mount.go

/*
Copyright 2014 The Kubernetes Authors.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/

// TODO(thockin): This whole pkg is pretty linux-centric.  As soon as we have
// an alternate platform, we will need to abstract further.
package mount

import (
	"fmt"
	"os"
	"path/filepath"
	"strings"
)

type FileType string

const (
	// Default mount command if mounter path is not specified
	defaultMountCommand           = "mount"
	MountsInGlobalPDPath          = "mounts"
	FileTypeDirectory    FileType = "Directory"
	FileTypeFile         FileType = "File"
	FileTypeSocket       FileType = "Socket"
	FileTypeCharDev      FileType = "CharDevice"
	FileTypeBlockDev     FileType = "BlockDevice"
)

type Interface interface {
	// Mount mounts source to target as fstype with given options.
	Mount(source string, target string, fstype string, options []string) error
	// Unmount unmounts given target.
	Unmount(target string) error
	// List returns a list of all mounted filesystems.  This can be large.
	// On some platforms, reading mounts is not guaranteed consistent (i.e.
	// it could change between chunked reads). This is guaranteed to be
	// consistent.
	List() ([]MountPoint, error)
	// IsMountPointMatch determines if the mountpoint matches the dir
	IsMountPointMatch(mp MountPoint, dir string) bool
	// IsNotMountPoint determines if a directory is a mountpoint.
	// It should return ErrNotExist when the directory does not exist.
	// IsNotMountPoint is more expensive than IsLikelyNotMountPoint.
	// IsNotMountPoint detects bind mounts in linux.
	// IsNotMountPoint enumerates all the mountpoints using List() and
	// the list of mountpoints may be large, then it uses
	// IsMountPointMatch to evaluate whether the directory is a mountpoint
	IsNotMountPoint(file string) (bool, error)
	// IsLikelyNotMountPoint uses heuristics to determine if a directory
	// is a mountpoint.
	// It should return ErrNotExist when the directory does not exist.
	// IsLikelyNotMountPoint does NOT properly detect all mountpoint types
	// most notably linux bind mounts.
	IsLikelyNotMountPoint(file string) (bool, error)
	// DeviceOpened determines if the device is in use elsewhere
	// on the system, i.e. still mounted.
	DeviceOpened(pathname string) (bool, error)
	// PathIsDevice determines if a path is a device.
	PathIsDevice(pathname string) (bool, error)
	// GetDeviceNameFromMount finds the device name by checking the mount path
	// to get the global mount path which matches its plugin directory
	GetDeviceNameFromMount(mountPath, pluginDir string) (string, error)
	// MakeRShared checks that given path is on a mount with 'rshared' mount
	// propagation. If not, it bind-mounts the path as rshared.
	MakeRShared(path string) error
	// GetFileType checks for file/directory/socket/block/character devices.
	// Will operate in the host mount namespace if kubelet is running in a container
	GetFileType(pathname string) (FileType, error)
	// MakeFile creates an empty file.
	// Will operate in the host mount namespace if kubelet is running in a container
	MakeFile(pathname string) error
	// MakeDir creates a new directory.
	// Will operate in the host mount namespace if kubelet is running in a container
	MakeDir(pathname string) error
	// SafeMakeDir creates subdir within given base. It makes sure that the
	// created directory does not escape given base directory mis-using
	// symlinks. Note that the function makes sure that it creates the directory
	// somewhere under the base, nothing else. E.g. if the directory already
	// exists, it may exist outside of the base due to symlinks.
	// This method should be used if the directory to create is inside volume
	// that's under user control. User must not be able to use symlinks to
	// escape the volume to create directories somewhere else.
	SafeMakeDir(subdir string, base string, perm os.FileMode) error
	// Will operate in the host mount namespace if kubelet is running in a container.
	// Error is returned on any other error than "file not found".
	ExistsPath(pathname string) (bool, error)
	// EvalHostSymlinks returns the path name after evaluating symlinks.
	// Will operate in the host mount namespace if kubelet is running in a container.
	EvalHostSymlinks(pathname string) (string, error)
	// CleanSubPaths removes any bind-mounts created by PrepareSafeSubpath in given
	// pod volume directory.
	CleanSubPaths(podDir string, volumeName string) error
	// PrepareSafeSubpath does everything that's necessary to prepare a subPath
	// that's 1) inside given volumePath and 2) immutable after this call.
	//
	// newHostPath - location of prepared subPath. It should be used instead of
	// hostName when running the container.
	// cleanupAction - action to run when the container is running or it failed to start.
	//
	// CleanupAction must be called immediately after the container with given
	// subpath starts. On the other hand, Interface.CleanSubPaths must be called
	// when the pod finishes.
	PrepareSafeSubpath(subPath Subpath) (newHostPath string, cleanupAction func(), err error)
	// GetMountRefs finds all mount references to the path, returns a
	// list of paths. Path could be a mountpoint path, device or a normal
	// directory (for bind mount).
	GetMountRefs(pathname string) ([]string, error)
	// GetFSGroup returns FSGroup of the path.
	GetFSGroup(pathname string) (int64, error)
	// GetSELinuxSupport returns true if given path is on a mount that supports
	// SELinux.
	GetSELinuxSupport(pathname string) (bool, error)
	// GetMode returns permissions of the path.
	GetMode(pathname string) (os.FileMode, error)
}

