client/mflag: remove use of docker/docker/pkg/homedir

The homedir package was only used to print default values for
flags that contained paths inside the user's home-directory in
a slightly nicer way (replace `/users/home` with `~`).

Given that this is not critical, we can replace this with golang's
function, which does not depend on libcontainer.

There's still one use of the homedir package in docker/docker/opts,
which is used by the dnet binary (but only requires the homedir
package when running in rootless mode)

Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This commit is contained in:
Sebastiaan van Stijn 2020-06-04 12:48:37 +02:00
parent 681196c8f2
commit 6796c2cc9b
93 changed files with 9 additions and 7205 deletions

View file

@ -91,8 +91,6 @@ import (
"strings"
"text/tabwriter"
"time"
"github.com/docker/docker/pkg/homedir"
)
// ErrHelp is the error returned if the flag -help is invoked but no such flag is defined.
@ -538,7 +536,7 @@ func isZeroValue(value string) bool {
// otherwise, the default values of all defined flags in the set.
func (fs *FlagSet) PrintDefaults() {
writer := tabwriter.NewWriter(fs.Out(), 20, 1, 3, ' ', 0)
home := homedir.Get()
home, _ := os.UserHomeDir()
// Don't substitute when HOME is /
if runtime.GOOS != "windows" && home == "/" {
@ -561,7 +559,7 @@ func (fs *FlagSet) PrintDefaults() {
val := flag.DefValue
if home != "" && strings.HasPrefix(val, home) {
val = homedir.GetShortcutString() + val[len(home):]
val = getShortcutString() + val[len(home):]
}
if isZeroValue(val) {
@ -579,6 +577,13 @@ func (fs *FlagSet) PrintDefaults() {
writer.Flush()
}
func getShortcutString() string {
if runtime.GOOS == "windows" {
return "%USERPROFILE%"
}
return "~"
}
// PrintDefaults prints to standard error the default values of all defined command-line flags.
func PrintDefaults() {
CommandLine.PrintDefaults()

View file

@ -14,7 +14,6 @@ github.com/docker/docker dbe4a30928d418e0570891a09703
github.com/docker/distribution 0d3efadf0154c2b8a4e7b6621fff9809655cc580
github.com/docker/go-connections 7395e3f8aa162843a74ed6d48e79627d9792ac55 # v0.4.0
github.com/docker/go-events 9461782956ad83b30282bf90e31fa6a70c255ba9
github.com/docker/go-units 47565b4f722fb6ceae66b95f853feed578a4a51c # v0.3.3
github.com/docker/libkv 458977154600b9f23984d9f4b82e79570b5ae12b
github.com/gogo/protobuf ba06b47c162d49f2af050fb4c75bcbc86a159d5c # v1.2.1
@ -35,7 +34,6 @@ github.com/hashicorp/go-sockaddr c7188e74f6acae5a989bdc959aa7
github.com/hashicorp/serf 598c54895cc5a7b1a24a398d635e8c0ea0959870
github.com/mattn/go-shellwords 02e3cf038dcea8290e44424da473dd12be796a8a # v1.0.3
github.com/miekg/dns 6c0c4e6581f8e173cc562c8b3363ab984e4ae071 # v1.1.27
github.com/opencontainers/runc 2b18fe1d885ee5083ef9f0838fee39b62d653e30
github.com/opencontainers/runtime-spec 29686dbc5559d93fb1ef402eeda3e35c38d75af4 # v1.0.1-59-g29686db
github.com/samuel/go-zookeeper d0e0d8e11f318e000a8cc434616d69e329edc374
github.com/sirupsen/logrus 8bdbc7bcc01dcbb8ec23dc8a28e332258d25251f # v1.4.1

View file

@ -1,109 +0,0 @@
package homedir // import "github.com/docker/docker/pkg/homedir"
import (
"errors"
"os"
"path/filepath"
"strings"
"github.com/docker/docker/pkg/idtools"
)
// GetStatic returns the home directory for the current user without calling
// os/user.Current(). This is useful for static-linked binary on glibc-based
// system, because a call to os/user.Current() in a static binary leads to
// segfault due to a glibc issue that won't be fixed in a short term.
// (#29344, golang/go#13470, https://sourceware.org/bugzilla/show_bug.cgi?id=19341)
func GetStatic() (string, error) {
uid := os.Getuid()
usr, err := idtools.LookupUID(uid)
if err != nil {
return "", err
}
return usr.Home, nil
}
// GetRuntimeDir returns XDG_RUNTIME_DIR.
// XDG_RUNTIME_DIR is typically configured via pam_systemd.
// GetRuntimeDir returns non-nil error if XDG_RUNTIME_DIR is not set.
//
// See also https://standards.freedesktop.org/basedir-spec/latest/ar01s03.html
func GetRuntimeDir() (string, error) {
if xdgRuntimeDir := os.Getenv("XDG_RUNTIME_DIR"); xdgRuntimeDir != "" {
return xdgRuntimeDir, nil
}
return "", errors.New("could not get XDG_RUNTIME_DIR")
}
// StickRuntimeDirContents sets the sticky bit on files that are under
// XDG_RUNTIME_DIR, so that the files won't be periodically removed by the system.
//
// StickyRuntimeDir returns slice of sticked files.
// StickyRuntimeDir returns nil error if XDG_RUNTIME_DIR is not set.
//
// See also https://standards.freedesktop.org/basedir-spec/latest/ar01s03.html
func StickRuntimeDirContents(files []string) ([]string, error) {
runtimeDir, err := GetRuntimeDir()
if err != nil {
// ignore error if runtimeDir is empty
return nil, nil
}
runtimeDir, err = filepath.Abs(runtimeDir)
if err != nil {
return nil, err
}
var sticked []string
for _, f := range files {
f, err = filepath.Abs(f)
if err != nil {
return sticked, err
}
if strings.HasPrefix(f, runtimeDir+"/") {
if err = stick(f); err != nil {
return sticked, err
}
sticked = append(sticked, f)
}
}
return sticked, nil
}
func stick(f string) error {
st, err := os.Stat(f)
if err != nil {
return err
}
m := st.Mode()
m |= os.ModeSticky
return os.Chmod(f, m)
}
// GetDataHome returns XDG_DATA_HOME.
// GetDataHome returns $HOME/.local/share and nil error if XDG_DATA_HOME is not set.
//
// See also https://standards.freedesktop.org/basedir-spec/latest/ar01s03.html
func GetDataHome() (string, error) {
if xdgDataHome := os.Getenv("XDG_DATA_HOME"); xdgDataHome != "" {
return xdgDataHome, nil
}
home := os.Getenv("HOME")
if home == "" {
return "", errors.New("could not get either XDG_DATA_HOME or HOME")
}
return filepath.Join(home, ".local", "share"), nil
}
// GetConfigHome returns XDG_CONFIG_HOME.
// GetConfigHome returns $HOME/.config and nil error if XDG_CONFIG_HOME is not set.
//
// See also https://standards.freedesktop.org/basedir-spec/latest/ar01s03.html
func GetConfigHome() (string, error) {
if xdgConfigHome := os.Getenv("XDG_CONFIG_HOME"); xdgConfigHome != "" {
return xdgConfigHome, nil
}
home := os.Getenv("HOME")
if home == "" {
return "", errors.New("could not get either XDG_CONFIG_HOME or HOME")
}
return filepath.Join(home, ".config"), nil
}

View file

@ -1,33 +0,0 @@
// +build !linux
package homedir // import "github.com/docker/docker/pkg/homedir"
import (
"errors"
)
// GetStatic is not needed for non-linux systems.
// (Precisely, it is needed only for glibc-based linux systems.)
func GetStatic() (string, error) {
return "", errors.New("homedir.GetStatic() is not supported on this system")
}
// GetRuntimeDir is unsupported on non-linux system.
func GetRuntimeDir() (string, error) {
return "", errors.New("homedir.GetRuntimeDir() is not supported on this system")
}
// StickRuntimeDirContents is unsupported on non-linux system.
func StickRuntimeDirContents(files []string) ([]string, error) {
return nil, errors.New("homedir.StickRuntimeDirContents() is not supported on this system")
}
// GetDataHome is unsupported on non-linux system.
func GetDataHome() (string, error) {
return "", errors.New("homedir.GetDataHome() is not supported on this system")
}
// GetConfigHome is unsupported on non-linux system.
func GetConfigHome() (string, error) {
return "", errors.New("homedir.GetConfigHome() is not supported on this system")
}

View file

@ -1,34 +0,0 @@
// +build !windows
package homedir // import "github.com/docker/docker/pkg/homedir"
import (
"os"
"github.com/opencontainers/runc/libcontainer/user"
)
// Key returns the env var name for the user's home dir based on
// the platform being run on
func Key() string {
return "HOME"
}
// Get returns the home directory of the current user with the help of
// environment variables depending on the target operating system.
// Returned path should be used with "path/filepath" to form new paths.
func Get() string {
home := os.Getenv(Key())
if home == "" {
if u, err := user.CurrentUser(); err == nil {
return u.Home
}
}
return home
}
// GetShortcutString returns the string that is shortcut to user's home directory
// in the native shell of the platform running on.
func GetShortcutString() string {
return "~"
}

View file

@ -1,24 +0,0 @@
package homedir // import "github.com/docker/docker/pkg/homedir"
import (
"os"
)
// Key returns the env var name for the user's home dir based on
// the platform being run on
func Key() string {
return "USERPROFILE"
}
// Get returns the home directory of the current user with the help of
// environment variables depending on the target operating system.
// Returned path should be used with "path/filepath" to form new paths.
func Get() string {
return os.Getenv(Key())
}
// GetShortcutString returns the string that is shortcut to user's home directory
// in the native shell of the platform running on.
func GetShortcutString() string {
return "%USERPROFILE%" // be careful while using in format functions
}

