moby/daemon/volumes.go
Erik Hollensbe 09b700288e Allow /etc/hosts and /etc/resolv.conf to be updated both outside and
inside the container.

Docker-DCO-1.1-Signed-off-by: Erik Hollensbe <github@hollensbe.org> (github: erikh)
2014-09-13 11:27:34 -07:00

330 lines
7.3 KiB
Go

package daemon
import (
"fmt"
"io/ioutil"
"os"
"path/filepath"
"strings"
"syscall"
"github.com/docker/docker/archive"
"github.com/docker/docker/daemon/execdriver"
"github.com/docker/docker/pkg/symlink"
)
type Volume struct {
HostPath string
VolPath string
isReadWrite bool
isBindMount bool
}
func (v *Volume) isDir() (bool, error) {
stat, err := os.Stat(v.HostPath)
if err != nil {
return false, err
}
return stat.IsDir(), nil
}
func prepareVolumesForContainer(container *Container) error {
if container.Volumes == nil || len(container.Volumes) == 0 {
container.Volumes = make(map[string]string)
container.VolumesRW = make(map[string]bool)
if err := applyVolumesFrom(container); err != nil {
return err
}
}
return createVolumes(container)
}
func setupMountsForContainer(container *Container) error {
mounts := []execdriver.Mount{
{
Source: container.ResolvConfPath,
Destination: "/etc/resolv.conf",
Writable: true,
Slave: true,
},
}
if container.HostnamePath != "" {
mounts = append(mounts, execdriver.Mount{
Source: container.HostnamePath,
Destination: "/etc/hostname",
Writable: true,
Private: true,
})
}
if container.HostsPath != "" {
mounts = append(mounts, execdriver.Mount{
Source: container.HostsPath,
Destination: "/etc/hosts",
Writable: true,
Slave: true,
})
}
// Mount user specified volumes
// Note, these are not private because you may want propagation of (un)mounts from host
// volumes. For instance if you use -v /usr:/usr and the host later mounts /usr/share you
// want this new mount in the container
for r, v := range container.Volumes {
mounts = append(mounts, execdriver.Mount{
Source: v,
Destination: r,
Writable: container.VolumesRW[r],
})
}
container.command.Mounts = mounts
return nil
}
func parseVolumesFromSpec(container *Container, spec string) (map[string]*Volume, error) {
specParts := strings.SplitN(spec, ":", 2)
if len(specParts) == 0 {
return nil, fmt.Errorf("Malformed volumes-from specification: %s", spec)
}
c := container.daemon.Get(specParts[0])
if c == nil {
return nil, fmt.Errorf("Container %s not found. Impossible to mount its volumes", specParts[0])
}
volumes, err := c.GetVolumes()
if err != nil {
return nil, err
}
if len(specParts) == 2 {
mode := specParts[1]
if validVolumeMode(mode) {
return nil, fmt.Errorf("Invalid mode for volumes-from: %s", mode)
}
// Set the mode for the inheritted volume
for _, v := range volumes {
v.isReadWrite = mode != "ro"
}
}
return volumes, nil
}
func applyVolumesFrom(container *Container) error {
volumesFrom := container.hostConfig.VolumesFrom
for _, spec := range volumesFrom {
volumes, err := parseVolumesFromSpec(container, spec)
if err != nil {
return err
}
for _, v := range volumes {
if err = v.initialize(container); err != nil {
return err
}
}
}
return nil
}
func validVolumeMode(mode string) bool {
validModes := map[string]bool{
"rw": true,
"ro": true,
}
return validModes[mode]
}
func parseBindVolumeSpec(spec string) (Volume, error) {
var (
arr = strings.Split(spec, ":")
vol Volume
)
switch len(arr) {
case 1:
vol.VolPath = spec
vol.isReadWrite = true
case 2:
vol.HostPath = arr[0]
vol.VolPath = arr[1]
vol.isReadWrite = true
case 3:
vol.HostPath = arr[0]
vol.VolPath = arr[1]
vol.isReadWrite = validVolumeMode(arr[2]) && arr[2] == "rw"
default:
return vol, fmt.Errorf("Invalid volume specification: %s", spec)
}
if !filepath.IsAbs(vol.HostPath) {