type Subpath struct {
	// index of the VolumeMount for this container
	VolumeMountIndex int
	// Full path to the subpath directory on the host
	Path string
	// name of the volume that is a valid directory name.
	VolumeName string
	// Full path to the volume path
	VolumePath string
	// Path to the pod's directory, including pod UID
	PodDir string
	// Name of the container
	ContainerName string
}

// Exec executes command where mount utilities are. This can be either the host,
// container where kubelet runs or even a remote pod with mount utilities.
// Usual k8s.io/utils/exec interface is not used because kubelet.RunInContainer does
// not provide stdin/stdout/stderr streams.
type Exec interface {
	// Run executes a command and returns its stdout + stderr combined in one
	// stream.
	Run(cmd string, args ...string) ([]byte, error)
}

// Compile-time check to ensure all Mounter implementations satisfy
// the mount interface
var _ Interface = &Mounter{}

// This represents a single line in /proc/mounts or /etc/fstab.
type MountPoint struct {
	Device string
	Path   string
	Type   string
	Opts   []string
	Freq   int
	Pass   int
}

// SafeFormatAndMount probes a device to see if it is formatted.
// Namely it checks to see if a file system is present. If so it
// mounts it otherwise the device is formatted first then mounted.
type SafeFormatAndMount struct {
	Interface
	Exec
}

// FormatAndMount formats the given disk, if needed, and mounts it.
// That is if the disk is not formatted and it is not being mounted as
// read-only it will format it first then mount it. Otherwise, if the
// disk is already formatted or it is being mounted as read-only, it
// will be mounted without formatting.
func (mounter *SafeFormatAndMount) FormatAndMount(source string, target string, fstype string, options []string) error {
	return mounter.formatAndMount(source, target, fstype, options)
}

// getMountRefsByDev finds all references to the device provided
// by mountPath; returns a list of paths.
// Note that mountPath should be path after the evaluation of any symblolic links.
func getMountRefsByDev(mounter Interface, mountPath string) ([]string, error) {
	mps, err := mounter.List()
	if err != nil {
		return nil, err
	}

	// Finding the device mounted to mountPath
	diskDev := ""
	for i := range mps {
		if mountPath == mps[i].Path {
			diskDev = mps[i].Device
			break
		}
	}

	// Find all references to the device.
	var refs []string
	for i := range mps {
		if mps[i].Device == diskDev || mps[i].Device == mountPath {
			if mps[i].Path != mountPath {
				refs = append(refs, mps[i].Path)
			}
		}
	}
	return refs, nil
}

// GetDeviceNameFromMount: given a mnt point, find the device from /proc/mounts
// returns the device name, reference count, and error code
func GetDeviceNameFromMount(mounter Interface, mountPath string) (string, int, error) {
	mps, err := mounter.List()
	if err != nil {
		return "", 0, err
	}

	// Find the device name.
	// FIXME if multiple devices mounted on the same mount path, only the first one is returned
	device := ""
	// If mountPath is symlink, need get its target path.
	slTarget, err := filepath.EvalSymlinks(mountPath)
	if err != nil {
		slTarget = mountPath
	}
	for i := range mps {
		if mps[i].Path == slTarget {
			device = mps[i].Device
			break
		}
	}

	// Find all references to the device.
	refCount := 0
	for i := range mps {
		if mps[i].Device == device {
			refCount++
		}
	}
	return device, refCount, nil
}