View file

@ -1,267 +0,0 @@
package idtools // import "github.com/docker/docker/pkg/idtools"
import (
"bufio"
"fmt"
"os"
"sort"
"strconv"
"strings"
)
// IDMap contains a single entry for user namespace range remapping. An array
// of IDMap entries represents the structure that will be provided to the Linux
// kernel for creating a user namespace.
type IDMap struct {
ContainerID int `json:"container_id"`
HostID int `json:"host_id"`
Size int `json:"size"`
}
type subIDRange struct {
Start int
Length int
}
type ranges []subIDRange
func (e ranges) Len() int { return len(e) }
func (e ranges) Swap(i, j int) { e[i], e[j] = e[j], e[i] }
func (e ranges) Less(i, j int) bool { return e[i].Start < e[j].Start }
const (
subuidFileName = "/etc/subuid"
subgidFileName = "/etc/subgid"
)
// MkdirAllAndChown creates a directory (include any along the path) and then modifies
// ownership to the requested uid/gid. If the directory already exists, this
// function will still change ownership to the requested uid/gid pair.
func MkdirAllAndChown(path string, mode os.FileMode, owner Identity) error {
return mkdirAs(path, mode, owner, true, true)
}
// MkdirAndChown creates a directory and then modifies ownership to the requested uid/gid.
// If the directory already exists, this function still changes ownership.
// Note that unlike os.Mkdir(), this function does not return IsExist error
// in case path already exists.
func MkdirAndChown(path string, mode os.FileMode, owner Identity) error {
return mkdirAs(path, mode, owner, false, true)
}
// MkdirAllAndChownNew creates a directory (include any along the path) and then modifies
// ownership ONLY of newly created directories to the requested uid/gid. If the
// directories along the path exist, no change of ownership will be performed
func MkdirAllAndChownNew(path string, mode os.FileMode, owner Identity) error {
return mkdirAs(path, mode, owner, true, false)
}
// GetRootUIDGID retrieves the remapped root uid/gid pair from the set of maps.
// If the maps are empty, then the root uid/gid will default to "real" 0/0
func GetRootUIDGID(uidMap, gidMap []IDMap) (int, int, error) {
uid, err := toHost(0, uidMap)
if err != nil {
return -1, -1, err
}
gid, err := toHost(0, gidMap)
if err != nil {
return -1, -1, err
}
return uid, gid, nil
}
// toContainer takes an id mapping, and uses it to translate a
// host ID to the remapped ID. If no map is provided, then the translation
// assumes a 1-to-1 mapping and returns the passed in id
func toContainer(hostID int, idMap []IDMap) (int, error) {
if idMap == nil {
return hostID, nil
}
for _, m := range idMap {
if (hostID >= m.HostID) && (hostID <= (m.HostID + m.Size - 1)) {
contID := m.ContainerID + (hostID - m.HostID)
return contID, nil
}
}
return -1, fmt.Errorf("Host ID %d cannot be mapped to a container ID", hostID)
}
// toHost takes an id mapping and a remapped ID, and translates the
// ID to the mapped host ID. If no map is provided, then the translation
// assumes a 1-to-1 mapping and returns the passed in id #
func toHost(contID int, idMap []IDMap) (int, error) {
if idMap == nil {
return contID, nil
}
for _, m := range idMap {
if (contID >= m.ContainerID) && (contID <= (m.ContainerID + m.Size - 1)) {
hostID := m.HostID + (contID - m.ContainerID)
return hostID, nil
}
}
return -1, fmt.Errorf("Container ID %d cannot be mapped to a host ID", contID)
}
// Identity is either a UID and GID pair or a SID (but not both)
type Identity struct {
UID int
GID int
SID string
}
// IdentityMapping contains a mappings of UIDs and GIDs
type IdentityMapping struct {
uids []IDMap
gids []IDMap
}
// NewIdentityMapping takes a requested user and group name and
// using the data from /etc/sub{uid,gid} ranges, creates the
// proper uid and gid remapping ranges for that user/group pair
func NewIdentityMapping(username, groupname string) (*IdentityMapping, error) {
subuidRanges, err := parseSubuid(username)
if err != nil {
return nil, err
}
subgidRanges, err := parseSubgid(groupname)
if err != nil {
return nil, err
}
if len(subuidRanges) == 0 {
return nil, fmt.Errorf("No subuid ranges found for user %q", username)
}
if len(subgidRanges) == 0 {
return nil, fmt.Errorf("No subgid ranges found for group %q", groupname)
}
return &IdentityMapping{
uids: createIDMap(subuidRanges),
gids: createIDMap(subgidRanges),
}, nil
}
// NewIDMappingsFromMaps creates a new mapping from two slices
// Deprecated: this is a temporary shim while transitioning to IDMapping
func NewIDMappingsFromMaps(uids []IDMap, gids []IDMap) *IdentityMapping {
return &IdentityMapping{uids: uids, gids: gids}
}
// RootPair returns a uid and gid pair for the root user. The error is ignored
// because a root user always exists, and the defaults are correct when the uid
// and gid maps are empty.
func (i *IdentityMapping) RootPair() Identity {
uid, gid, _ := GetRootUIDGID(i.uids, i.gids)
return Identity{UID: uid, GID: gid}
}
// ToHost returns the host UID and GID for the container uid, gid.
// Remapping is only performed if the ids aren't already the remapped root ids
func (i *IdentityMapping) ToHost(pair Identity) (Identity, error) {
var err error
target := i.RootPair()
if pair.UID != target.UID {
target.UID, err = toHost(pair.UID, i.uids)
if err != nil {
return target, err
}
}
if pair.GID != target.GID {
target.GID, err = toHost(pair.GID, i.gids)
}
return target, err
}
// ToContainer returns the container UID and GID for the host uid and gid
func (i *IdentityMapping) ToContainer(pair Identity) (int, int, error) {
uid, err := toContainer(pair.UID, i.uids)
if err != nil {
return -1, -1, err
}
gid, err := toContainer(pair.GID, i.gids)
return uid, gid, err
}
// Empty returns true if there are no id mappings
func (i *IdentityMapping) Empty() bool {
return len(i.uids) == 0 && len(i.gids) == 0
}
// UIDs return the UID mapping
// TODO: remove this once everything has been refactored to use pairs
func (i *IdentityMapping) UIDs() []IDMap {
return i.uids
}
// GIDs return the UID mapping
// TODO: remove this once everything has been refactored to use pairs
func (i *IdentityMapping) GIDs() []IDMap {
return i.gids
}
func createIDMap(subidRanges ranges) []IDMap {
idMap := []IDMap{}
// sort the ranges by lowest ID first
sort.Sort(subidRanges)
containerID := 0
for _, idrange := range subidRanges {
idMap = append(idMap, IDMap{
ContainerID: containerID,
HostID: idrange.Start,
Size: idrange.Length,
})
containerID = containerID + idrange.Length
}
return idMap
}
func parseSubuid(username string) (ranges, error) {
return parseSubidFile(subuidFileName, username)
}
func parseSubgid(username string) (ranges, error) {
return parseSubidFile(subgidFileName, username)
}
// parseSubidFile will read the appropriate file (/etc/subuid or /etc/subgid)
// and return all found ranges for a specified username. If the special value
// "ALL" is supplied for username, then all ranges in the file will be returned
func parseSubidFile(path, username string) (ranges, error) {
var rangeList ranges
subidFile, err := os.Open(path)
if err != nil {
return rangeList, err
}
defer subidFile.Close()
s := bufio.NewScanner(subidFile)
for s.Scan() {
if err := s.Err(); err != nil {
return rangeList, err
}
text := strings.TrimSpace(s.Text())
if text == "" || strings.HasPrefix(text, "#") {
continue
}
parts := strings.Split(text, ":")
if len(parts) != 3 {
return rangeList, fmt.Errorf("Cannot parse subuid/gid information: Format not correct for %s file", path)
}
if parts[0] == username || username == "ALL" {
startid, err := strconv.Atoi(parts[1])
if err != nil {
return rangeList, fmt.Errorf("String to int conversion failed during subuid/gid parsing of %s: %v", path, err)
}
length, err := strconv.Atoi(parts[2])
if err != nil {
return rangeList, fmt.Errorf("String to int conversion failed during subuid/gid parsing of %s: %v", path, err)
}
rangeList = append(rangeList, subIDRange{startid, length})
}
}
return rangeList, nil
}

