snapshots.Snapshot[0].Tags.Tag)

	volumeAZ, err = getMetaData(metadataZoneKey)
	if err != nil {
		return "", errors.Errorf("failed to get zone-id, got %v", err)
	}
.........
}

调用了getMetadata 来读取了velero pod 所在的instance的metadata， 并且用velero pod所在的instance的zone 覆盖了velero传进来的参数：volumeAZ，所以会导致插件创建的disk和velero传的参数不在一个zone上，是否能删除这行代码来修复这个问题？谢谢

Environment:

• Velero version: (use ark version):
• Kubernetes version: (use kubectl version):
• Kubernetes installer & version:
• Cloud provider or hardware configuration:
• OS (e.g. from /etc/os-release):