return vol, fmt.Errorf("cannot bind mount volume: %s volume paths must be absolute.", vol.HostPath)
}
return vol, nil
}
func createVolumes(container *Container) error {
// Get all the bindmounts
volumes, err := container.GetVolumes()
if err != nil {
return err
}
// Get all the rest of the volumes
for volPath := range container.Config.Volumes {
// Make sure the the volume isn't already specified as a bindmount
if _, exists := volumes[volPath]; !exists {
volumes[volPath] = &Volume{
VolPath: volPath,
isReadWrite: true,
isBindMount: false,
}
}
}
for _, vol := range volumes {
if err = vol.initialize(container); err != nil {
return err
}
}
return nil
}
func createVolumeHostPath(container *Container) (string, error) {
volumesDriver := container.daemon.volumes.Driver()
// Do not pass a container as the parameter for the volume creation.
// The graph driver using the container's information ( Image ) to
// create the parent.
c, err := container.daemon.volumes.Create(nil, "", "", "", "", nil, nil)
if err != nil {
return "", err
}
hostPath, err := volumesDriver.Get(c.ID, "")
if err != nil {
return hostPath, fmt.Errorf("Driver %s failed to get volume rootfs %s: %s", volumesDriver, c.ID, err)
}
return hostPath, nil
}
func (v *Volume) initialize(container *Container) error {
var err error
v.VolPath = filepath.Clean(v.VolPath)
// Do not initialize an existing volume
if _, exists := container.Volumes[v.VolPath]; exists {
return nil
}
// If it's not a bindmount we need to create the dir on the host
if !v.isBindMount && v.HostPath == "" {
v.HostPath, err = createVolumeHostPath(container)
if err != nil {
return err
}
}
hostPath, err := filepath.EvalSymlinks(v.HostPath)
if err != nil {
return err
}
// Create the mountpoint
// This is the path to the volume within the container FS
// This differs from `hostPath` in that `hostPath` refers to the place where
// the volume data is actually stored on the host
fullVolPath, err := symlink.FollowSymlinkInScope(filepath.Join(container.basefs, v.VolPath), container.basefs)
if err != nil {
return err
}
container.Volumes[v.VolPath] = hostPath
container.VolumesRW[v.VolPath] = v.isReadWrite
volIsDir, err := v.isDir()
if err != nil {
return err
}
if err := createIfNotExists(fullVolPath, volIsDir); err != nil {
return err
}
// Do not copy or change permissions if we are mounting from the host
if v.isReadWrite && !v.isBindMount {
return copyExistingContents(fullVolPath, hostPath)
}
return nil
}
func createIfNotExists(destination string, isDir bool) error {
if _, err := os.Stat(destination); err == nil || !os.IsNotExist(err) {
return nil
}
if isDir {
return os.MkdirAll(destination, 0755)
}
if err := os.MkdirAll(filepath.Dir(destination), 0755); err != nil {
return err
}
f, err := os.OpenFile(destination, os.O_CREATE, 0755)
if err != nil {
return err
}
f.Close()
return nil
}
func copyExistingContents(source, destination string) error {
volList, err := ioutil.ReadDir(source)
if err != nil {
return err
}
if len(volList) > 0 {
srcList, err := ioutil.ReadDir(destination)
if err != nil {
return err
}
if len(srcList) == 0 {
// If the source volume is empty copy files from the root into the volume
if err := archive.CopyWithTar(source, destination); err != nil {
return err
}
}
}
return copyOwnership(source, destination)
}
// copyOwnership copies the permissions and uid:gid of the source file
// into the destination file
func copyOwnership(source, destination string) error {
var stat syscall.Stat_t
if err := syscall.Stat(source, &stat); err != nil {
return err
}
if err := os.Chown(destination, int(stat.Uid), int(stat.Gid)); err != nil {
return err
}
return os.Chmod(destination, os.FileMode(stat.Mode))
}