View file

@ -1,231 +0,0 @@
// +build !windows
package idtools // import "github.com/docker/docker/pkg/idtools"
import (
"bytes"
"fmt"
"io"
"os"
"path/filepath"
"strings"
"sync"
"syscall"
"github.com/docker/docker/pkg/system"
"github.com/opencontainers/runc/libcontainer/user"
)
var (
entOnce sync.Once
getentCmd string
)
func mkdirAs(path string, mode os.FileMode, owner Identity, mkAll, chownExisting bool) error {
// make an array containing the original path asked for, plus (for mkAll == true)
// all path components leading up to the complete path that don't exist before we MkdirAll
// so that we can chown all of them properly at the end. If chownExisting is false, we won't
// chown the full directory path if it exists
var paths []string
stat, err := system.Stat(path)
if err == nil {
if !stat.IsDir() {
return &os.PathError{Op: "mkdir", Path: path, Err: syscall.ENOTDIR}
}
if !chownExisting {
return nil
}
// short-circuit--we were called with an existing directory and chown was requested
return lazyChown(path, owner.UID, owner.GID, stat)
}
if os.IsNotExist(err) {
paths = []string{path}
}
if mkAll {
// walk back to "/" looking for directories which do not exist
// and add them to the paths array for chown after creation
dirPath := path
for {
dirPath = filepath.Dir(dirPath)
if dirPath == "/" {
break
}
if _, err := os.Stat(dirPath); err != nil && os.IsNotExist(err) {
paths = append(paths, dirPath)
}
}
if err := system.MkdirAll(path, mode, ""); err != nil {
return err
}
} else {
if err := os.Mkdir(path, mode); err != nil && !os.IsExist(err) {
return err
}
}
// even if it existed, we will chown the requested path + any subpaths that
// didn't exist when we called MkdirAll
for _, pathComponent := range paths {
if err := lazyChown(pathComponent, owner.UID, owner.GID, nil); err != nil {
return err
}
}
return nil
}
// CanAccess takes a valid (existing) directory and a uid, gid pair and determines
// if that uid, gid pair has access (execute bit) to the directory
func CanAccess(path string, pair Identity) bool {
statInfo, err := system.Stat(path)
if err != nil {
return false
}
fileMode := os.FileMode(statInfo.Mode())
permBits := fileMode.Perm()
return accessible(statInfo.UID() == uint32(pair.UID),
statInfo.GID() == uint32(pair.GID), permBits)
}
func accessible(isOwner, isGroup bool, perms os.FileMode) bool {
if isOwner && (perms&0100 == 0100) {
return true
}
if isGroup && (perms&0010 == 0010) {
return true
}
if perms&0001 == 0001 {
return true
}
return false
}
// LookupUser uses traditional local system files lookup (from libcontainer/user) on a username,
// followed by a call to `getent` for supporting host configured non-files passwd and group dbs
func LookupUser(username string) (user.User, error) {
// first try a local system files lookup using existing capabilities
usr, err := user.LookupUser(username)
if err == nil {
return usr, nil
}
// local files lookup failed; attempt to call `getent` to query configured passwd dbs
usr, err = getentUser(fmt.Sprintf("%s %s", "passwd", username))
if err != nil {
return user.User{}, err
}
return usr, nil
}
// LookupUID uses traditional local system files lookup (from libcontainer/user) on a uid,
// followed by a call to `getent` for supporting host configured non-files passwd and group dbs
func LookupUID(uid int) (user.User, error) {
// first try a local system files lookup using existing capabilities
usr, err := user.LookupUid(uid)
if err == nil {
return usr, nil
}
// local files lookup failed; attempt to call `getent` to query configured passwd dbs
return getentUser(fmt.Sprintf("%s %d", "passwd", uid))
}
func getentUser(args string) (user.User, error) {
reader, err := callGetent(args)
if err != nil {
return user.User{}, err
}
users, err := user.ParsePasswd(reader)
if err != nil {
return user.User{}, err
}
if len(users) == 0 {
return user.User{}, fmt.Errorf("getent failed to find passwd entry for %q", strings.Split(args, " ")[1])
}
return users[0], nil
}
// LookupGroup uses traditional local system files lookup (from libcontainer/user) on a group name,
// followed by a call to `getent` for supporting host configured non-files passwd and group dbs
func LookupGroup(groupname string) (user.Group, error) {
// first try a local system files lookup using existing capabilities
group, err := user.LookupGroup(groupname)
if err == nil {
return group, nil
}
// local files lookup failed; attempt to call `getent` to query configured group dbs
return getentGroup(fmt.Sprintf("%s %s", "group", groupname))
}
// LookupGID uses traditional local system files lookup (from libcontainer/user) on a group ID,
// followed by a call to `getent` for supporting host configured non-files passwd and group dbs
func LookupGID(gid int) (user.Group, error) {
// first try a local system files lookup using existing capabilities
group, err := user.LookupGid(gid)
if err == nil {
return group, nil
}
// local files lookup failed; attempt to call `getent` to query configured group dbs
return getentGroup(fmt.Sprintf("%s %d", "group", gid))
}
func getentGroup(args string) (user.Group, error) {
reader, err := callGetent(args)
if err != nil {
return user.Group{}, err
}
groups, err := user.ParseGroup(reader)
if err != nil {
return user.Group{}, err
}
if len(groups) == 0 {
return user.Group{}, fmt.Errorf("getent failed to find groups entry for %q", strings.Split(args, " ")[1])
}
return groups[0], nil
}
func callGetent(args string) (io.Reader, error) {
entOnce.Do(func() { getentCmd, _ = resolveBinary("getent") })
// if no `getent` command on host, can't do anything else
if getentCmd == "" {
return nil, fmt.Errorf("")
}
out, err := execCmd(getentCmd, args)
if err != nil {
exitCode, errC := system.GetExitCode(err)
if errC != nil {
return nil, err
}
switch exitCode {
case 1:
return nil, fmt.Errorf("getent reported invalid parameters/database unknown")
case 2:
terms := strings.Split(args, " ")
return nil, fmt.Errorf("getent unable to find entry %q in %s database", terms[1], terms[0])
case 3:
return nil, fmt.Errorf("getent database doesn't support enumeration")
default:
return nil, err
}
}
return bytes.NewReader(out), nil
}
// lazyChown performs a chown only if the uid/gid don't match what's requested
// Normally a Chown is a no-op if uid/gid match, but in some cases this can still cause an error, e.g. if the
// dir is on an NFS share, so don't call chown unless we absolutely must.
func lazyChown(p string, uid, gid int, stat *system.StatT) error {
if stat == nil {
var err error
stat, err = system.Stat(p)
if err != nil {
return err
}
}
if stat.UID() == uint32(uid) && stat.GID() == uint32(gid) {
return nil
}
return os.Chown(p, uid, gid)
}

View file

@ -1,25 +0,0 @@
package idtools // import "github.com/docker/docker/pkg/idtools"
import (
"os"
"github.com/docker/docker/pkg/system"
)
// This is currently a wrapper around MkdirAll, however, since currently
// permissions aren't set through this path, the identity isn't utilized.
// Ownership is handled elsewhere, but in the future could be support here
// too.
func mkdirAs(path string, mode os.FileMode, owner Identity, mkAll, chownExisting bool) error {
if err := system.MkdirAll(path, mode, ""); err != nil {
return err
}
return nil
}
// CanAccess takes a valid (existing) directory and a uid, gid pair and determines
// if that uid, gid pair has access (execute bit) to the directory
// Windows does not require/support this function, so always return true
func CanAccess(path string, identity Identity) bool {
return true
}

View file

@ -1,164 +0,0 @@
package idtools // import "github.com/docker/docker/pkg/idtools"
import (
"fmt"
"regexp"
"sort"
"strconv"
"strings"
"sync"
)
// add a user and/or group to Linux /etc/passwd, /etc/group using standard
// Linux distribution commands:
// adduser --system --shell /bin/false --disabled-login --disabled-password --no-create-home --group <username>
// useradd -r -s /bin/false <username>
var (
once sync.Once
userCommand string
cmdTemplates = map[string]string{
"adduser": "--system --shell /bin/false --no-create-home --disabled-login --disabled-password --group %s",
"useradd": "-r -s /bin/false %s",
"usermod": "-%s %d-%d %s",
}
idOutRegexp = regexp.MustCompile(`uid=([0-9]+).*gid=([0-9]+)`)
// default length for a UID/GID subordinate range
defaultRangeLen = 65536
defaultRangeStart = 100000
userMod = "usermod"
)
// AddNamespaceRangesUser takes a username and uses the standard system
// utility to create a system user/group pair used to hold the
// /etc/sub{uid,gid} ranges which will be used for user namespace
// mapping ranges in containers.
func AddNamespaceRangesUser(name string) (int, int, error) {
if err := addUser(name); err != nil {
return -1, -1, fmt.Errorf("Error adding user %q: %v", name, err)
}
// Query the system for the created uid and gid pair
out, err := execCmd("id", name)
if err != nil {
return -1, -1, fmt.Errorf("Error trying to find uid/gid for new user %q: %v", name, err)
}
matches := idOutRegexp.FindStringSubmatch(strings.TrimSpace(string(out)))
if len(matches) != 3 {
return -1, -1, fmt.Errorf("Can't find uid, gid from `id` output: %q", string(out))
}
uid, err := strconv.Atoi(matches[1])
if err != nil {
return -1, -1, fmt.Errorf("Can't convert found uid (%s) to int: %v", matches[1], err)
}
gid, err := strconv.Atoi(matches[2])
if err != nil {
return -1, -1, fmt.Errorf("Can't convert found gid (%s) to int: %v", matches[2], err)
}
// Now we need to create the subuid/subgid ranges for our new user/group (system users
// do not get auto-created ranges in subuid/subgid)
if err := createSubordinateRanges(name); err != nil {
return -1, -1, fmt.Errorf("Couldn't create subordinate ID ranges: %v", err)
}
return uid, gid, nil
}
func addUser(userName string) error {
once.Do(func() {
// set up which commands are used for adding users/groups dependent on distro
if _, err := resolveBinary("adduser"); err == nil {
userCommand = "adduser"
} else if _, err := resolveBinary("useradd"); err == nil {
userCommand = "useradd"
}
})
if userCommand == "" {
return fmt.Errorf("Cannot add user; no useradd/adduser binary found")
}
args := fmt.Sprintf(cmdTemplates[userCommand], userName)
out, err := execCmd(userCommand, args)
if err != nil {
return fmt.Errorf("Failed to add user with error: %v; output: %q", err, string(out))
}
return nil
}
func createSubordinateRanges(name string) error {
// first, we should verify that ranges weren't automatically created
// by the distro tooling
ranges, err := parseSubuid(name)
if err != nil {
return fmt.Errorf("Error while looking for subuid ranges for user %q: %v", name, err)
}
if len(ranges) == 0 {
// no UID ranges; let's create one
startID, err := findNextUIDRange()
if err != nil {
return fmt.Errorf("Can't find available subuid range: %v", err)
}
out, err := execCmd(userMod, fmt.Sprintf(cmdTemplates[userMod], "v", startID, startID+defaultRangeLen-1, name))
if err != nil {
return fmt.Errorf("Unable to add subuid range to user: %q; output: %s, err: %v", name, out, err)
}
}
ranges, err = parseSubgid(name)
if err != nil {
return fmt.Errorf("Error while looking for subgid ranges for user %q: %v", name, err)
}
if len(ranges) == 0 {
// no GID ranges; let's create one
startID, err := findNextGIDRange()
if err != nil {
return fmt.Errorf("Can't find available subgid range: %v", err)
}
out, err := execCmd(userMod, fmt.Sprintf(cmdTemplates[userMod], "w", startID, startID+defaultRangeLen-1, name))
if err != nil {
return fmt.Errorf("Unable to add subgid range to user: %q; output: %s, err: %v", name, out, err)
}
}
return nil
}
func findNextUIDRange() (int, error) {
ranges, err := parseSubuid("ALL")
if err != nil {
return -1, fmt.Errorf("Couldn't parse all ranges in /etc/subuid file: %v", err)
}
sort.Sort(ranges)
return findNextRangeStart(ranges)
}
func findNextGIDRange() (int, error) {
ranges, err := parseSubgid("ALL")
if err != nil {
return -1, fmt.Errorf("Couldn't parse all ranges in /etc/subgid file: %v", err)
}
sort.Sort(ranges)
return findNextRangeStart(ranges)
}
func findNextRangeStart(rangeList ranges) (int, error) {
startID := defaultRangeStart
for _, arange := range rangeList {
if wouldOverlap(arange, startID) {
startID = arange.Start + arange.Length
}
}
return startID, nil
}
func wouldOverlap(arange subIDRange, ID int) bool {
low := ID
high := ID + defaultRangeLen
if (low >= arange.Start && low <= arange.Start+arange.Length) ||
(high <= arange.Start+arange.Length && high >= arange.Start) {
return true
}
return false
}