// isNotMountPoint implements Mounter.IsNotMountPoint and is shared by mounter
// implementations.
func isNotMountPoint(mounter Interface, file string) (bool, error) {
	// IsLikelyNotMountPoint provides a quick check
	// to determine whether file IS A mountpoint
	notMnt, notMntErr := mounter.IsLikelyNotMountPoint(file)
	if notMntErr != nil && os.IsPermission(notMntErr) {
		// We were not allowed to do the simple stat() check, e.g. on NFS with
		// root_squash. Fall back to /proc/mounts check below.
		notMnt = true
		notMntErr = nil
	}
	if notMntErr != nil {
		return notMnt, notMntErr
	}
	// identified as mountpoint, so return this fact
	if notMnt == false {
		return notMnt, nil
	}

	// Resolve any symlinks in file, kernel would do the same and use the resolved path in /proc/mounts
	resolvedFile, err := mounter.EvalHostSymlinks(file)
	if err != nil {
		return true, err
	}

	// check all mountpoints since IsLikelyNotMountPoint
	// is not reliable for some mountpoint types
	mountPoints, mountPointsErr := mounter.List()
	if mountPointsErr != nil {
		return notMnt, mountPointsErr
	}
	for _, mp := range mountPoints {
		if mounter.IsMountPointMatch(mp, resolvedFile) {
			notMnt = false
			break
		}
	}
	return notMnt, nil
}

// isBind detects whether a bind mount is being requested and makes the remount options to
// use in case of bind mount, due to the fact that bind mount doesn't respect mount options.
// The list equals:
//   options - 'bind' + 'remount' (no duplicate)
func isBind(options []string) (bool, []string, []string) {
	// Because we have an FD opened on the subpath bind mount, the "bind" option
	// needs to be included, otherwise the mount target will error as busy if you
	// remount as readonly.
	//
	// As a consequence, all read only bind mounts will no longer change the underlying
	// volume mount to be read only.
	bindRemountOpts := []string{"bind", "remount"}
	bind := false
	bindOpts := []string{"bind"}

	// _netdev is a userspace mount option and does not automatically get added when
	// bind mount is created and hence we must carry it over.
	if checkForNetDev(options) {
		bindOpts = append(bindOpts, "_netdev")
	}

	for _, option := range options {
		switch option {
		case "bind":
			bind = true
			break
		case "remount":
			break
		default:
			bindRemountOpts = append(bindRemountOpts, option)
		}
	}

	return bind, bindOpts, bindRemountOpts
}

func checkForNetDev(options []string) bool {
	for _, option := range options {
		if option == "_netdev" {
			return true
		}
	}
	return false
}

// TODO: this is a workaround for the unmount device issue caused by gci mounter.
// In GCI cluster, if gci mounter is used for mounting, the container started by mounter
// script will cause additional mounts created in the container. Since these mounts are
// irrelevant to the original mounts, they should be not considered when checking the
// mount references. Current solution is to filter out those mount paths that contain
// the string of original mount path.
// Plan to work on better approach to solve this issue.

func HasMountRefs(mountPath string, mountRefs []string) bool {
	count := 0
	for _, ref := range mountRefs {
		if !strings.Contains(ref, mountPath) {
			count = count + 1
		}
	}
	return count > 0
}

// PathWithinBase checks if give path is within given base directory.
func PathWithinBase(fullPath, basePath string) bool {
	rel, err := filepath.Rel(basePath, fullPath)
	if err != nil {
		return false
	}
	if startsWithBackstep(rel) {
		// Needed to escape the base path
		return false
	}
	return true
}

// startsWithBackstep checks if the given path starts with a backstep segment
func startsWithBackstep(rel string) bool {
	// normalize to / and check for ../
	return rel == ".." || strings.HasPrefix(filepath.ToSlash(rel), "../")
}

// getFileType checks for file/directory/socket and block/character devices
func getFileType(pathname string) (FileType, error) {
	var pathType FileType
	info, err := os.Stat(pathname)
	if os.IsNotExist(err) {
		return pathType, fmt.Errorf("path %q does not exist", pathname)
	}
	// err in call to os.Stat
	if err != nil {
		return pathType, err
	}

	// checks whether the mode is the target mode
	isSpecificMode := func(mode, targetMode os.FileMode) bool {
		return mode&targetMode == targetMode
	}

	mode := info.Mode()
	if mode.IsDir() {
		return FileTypeDirectory, nil
	} else if mode.IsRegular() {
		return FileTypeFile, nil
	} else if isSpecificMode(mode, os.ModeSocket) {
		return FileTypeSocket, nil
	} else if isSpecificMode(mode, os.ModeDevice) {
		if isSpecificMode(mode, os.ModeCharDevice) {
			return FileTypeCharDev, nil
		}
		return FileTypeBlockDev, nil
	}

	return pathType, fmt.Errorf("only recognise file, directory, socket, block device and character device")
}