View file

@ -1,12 +0,0 @@
// +build !linux
package idtools // import "github.com/docker/docker/pkg/idtools"
import "fmt"
// AddNamespaceRangesUser takes a name and finds an unused uid, gid pair
// and calls the appropriate helper function to add the group and then
// the user to the group in /etc/group and /etc/passwd respectively.
func AddNamespaceRangesUser(name string) (int, int, error) {
return -1, -1, fmt.Errorf("No support for adding users or groups on this OS")
}

View file

@ -1,32 +0,0 @@
// +build !windows
package idtools // import "github.com/docker/docker/pkg/idtools"
import (
"fmt"
"os/exec"
"path/filepath"
"strings"
)
func resolveBinary(binname string) (string, error) {
binaryPath, err := exec.LookPath(binname)
if err != nil {
return "", err
}
resolvedPath, err := filepath.EvalSymlinks(binaryPath)
if err != nil {
return "", err
}
//only return no error if the final resolved binary basename
//matches what was searched for
if filepath.Base(resolvedPath) == binname {
return resolvedPath, nil
}
return "", fmt.Errorf("Binary %q does not resolve to a binary of that name in $PATH (%q)", binname, resolvedPath)
}
func execCmd(cmd, args string) ([]byte, error) {
execCmd := exec.Command(cmd, strings.Split(args, " ")...)
return execCmd.CombinedOutput()
}

View file

@ -1,137 +0,0 @@
package mount // import "github.com/docker/docker/pkg/mount"
import (
"fmt"
"strings"
)
var flags = map[string]struct {
clear bool
flag int
}{
"defaults": {false, 0},
"ro": {false, RDONLY},
"rw": {true, RDONLY},
"suid": {true, NOSUID},
"nosuid": {false, NOSUID},
"dev": {true, NODEV},
"nodev": {false, NODEV},
"exec": {true, NOEXEC},
"noexec": {false, NOEXEC},
"sync": {false, SYNCHRONOUS},
"async": {true, SYNCHRONOUS},
"dirsync": {false, DIRSYNC},
"remount": {false, REMOUNT},
"mand": {false, MANDLOCK},
"nomand": {true, MANDLOCK},
"atime": {true, NOATIME},
"noatime": {false, NOATIME},
"diratime": {true, NODIRATIME},
"nodiratime": {false, NODIRATIME},
"bind": {false, BIND},
"rbind": {false, RBIND},
"unbindable": {false, UNBINDABLE},
"runbindable": {false, RUNBINDABLE},
"private": {false, PRIVATE},
"rprivate": {false, RPRIVATE},
"shared": {false, SHARED},
"rshared": {false, RSHARED},
"slave": {false, SLAVE},
"rslave": {false, RSLAVE},
"relatime": {false, RELATIME},
"norelatime": {true, RELATIME},
"strictatime": {false, STRICTATIME},
"nostrictatime": {true, STRICTATIME},
}
var validFlags = map[string]bool{
"": true,
"size": true,
"mode": true,
"uid": true,
"gid": true,
"nr_inodes": true,
"nr_blocks": true,
"mpol": true,
}
var propagationFlags = map[string]bool{
"bind": true,
"rbind": true,
"unbindable": true,
"runbindable": true,
"private": true,
"rprivate": true,
"shared": true,
"rshared": true,
"slave": true,
"rslave": true,
}
// MergeTmpfsOptions merge mount options to make sure there is no duplicate.
func MergeTmpfsOptions(options []string) ([]string, error) {
// We use collisions maps to remove duplicates.
// For flag, the key is the flag value (the key for propagation flag is -1)
// For data=value, the key is the data
flagCollisions := map[int]bool{}
dataCollisions := map[string]bool{}
var newOptions []string
// We process in reverse order
for i := len(options) - 1; i >= 0; i-- {
option := options[i]
if option == "defaults" {
continue
}
if f, ok := flags[option]; ok && f.flag != 0 {
// There is only one propagation mode
key := f.flag
if propagationFlags[option] {
key = -1
}
// Check to see if there is collision for flag
if !flagCollisions[key] {
// We prepend the option and add to collision map
newOptions = append([]string{option}, newOptions...)
flagCollisions[key] = true
}
continue
}
opt := strings.SplitN(option, "=", 2)
if len(opt) != 2 || !validFlags[opt[0]] {
return nil, fmt.Errorf("Invalid tmpfs option %q", opt)
}
if !dataCollisions[opt[0]] {
// We prepend the option and add to collision map
newOptions = append([]string{option}, newOptions...)
dataCollisions[opt[0]] = true
}
}
return newOptions, nil
}
// Parse fstab type mount options into mount() flags
// and device specific data
func parseOptions(options string) (int, string) {
var (
flag int
data []string
)
for _, o := range strings.Split(options, ",") {
// If the option does not exist in the flags table or the flag
// is not supported on the platform,
// then it is a data value for a specific fs type
if f, exists := flags[o]; exists && f.flag != 0 {
if f.clear {
flag &= ^f.flag
} else {
flag |= f.flag
}
} else {
data = append(data, o)
}
}
return flag, strings.Join(data, ",")
}

View file

@ -1,49 +0,0 @@
// +build freebsd,cgo
package mount // import "github.com/docker/docker/pkg/mount"
/*
#include <sys/mount.h>
*/
import "C"
const (
// RDONLY will mount the filesystem as read-only.
RDONLY = C.MNT_RDONLY
// NOSUID will not allow set-user-identifier or set-group-identifier bits to
// take effect.
NOSUID = C.MNT_NOSUID
// NOEXEC will not allow execution of any binaries on the mounted file system.
NOEXEC = C.MNT_NOEXEC
// SYNCHRONOUS will allow any I/O to the file system to be done synchronously.
SYNCHRONOUS = C.MNT_SYNCHRONOUS
// NOATIME will not update the file access time when reading from a file.
NOATIME = C.MNT_NOATIME
)
// These flags are unsupported.
const (
BIND = 0
DIRSYNC = 0
MANDLOCK = 0
NODEV = 0
NODIRATIME = 0
UNBINDABLE = 0
RUNBINDABLE = 0
PRIVATE = 0
RPRIVATE = 0
SHARED = 0
RSHARED = 0
SLAVE = 0
RSLAVE = 0
RBIND = 0
RELATIVE = 0
RELATIME = 0
REMOUNT = 0
STRICTATIME = 0
mntDetach = 0
)

View file

@ -1,87 +0,0 @@
package mount // import "github.com/docker/docker/pkg/mount"
import (
"golang.org/x/sys/unix"
)
const (
// RDONLY will mount the file system read-only.
RDONLY = unix.MS_RDONLY
// NOSUID will not allow set-user-identifier or set-group-identifier bits to
// take effect.
NOSUID = unix.MS_NOSUID
// NODEV will not interpret character or block special devices on the file
// system.
NODEV = unix.MS_NODEV
// NOEXEC will not allow execution of any binaries on the mounted file system.
NOEXEC = unix.MS_NOEXEC
// SYNCHRONOUS will allow I/O to the file system to be done synchronously.
SYNCHRONOUS = unix.MS_SYNCHRONOUS
// DIRSYNC will force all directory updates within the file system to be done
// synchronously. This affects the following system calls: create, link,
// unlink, symlink, mkdir, rmdir, mknod and rename.
DIRSYNC = unix.MS_DIRSYNC
// REMOUNT will attempt to remount an already-mounted file system. This is
// commonly used to change the mount flags for a file system, especially to
// make a readonly file system writeable. It does not change device or mount
// point.
REMOUNT = unix.MS_REMOUNT
// MANDLOCK will force mandatory locks on a filesystem.
MANDLOCK = unix.MS_MANDLOCK
// NOATIME will not update the file access time when reading from a file.
NOATIME = unix.MS_NOATIME
// NODIRATIME will not update the directory access time.
NODIRATIME = unix.MS_NODIRATIME
// BIND remounts a subtree somewhere else.
BIND = unix.MS_BIND
// RBIND remounts a subtree and all possible submounts somewhere else.
RBIND = unix.MS_BIND | unix.MS_REC
// UNBINDABLE creates a mount which cannot be cloned through a bind operation.
UNBINDABLE = unix.MS_UNBINDABLE
// RUNBINDABLE marks the entire mount tree as UNBINDABLE.
RUNBINDABLE = unix.MS_UNBINDABLE | unix.MS_REC
// PRIVATE creates a mount which carries no propagation abilities.
PRIVATE = unix.MS_PRIVATE
// RPRIVATE marks the entire mount tree as PRIVATE.
RPRIVATE = unix.MS_PRIVATE | unix.MS_REC
// SLAVE creates a mount which receives propagation from its master, but not
// vice versa.
SLAVE = unix.MS_SLAVE
// RSLAVE marks the entire mount tree as SLAVE.
RSLAVE = unix.MS_SLAVE | unix.MS_REC
// SHARED creates a mount which provides the ability to create mirrors of
// that mount such that mounts and unmounts within any of the mirrors
// propagate to the other mirrors.
SHARED = unix.MS_SHARED
// RSHARED marks the entire mount tree as SHARED.
RSHARED = unix.MS_SHARED | unix.MS_REC
// RELATIME updates inode access times relative to modify or change time.
RELATIME = unix.MS_RELATIME
// STRICTATIME allows to explicitly request full atime updates. This makes
// it possible for the kernel to default to relatime or noatime but still
// allow userspace to override it.
STRICTATIME = unix.MS_STRICTATIME
mntDetach = unix.MNT_DETACH
)

View file

@ -1,31 +0,0 @@
// +build !linux,!freebsd freebsd,!cgo
package mount // import "github.com/docker/docker/pkg/mount"
// These flags are unsupported.
const (
BIND = 0
DIRSYNC = 0
MANDLOCK = 0
NOATIME = 0
NODEV = 0
NODIRATIME = 0
NOEXEC = 0
NOSUID = 0
UNBINDABLE = 0
RUNBINDABLE = 0
PRIVATE = 0
RPRIVATE = 0
SHARED = 0
RSHARED = 0
SLAVE = 0
RSLAVE = 0
RBIND = 0
RELATIME = 0
RELATIVE = 0
REMOUNT = 0
STRICTATIME = 0
SYNCHRONOUS = 0
RDONLY = 0
mntDetach = 0
)

View file

@ -1,159 +0,0 @@
package mount // import "github.com/docker/docker/pkg/mount"
import (
"sort"
"strconv"
"strings"
"github.com/sirupsen/logrus"
)
// mountError records an error from mount or unmount operation
type mountError struct {
op string
source, target string
flags uintptr
data string
err error
}
func (e *mountError) Error() string {
out := e.op + " "
if e.source != "" {
out += e.source + ":" + e.target
} else {
out += e.target
}
if e.flags != uintptr(0) {
out += ", flags: 0x" + strconv.FormatUint(uint64(e.flags), 16)
}
if e.data != "" {
out += ", data: " + e.data
}
out += ": " + e.err.Error()
return out
}
// Cause returns the underlying cause of the error
func (e *mountError) Cause() error {
return e.err
}
// FilterFunc is a type defining a callback function
// to filter out unwanted entries. It takes a pointer
// to an Info struct (not fully populated, currently
// only Mountpoint is filled in), and returns two booleans:
// - skip: true if the entry should be skipped
// - stop: true if parsing should be stopped after the entry
type FilterFunc func(*Info) (skip, stop bool)
// PrefixFilter discards all entries whose mount points
// do not start with a prefix specified
func PrefixFilter(prefix string) FilterFunc {
return func(m *Info) (bool, bool) {
skip := !strings.HasPrefix(m.Mountpoint, prefix)
return skip, false
}
}
// SingleEntryFilter looks for a specific entry
func SingleEntryFilter(mp string) FilterFunc {
return func(m *Info) (bool, bool) {
if m.Mountpoint == mp {
return false, true // don't skip, stop now
}
return true, false // skip, keep going
}
}
// ParentsFilter returns all entries whose mount points
// can be parents of a path specified, discarding others.
// For example, given `/var/lib/docker/something`, entries
// like `/var/lib/docker`, `/var` and `/` are returned.
func ParentsFilter(path string) FilterFunc {
return func(m *Info) (bool, bool) {
skip := !strings.HasPrefix(path, m.Mountpoint)
return skip, false
}
}
// GetMounts retrieves a list of mounts for the current running process,
// with an optional filter applied (use nil for no filter).
func GetMounts(f FilterFunc) ([]*Info, error) {
return parseMountTable(f)
}
// Mounted determines if a specified mountpoint has been mounted.
// On Linux it looks at /proc/self/mountinfo.
func Mounted(mountpoint string) (bool, error) {
entries, err := GetMounts(SingleEntryFilter(mountpoint))
if err != nil {
return false, err
}
return len(entries) > 0, nil
}
// Mount will mount filesystem according to the specified configuration, on the
// condition that the target path is *not* already mounted. Options must be
// specified like the mount or fstab unix commands: "opt1=val1,opt2=val2". See
// flags.go for supported option flags.
func Mount(device, target, mType, options string) error {
flag, _ := parseOptions(options)
if flag&REMOUNT != REMOUNT {
if mounted, err := Mounted(target); err != nil || mounted {
return err
}
}
return ForceMount(device, target, mType, options)
}
// ForceMount will mount a filesystem according to the specified configuration,
// *regardless* if the target path is not already mounted. Options must be
// specified like the mount or fstab unix commands: "opt1=val1,opt2=val2". See
// flags.go for supported option flags.
func ForceMount(device, target, mType, options string) error {
flag, data := parseOptions(options)
return mount(device, target, mType, uintptr(flag), data)
}
// Unmount lazily unmounts a filesystem on supported platforms, otherwise
// does a normal unmount.
func Unmount(target string) error {
return unmount(target, mntDetach)
}
// RecursiveUnmount unmounts the target and all mounts underneath, starting with
// the deepsest mount first.
func RecursiveUnmount(target string) error {
mounts, err := parseMountTable(PrefixFilter(target))
if err != nil {
return err
}
// Make the deepest mount be first
sort.Slice(mounts, func(i, j int) bool {
return len(mounts[i].Mountpoint) > len(mounts[j].Mountpoint)
})
for i, m := range mounts {
logrus.Debugf("Trying to unmount %s", m.Mountpoint)
err = unmount(m.Mountpoint, mntDetach)
if err != nil {
if i == len(mounts)-1 { // last mount
if mounted, e := Mounted(m.Mountpoint); e != nil || mounted {
return err
}
} else {
// This is some submount, we can ignore this error for now, the final unmount will fail if this is a real problem
logrus.WithError(err).Warnf("Failed to unmount submount %s", m.Mountpoint)
}
}
logrus.Debugf("Unmounted %s", m.Mountpoint)
}
return nil
}

View file

@ -1,59 +0,0 @@
package mount // import "github.com/docker/docker/pkg/mount"
/*
#include <errno.h>
#include <stdlib.h>
#include <string.h>
#include <sys/_iovec.h>
#include <sys/mount.h>
#include <sys/param.h>
*/
import "C"
import (
"strings"
"syscall"
"unsafe"
)
func allocateIOVecs(options []string) []C.struct_iovec {
out := make([]C.struct_iovec, len(options))
for i, option := range options {
out[i].iov_base = unsafe.Pointer(C.CString(option))
out[i].iov_len = C.size_t(len(option) + 1)
}
return out
}
func mount(device, target, mType string, flag uintptr, data string) error {
isNullFS := false
xs := strings.Split(data, ",")
for _, x := range xs {
if x == "bind" {
isNullFS = true
}
}
options := []string{"fspath", target}
if isNullFS {
options = append(options, "fstype", "nullfs", "target", device)
} else {
options = append(options, "fstype", mType, "from", device)
}
rawOptions := allocateIOVecs(options)
for _, rawOption := range rawOptions {
defer C.free(rawOption.iov_base)
}
if errno := C.nmount(&rawOptions[0], C.uint(len(options)), C.int(flag)); errno != 0 {
return &mountError{
op: "mount",
source: device,
target: target,
flags: flag,
err: syscall.Errno(errno),
}
}
return nil
}

View file

@ -1,73 +0,0 @@
package mount // import "github.com/docker/docker/pkg/mount"
import (
"golang.org/x/sys/unix"
)
const (
// ptypes is the set propagation types.
ptypes = unix.MS_SHARED | unix.MS_PRIVATE | unix.MS_SLAVE | unix.MS_UNBINDABLE
// pflags is the full set valid flags for a change propagation call.
pflags = ptypes | unix.MS_REC | unix.MS_SILENT
// broflags is the combination of bind and read only
broflags = unix.MS_BIND | unix.MS_RDONLY
)
// isremount returns true if either device name or flags identify a remount request, false otherwise.
func isremount(device string, flags uintptr) bool {
switch {
// We treat device "" and "none" as a remount request to provide compatibility with
// requests that don't explicitly set MS_REMOUNT such as those manipulating bind mounts.
case flags&unix.MS_REMOUNT != 0, device == "", device == "none":
return true
default:
return false
}
}
func mount(device, target, mType string, flags uintptr, data string) error {
oflags := flags &^ ptypes
if !isremount(device, flags) || data != "" {
// Initial call applying all non-propagation flags for mount
// or remount with changed data
if err := unix.Mount(device, target, mType, oflags, data); err != nil {
return &mountError{
op: "mount",
source: device,
target: target,
flags: oflags,
data: data,
err: err,
}
}
}
if flags&ptypes != 0 {
// Change the propagation type.
if err := unix.Mount("", target, "", flags&pflags, ""); err != nil {
return &mountError{
op: "remount",
target: target,
flags: flags & pflags,
err: err,
}
}
}
if oflags&broflags == broflags {
// Remount the bind to apply read only.
if err := unix.Mount("", target, "", oflags|unix.MS_REMOUNT, ""); err != nil {
return &mountError{
op: "remount-ro",
target: target,
flags: oflags | unix.MS_REMOUNT,
err: err,
}
}
}
return nil
}

View file

@ -1,7 +0,0 @@
// +build !linux,!freebsd freebsd,!cgo
package mount // import "github.com/docker/docker/pkg/mount"
func mount(device, target, mType string, flag uintptr, data string) error {
panic("Not implemented")
}

View file

@ -1,40 +0,0 @@
package mount // import "github.com/docker/docker/pkg/mount"
// Info reveals information about a particular mounted filesystem. This
// struct is populated from the content in the /proc/<pid>/mountinfo file.
type Info struct {
// ID is a unique identifier of the mount (may be reused after umount).
ID int
// Parent indicates the ID of the mount parent (or of self for the top of the
// mount tree).
Parent int
// Major indicates one half of the device ID which identifies the device class.
Major int
// Minor indicates one half of the device ID which identifies a specific
// instance of device.
Minor int
// Root of the mount within the filesystem.
Root string
// Mountpoint indicates the mount point relative to the process's root.
Mountpoint string
// Opts represents mount-specific options.
Opts string
// Optional represents optional fields.
Optional string
// Fstype indicates the type of filesystem, such as EXT3.
Fstype string
// Source indicates filesystem specific information or "none".
Source string
// VfsOpts represents per super block options.
VfsOpts string
}

View file

@ -1,55 +0,0 @@
package mount // import "github.com/docker/docker/pkg/mount"
/*
#include <sys/param.h>
#include <sys/ucred.h>
#include <sys/mount.h>
*/
import "C"
import (
"fmt"
"reflect"
"unsafe"
)
// Parse /proc/self/mountinfo because comparing Dev and ino does not work from
// bind mounts.
func parseMountTable(filter FilterFunc) ([]*Info, error) {
var rawEntries *C.struct_statfs
count := int(C.getmntinfo(&rawEntries, C.MNT_WAIT))
if count == 0 {
return nil, fmt.Errorf("Failed to call getmntinfo")
}
var entries []C.struct_statfs
header := (*reflect.SliceHeader)(unsafe.Pointer(&entries))
header.Cap = count
header.Len = count
header.Data = uintptr(unsafe.Pointer(rawEntries))
var out []*Info
for _, entry := range entries {
var mountinfo Info
var skip, stop bool
mountinfo.Mountpoint = C.GoString(&entry.f_mntonname[0])
if filter != nil {
// filter out entries we're not interested in
skip, stop = filter(p)
if skip {
continue
}
}
mountinfo.Source = C.GoString(&entry.f_mntfromname[0])
mountinfo.Fstype = C.GoString(&entry.f_fstypename[0])
out = append(out, &mountinfo)
if stop {
break
}
}
return out, nil
}

View file

@ -1,144 +0,0 @@
package mount // import "github.com/docker/docker/pkg/mount"
import (
"bufio"
"fmt"
"io"
"os"
"strconv"
"strings"
"github.com/pkg/errors"
)
func parseInfoFile(r io.Reader, filter FilterFunc) ([]*Info, error) {
s := bufio.NewScanner(r)
out := []*Info{}
var err error
for s.Scan() {
if err = s.Err(); err != nil {
return nil, err
}
/*
See http://man7.org/linux/man-pages/man5/proc.5.html
36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
(1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
(1) mount ID: unique identifier of the mount (may be reused after umount)
(2) parent ID: ID of parent (or of self for the top of the mount tree)
(3) major:minor: value of st_dev for files on filesystem
(4) root: root of the mount within the filesystem
(5) mount point: mount point relative to the process's root
(6) mount options: per mount options
(7) optional fields: zero or more fields of the form "tag[:value]"
(8) separator: marks the end of the optional fields
(9) filesystem type: name of filesystem of the form "type[.subtype]"
(10) mount source: filesystem specific information or "none"
(11) super options: per super block options
*/
text := s.Text()
fields := strings.Split(text, " ")
numFields := len(fields)
if numFields < 10 {
// should be at least 10 fields
return nil, fmt.Errorf("Parsing '%s' failed: not enough fields (%d)", text, numFields)
}
p := &Info{}
// ignore any numbers parsing errors, as there should not be any
p.ID, _ = strconv.Atoi(fields[0])
p.Parent, _ = strconv.Atoi(fields[1])
mm := strings.Split(fields[2], ":")
if len(mm) != 2 {
return nil, fmt.Errorf("Parsing '%s' failed: unexpected minor:major pair %s", text, mm)
}
p.Major, _ = strconv.Atoi(mm[0])
p.Minor, _ = strconv.Atoi(mm[1])
p.Root, err = strconv.Unquote(`"` + fields[3] + `"`)
if err != nil {
return nil, errors.Wrapf(err, "Parsing '%s' failed: unable to unquote root field", fields[3])
}
p.Mountpoint, err = strconv.Unquote(`"` + fields[4] + `"`)
if err != nil {
return nil, errors.Wrapf(err, "Parsing '%s' failed: unable to unquote mount point field", fields[4])
}
p.Opts = fields[5]
var skip, stop bool
if filter != nil {
// filter out entries we're not interested in
skip, stop = filter(p)
if skip {
continue
}
}
// one or more optional fields, when a separator (-)
i := 6
for ; i < numFields && fields[i] != "-"; i++ {
switch i {
case 6:
p.Optional = fields[6]
default:
/* NOTE there might be more optional fields before the such as
fields[7]...fields[N] (where N < sepIndex), although
as of Linux kernel 4.15 the only known ones are
mount propagation flags in fields[6]. The correct
behavior is to ignore any unknown optional fields.
*/
break
}
}
if i == numFields {
return nil, fmt.Errorf("Parsing '%s' failed: missing separator ('-')", text)
}
// There should be 3 fields after the separator...
if i+4 > numFields {
return nil, fmt.Errorf("Parsing '%s' failed: not enough fields after a separator", text)
}
// ... but in Linux <= 3.9 mounting a cifs with spaces in a share name
// (like "//serv/My Documents") _may_ end up having a space in the last field
// of mountinfo (like "unc=//serv/My Documents"). Since kernel 3.10-rc1, cifs
// option unc= is ignored, so a space should not appear. In here we ignore
// those "extra" fields caused by extra spaces.
p.Fstype = fields[i+1]
p.Source = fields[i+2]
p.VfsOpts = fields[i+3]
out = append(out, p)
if stop {
break
}
}
return out, nil
}
// Parse /proc/self/mountinfo because comparing Dev and ino does not work from
// bind mounts
func parseMountTable(filter FilterFunc) ([]*Info, error) {
f, err := os.Open("/proc/self/mountinfo")
if err != nil {
return nil, err
}
defer f.Close()
return parseInfoFile(f, filter)
}
// PidMountInfo collects the mounts for a specific process ID. If the process
// ID is unknown, it is better to use `GetMounts` which will inspect
// "/proc/self/mountinfo" instead.
func PidMountInfo(pid int) ([]*Info, error) {
f, err := os.Open(fmt.Sprintf("/proc/%d/mountinfo", pid))
if err != nil {
return nil, err
}
defer f.Close()
return parseInfoFile(f, nil)
}

View file

@ -1,12 +0,0 @@
// +build !windows,!linux,!freebsd freebsd,!cgo
package mount // import "github.com/docker/docker/pkg/mount"
import (
"fmt"
"runtime"
)
func parseMountTable(f FilterFunc) ([]*Info, error) {
return nil, fmt.Errorf("mount.parseMountTable is not implemented on %s/%s", runtime.GOOS, runtime.GOARCH)
}

View file

@ -1,6 +0,0 @@
package mount // import "github.com/docker/docker/pkg/mount"
func parseMountTable(f FilterFunc) ([]*Info, error) {
// Do NOT return an error!
return nil, nil
}

View file

@ -1,71 +0,0 @@
package mount // import "github.com/docker/docker/pkg/mount"
// MakeShared ensures a mounted filesystem has the SHARED mount option enabled.
// See the supported options in flags.go for further reference.
func MakeShared(mountPoint string) error {
return ensureMountedAs(mountPoint, "shared")
}
// MakeRShared ensures a mounted filesystem has the RSHARED mount option enabled.
// See the supported options in flags.go for further reference.
func MakeRShared(mountPoint string) error {
return ensureMountedAs(mountPoint, "rshared")
}
// MakePrivate ensures a mounted filesystem has the PRIVATE mount option enabled.
// See the supported options in flags.go for further reference.
func MakePrivate(mountPoint string) error {
return ensureMountedAs(mountPoint, "private")
}
// MakeRPrivate ensures a mounted filesystem has the RPRIVATE mount option
// enabled. See the supported options in flags.go for further reference.
func MakeRPrivate(mountPoint string) error {
return ensureMountedAs(mountPoint, "rprivate")
}
// MakeSlave ensures a mounted filesystem has the SLAVE mount option enabled.
// See the supported options in flags.go for further reference.
func MakeSlave(mountPoint string) error {
return ensureMountedAs(mountPoint, "slave")
}
// MakeRSlave ensures a mounted filesystem has the RSLAVE mount option enabled.
// See the supported options in flags.go for further reference.
func MakeRSlave(mountPoint string) error {
return ensureMountedAs(mountPoint, "rslave")
}
// MakeUnbindable ensures a mounted filesystem has the UNBINDABLE mount option
// enabled. See the supported options in flags.go for further reference.
func MakeUnbindable(mountPoint string) error {
return ensureMountedAs(mountPoint, "unbindable")
}
// MakeRUnbindable ensures a mounted filesystem has the RUNBINDABLE mount
// option enabled. See the supported options in flags.go for further reference.
func MakeRUnbindable(mountPoint string) error {
return ensureMountedAs(mountPoint, "runbindable")
}
// MakeMount ensures that the file or directory given is a mount point,
// bind mounting it to itself it case it is not.
func MakeMount(mnt string) error {
mounted, err := Mounted(mnt)
if err != nil {
return err
}
if mounted {
return nil
}
return Mount(mnt, mnt, "none", "bind")
}
func ensureMountedAs(mountPoint, options string) error {
if err := MakeMount(mountPoint); err != nil {
return err
}
return ForceMount("", mountPoint, "none", options)
}

View file

@ -1,22 +0,0 @@
// +build !windows
package mount // import "github.com/docker/docker/pkg/mount"
import "golang.org/x/sys/unix"
func unmount(target string, flags int) error {
err := unix.Unmount(target, flags)
if err == nil || err == unix.EINVAL {
// Ignore "not mounted" error here. Note the same error
// can be returned if flags are invalid, so this code
// assumes that the flags value is always correct.
return nil
}
return &mountError{
op: "umount",
target: target,
flags: uintptr(flags),
err: err,
}
}

View file

@ -1,7 +0,0 @@
// +build windows
package mount // import "github.com/docker/docker/pkg/mount"
func unmount(target string, flag int) error {
panic("Not implemented")
}

View file

@ -1,16 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"strings"
"golang.org/x/sys/windows"
)
// EscapeArgs makes a Windows-style escaped command line from a set of arguments
func EscapeArgs(args []string) string {
escapedArgs := make([]string, len(args))
for i, a := range args {
escapedArgs[i] = windows.EscapeArg(a)
}
return strings.Join(escapedArgs, " ")
}

View file

@ -1,31 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"os"
"time"
)
// Chtimes changes the access time and modified time of a file at the given path
func Chtimes(name string, atime time.Time, mtime time.Time) error {
unixMinTime := time.Unix(0, 0)
unixMaxTime := maxTime
// If the modified time is prior to the Unix Epoch, or after the
// end of Unix Time, os.Chtimes has undefined behavior
// default to Unix Epoch in this case, just in case
if atime.Before(unixMinTime) || atime.After(unixMaxTime) {
atime = unixMinTime
}
if mtime.Before(unixMinTime) || mtime.After(unixMaxTime) {
mtime = unixMinTime
}
if err := os.Chtimes(name, atime, mtime); err != nil {
return err
}
// Take platform specific action for setting create time.
return setCTime(name, mtime)
}

View file

@ -1,14 +0,0 @@
// +build !windows
package system // import "github.com/docker/docker/pkg/system"
import (
"time"
)
//setCTime will set the create time on a file. On Unix, the create
//time is updated as a side effect of setting the modified time, so
//no action is required.
func setCTime(path string, ctime time.Time) error {
return nil
}

View file

@ -1,26 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"time"
"golang.org/x/sys/windows"
)
//setCTime will set the create time on a file. On Windows, this requires
//calling SetFileTime and explicitly including the create time.
func setCTime(path string, ctime time.Time) error {
ctimespec := windows.NsecToTimespec(ctime.UnixNano())
pathp, e := windows.UTF16PtrFromString(path)
if e != nil {
return e
}
h, e := windows.CreateFile(pathp,
windows.FILE_WRITE_ATTRIBUTES, windows.FILE_SHARE_WRITE, nil,
windows.OPEN_EXISTING, windows.FILE_FLAG_BACKUP_SEMANTICS, 0)
if e != nil {
return e
}
defer windows.Close(h)
c := windows.NsecToFiletime(windows.TimespecToNsec(ctimespec))
return windows.SetFileTime(h, &c, nil, nil)
}

View file

@ -1,13 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"errors"
)
var (
// ErrNotSupportedPlatform means the platform is not supported.
ErrNotSupportedPlatform = errors.New("platform and architecture is not supported")
// ErrNotSupportedOperatingSystem means the operating system is not supported.
ErrNotSupportedOperatingSystem = errors.New("operating system is not supported")
)

View file

@ -1,19 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"fmt"
"os/exec"
"syscall"
)
// GetExitCode returns the ExitStatus of the specified error if its type is
// exec.ExitError, returns 0 and an error otherwise.
func GetExitCode(err error) (int, error) {
exitCode := 0
if exiterr, ok := err.(*exec.ExitError); ok {
if procExit, ok := exiterr.Sys().(syscall.WaitStatus); ok {
return procExit.ExitStatus(), nil
}
}
return exitCode, fmt.Errorf("failed to get exit code")
}

View file

@ -1,67 +0,0 @@
// +build !windows
package system // import "github.com/docker/docker/pkg/system"
import (
"io/ioutil"
"os"
"path/filepath"
)
// MkdirAllWithACL is a wrapper for MkdirAll on unix systems.
func MkdirAllWithACL(path string, perm os.FileMode, sddl string) error {
return MkdirAll(path, perm, sddl)
}
// MkdirAll creates a directory named path along with any necessary parents,
// with permission specified by attribute perm for all dir created.
func MkdirAll(path string, perm os.FileMode, sddl string) error {
return os.MkdirAll(path, perm)
}
// IsAbs is a platform-specific wrapper for filepath.IsAbs.
func IsAbs(path string) bool {
return filepath.IsAbs(path)
}
// The functions below here are wrappers for the equivalents in the os and ioutils packages.
// They are passthrough on Unix platforms, and only relevant on Windows.
// CreateSequential creates the named file with mode 0666 (before umask), truncating
// it if it already exists. If successful, methods on the returned
// File can be used for I/O; the associated file descriptor has mode
// O_RDWR.
// If there is an error, it will be of type *PathError.
func CreateSequential(name string) (*os.File, error) {
return os.Create(name)
}
// OpenSequential opens the named file for reading. If successful, methods on
// the returned file can be used for reading; the associated file
// descriptor has mode O_RDONLY.
// If there is an error, it will be of type *PathError.
func OpenSequential(name string) (*os.File, error) {
return os.Open(name)
}
// OpenFileSequential is the generalized open call; most users will use Open
// or Create instead. It opens the named file with specified flag
// (O_RDONLY etc.) and perm, (0666 etc.) if applicable. If successful,
// methods on the returned File can be used for I/O.
// If there is an error, it will be of type *PathError.
func OpenFileSequential(name string, flag int, perm os.FileMode) (*os.File, error) {
return os.OpenFile(name, flag, perm)
}
// TempFileSequential creates a new temporary file in the directory dir
// with a name beginning with prefix, opens the file for reading
// and writing, and returns the resulting *os.File.
// If dir is the empty string, TempFile uses the default directory
// for temporary files (see os.TempDir).
// Multiple programs calling TempFile simultaneously
// will not choose the same file. The caller can use f.Name()
// to find the pathname of the file. It is the caller's responsibility
// to remove the file when no longer needed.
func TempFileSequential(dir, prefix string) (f *os.File, err error) {
return ioutil.TempFile(dir, prefix)
}

View file

@ -1,294 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"os"
"path/filepath"
"regexp"
"strconv"
"strings"
"sync"
"syscall"
"time"
"unsafe"
winio "github.com/Microsoft/go-winio"
"golang.org/x/sys/windows"
)
const (
// SddlAdministratorsLocalSystem is local administrators plus NT AUTHORITY\System
SddlAdministratorsLocalSystem = "D:P(A;OICI;GA;;;BA)(A;OICI;GA;;;SY)"
)
// MkdirAllWithACL is a wrapper for MkdirAll that creates a directory
// with an appropriate SDDL defined ACL.
func MkdirAllWithACL(path string, perm os.FileMode, sddl string) error {
return mkdirall(path, true, sddl)
}
// MkdirAll implementation that is volume path aware for Windows.
func MkdirAll(path string, _ os.FileMode, sddl string) error {
return mkdirall(path, false, sddl)
}
// mkdirall is a custom version of os.MkdirAll modified for use on Windows
// so that it is both volume path aware, and can create a directory with
// a DACL.
func mkdirall(path string, applyACL bool, sddl string) error {
if re := regexp.MustCompile(`^\\\\\?\\Volume{[a-z0-9-]+}$`); re.MatchString(path) {
return nil
}
// The rest of this method is largely copied from os.MkdirAll and should be kept
// as-is to ensure compatibility.
// Fast path: if we can tell whether path is a directory or file, stop with success or error.
dir, err := os.Stat(path)
if err == nil {
if dir.IsDir() {
return nil
}
return &os.PathError{
Op: "mkdir",
Path: path,
Err: syscall.ENOTDIR,
}
}
// Slow path: make sure parent exists and then call Mkdir for path.
i := len(path)
for i > 0 && os.IsPathSeparator(path[i-1]) { // Skip trailing path separator.
i--
}
j := i
for j > 0 && !os.IsPathSeparator(path[j-1]) { // Scan backward over element.
j--
}
if j > 1 {
// Create parent
err = mkdirall(path[0:j-1], false, sddl)
if err != nil {
return err
}
}
// Parent now exists; invoke os.Mkdir or mkdirWithACL and use its result.
if applyACL {
err = mkdirWithACL(path, sddl)
} else {
err = os.Mkdir(path, 0)
}
if err != nil {
// Handle arguments like "foo/." by
// double-checking that directory doesn't exist.
dir, err1 := os.Lstat(path)
if err1 == nil && dir.IsDir() {
return nil
}
return err
}
return nil
}
// mkdirWithACL creates a new directory. If there is an error, it will be of
// type *PathError. .
//
// This is a modified and combined version of os.Mkdir and windows.Mkdir
// in golang to cater for creating a directory am ACL permitting full
// access, with inheritance, to any subfolder/file for Built-in Administrators
// and Local System.
func mkdirWithACL(name string, sddl string) error {
sa := windows.SecurityAttributes{Length: 0}
sd, err := winio.SddlToSecurityDescriptor(sddl)
if err != nil {
return &os.PathError{Op: "mkdir", Path: name, Err: err}
}
sa.Length = uint32(unsafe.Sizeof(sa))
sa.InheritHandle = 1
sa.SecurityDescriptor = uintptr(unsafe.Pointer(&sd[0]))
namep, err := windows.UTF16PtrFromString(name)
if err != nil {
return &os.PathError{Op: "mkdir", Path: name, Err: err}
}
e := windows.CreateDirectory(namep, &sa)
if e != nil {
return &os.PathError{Op: "mkdir", Path: name, Err: e}
}
return nil
}
// IsAbs is a platform-specific wrapper for filepath.IsAbs. On Windows,
// golang filepath.IsAbs does not consider a path \windows\system32 as absolute
// as it doesn't start with a drive-letter/colon combination. However, in
// docker we need to verify things such as WORKDIR /windows/system32 in
// a Dockerfile (which gets translated to \windows\system32 when being processed
// by the daemon. This SHOULD be treated as absolute from a docker processing
// perspective.
func IsAbs(path string) bool {
if !filepath.IsAbs(path) {
if !strings.HasPrefix(path, string(os.PathSeparator)) {
return false
}
}
return true
}
// The origin of the functions below here are the golang OS and windows packages,
// slightly modified to only cope with files, not directories due to the
// specific use case.
//
// The alteration is to allow a file on Windows to be opened with
// FILE_FLAG_SEQUENTIAL_SCAN (particular for docker load), to avoid eating
// the standby list, particularly when accessing large files such as layer.tar.
// CreateSequential creates the named file with mode 0666 (before umask), truncating
// it if it already exists. If successful, methods on the returned
// File can be used for I/O; the associated file descriptor has mode
// O_RDWR.
// If there is an error, it will be of type *PathError.
func CreateSequential(name string) (*os.File, error) {
return OpenFileSequential(name, os.O_RDWR|os.O_CREATE|os.O_TRUNC, 0)
}
// OpenSequential opens the named file for reading. If successful, methods on
// the returned file can be used for reading; the associated file
// descriptor has mode O_RDONLY.
// If there is an error, it will be of type *PathError.
func OpenSequential(name string) (*os.File, error) {
return OpenFileSequential(name, os.O_RDONLY, 0)
}
// OpenFileSequential is the generalized open call; most users will use Open
// or Create instead.
// If there is an error, it will be of type *PathError.
func OpenFileSequential(name string, flag int, _ os.FileMode) (*os.File, error) {
if name == "" {
return nil, &os.PathError{Op: "open", Path: name, Err: syscall.ENOENT}
}
r, errf := windowsOpenFileSequential(name, flag, 0)
if errf == nil {
return r, nil
}
return nil, &os.PathError{Op: "open", Path: name, Err: errf}
}
func windowsOpenFileSequential(name string, flag int, _ os.FileMode) (file *os.File, err error) {
r, e := windowsOpenSequential(name, flag|windows.O_CLOEXEC, 0)
if e != nil {
return nil, e
}
return os.NewFile(uintptr(r), name), nil
}
func makeInheritSa() *windows.SecurityAttributes {
var sa windows.SecurityAttributes
sa.Length = uint32(unsafe.Sizeof(sa))
sa.InheritHandle = 1
return &sa
}
func windowsOpenSequential(path string, mode int, _ uint32) (fd windows.Handle, err error) {
if len(path) == 0 {
return windows.InvalidHandle, windows.ERROR_FILE_NOT_FOUND
}
pathp, err := windows.UTF16PtrFromString(path)
if err != nil {
return windows.InvalidHandle, err
}
var access uint32
switch mode & (windows.O_RDONLY | windows.O_WRONLY | windows.O_RDWR) {
case windows.O_RDONLY:
access = windows.GENERIC_READ
case windows.O_WRONLY:
access = windows.GENERIC_WRITE
case windows.O_RDWR:
access = windows.GENERIC_READ | windows.GENERIC_WRITE
}
if mode&windows.O_CREAT != 0 {
access |= windows.GENERIC_WRITE
}
if mode&windows.O_APPEND != 0 {
access &^= windows.GENERIC_WRITE
access |= windows.FILE_APPEND_DATA
}
sharemode := uint32(windows.FILE_SHARE_READ | windows.FILE_SHARE_WRITE)
var sa *windows.SecurityAttributes
if mode&windows.O_CLOEXEC == 0 {
sa = makeInheritSa()
}
var createmode uint32
switch {
case mode&(windows.O_CREAT|windows.O_EXCL) == (windows.O_CREAT | windows.O_EXCL):
createmode = windows.CREATE_NEW
case mode&(windows.O_CREAT|windows.O_TRUNC) == (windows.O_CREAT | windows.O_TRUNC):
createmode = windows.CREATE_ALWAYS
case mode&windows.O_CREAT == windows.O_CREAT:
createmode = windows.OPEN_ALWAYS
case mode&windows.O_TRUNC == windows.O_TRUNC:
createmode = windows.TRUNCATE_EXISTING
default:
createmode = windows.OPEN_EXISTING
}
// Use FILE_FLAG_SEQUENTIAL_SCAN rather than FILE_ATTRIBUTE_NORMAL as implemented in golang.
//https://msdn.microsoft.com/en-us/library/windows/desktop/aa363858(v=vs.85).aspx
const fileFlagSequentialScan = 0x08000000 // FILE_FLAG_SEQUENTIAL_SCAN
h, e := windows.CreateFile(pathp, access, sharemode, sa, createmode, fileFlagSequentialScan, 0)
return h, e
}
// Helpers for TempFileSequential
var rand uint32
var randmu sync.Mutex
func reseed() uint32 {
return uint32(time.Now().UnixNano() + int64(os.Getpid()))
}
func nextSuffix() string {
randmu.Lock()
r := rand
if r == 0 {
r = reseed()
}
r = r*1664525 + 1013904223 // constants from Numerical Recipes
rand = r
randmu.Unlock()
return strconv.Itoa(int(1e9 + r%1e9))[1:]
}
// TempFileSequential is a copy of ioutil.TempFile, modified to use sequential
// file access. Below is the original comment from golang:
// TempFile creates a new temporary file in the directory dir
// with a name beginning with prefix, opens the file for reading
// and writing, and returns the resulting *os.File.
// If dir is the empty string, TempFile uses the default directory
// for temporary files (see os.TempDir).
// Multiple programs calling TempFile simultaneously
// will not choose the same file. The caller can use f.Name()
// to find the pathname of the file. It is the caller's responsibility
// to remove the file when no longer needed.
func TempFileSequential(dir, prefix string) (f *os.File, err error) {
if dir == "" {
dir = os.TempDir()
}
nconflict := 0
for i := 0; i < 10000; i++ {
name := filepath.Join(dir, prefix+nextSuffix())
f, err = OpenFileSequential(name, os.O_RDWR|os.O_CREATE|os.O_EXCL, 0600)
if os.IsExist(err) {
if nconflict++; nconflict > 10 {
randmu.Lock()
rand = reseed()
randmu.Unlock()
}
continue
}
break
}
return
}

View file

@ -1,22 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"syscall"
"time"
"unsafe"
)
// Used by chtimes
var maxTime time.Time
func init() {
// chtimes initialization
if unsafe.Sizeof(syscall.Timespec{}.Nsec) == 8 {
// This is a 64 bit timespec
// os.Chtimes limits time to the following
maxTime = time.Unix(0, 1<<63-1)
} else {
// This is a 32 bit timespec
maxTime = time.Unix(1<<31-1, 0)
}
}

View file

@ -1,12 +0,0 @@
// +build !windows
package system // import "github.com/docker/docker/pkg/system"
// InitLCOW does nothing since LCOW is a windows only feature
func InitLCOW(experimental bool) {
}
// ContainerdRuntimeSupported returns true if the use of ContainerD runtime is supported.
func ContainerdRuntimeSupported(_ bool, _ string) bool {
return true
}

View file

@ -1,40 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"os"
"github.com/sirupsen/logrus"
)
var (
// lcowSupported determines if Linux Containers on Windows are supported.
lcowSupported = false
// containerdRuntimeSupported determines if ContainerD should be the runtime.
// As of March 2019, this is an experimental feature.
containerdRuntimeSupported = false
)
// InitLCOW sets whether LCOW is supported or not
func InitLCOW(experimental bool) {
v := GetOSVersion()
if experimental && v.Build >= 16299 {
lcowSupported = true
}
}
// InitContainerdRuntime sets whether to use ContainerD for runtime
// on Windows. This is an experimental feature still in development, and
// also requires an environment variable to be set (so as not to turn the
// feature on from simply experimental which would also mean LCOW.
func InitContainerdRuntime(experimental bool, cdPath string) {
if experimental && len(cdPath) > 0 && len(os.Getenv("DOCKER_WINDOWS_CONTAINERD_RUNTIME")) > 0 {
logrus.Warnf("Using ContainerD runtime. This feature is experimental")
containerdRuntimeSupported = true
}
}
// ContainerdRuntimeSupported returns true if the use of ContainerD runtime is supported.
func ContainerdRuntimeSupported() bool {
return containerdRuntimeSupported
}

View file

@ -1,32 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"runtime"
"strings"
specs "github.com/opencontainers/image-spec/specs-go/v1"
"github.com/pkg/errors"
)
// IsOSSupported determines if an operating system is supported by the host
func IsOSSupported(os string) bool {
if strings.EqualFold(runtime.GOOS, os) {
return true
}
if LCOWSupported() && strings.EqualFold(os, "linux") {
return true
}
return false
}
// ValidatePlatform determines if a platform structure is valid.
// TODO This is a temporary windows-only function, should be replaced by
// comparison of worker capabilities
func ValidatePlatform(platform specs.Platform) error {
if runtime.GOOS == "windows" {
if !(platform.OS == runtime.GOOS || (LCOWSupported() && platform.OS == "linux")) {
return errors.Errorf("unsupported os %s", platform.OS)
}
}
return nil
}

View file

@ -1,8 +0,0 @@
// +build !windows
package system // import "github.com/docker/docker/pkg/system"
// LCOWSupported returns true if Linux containers on Windows are supported.
func LCOWSupported() bool {
return false
}

View file

@ -1,6 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
// LCOWSupported returns true if Linux containers on Windows are supported.
func LCOWSupported() bool {
return lcowSupported
}

View file

@ -1,20 +0,0 @@
// +build !windows
package system // import "github.com/docker/docker/pkg/system"
import (
"os"
"syscall"
)
// Lstat takes a path to a file and returns
// a system.StatT type pertaining to that file.
//
// Throws an error if the file does not exist
func Lstat(path string) (*StatT, error) {
s := &syscall.Stat_t{}
if err := syscall.Lstat(path, s); err != nil {
return nil, &os.PathError{Op: "Lstat", Path: path, Err: err}
}
return fromStatT(s)
}

View file

@ -1,14 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import "os"
// Lstat calls os.Lstat to get a fileinfo interface back.
// This is then copied into our own locally defined structure.
func Lstat(path string) (*StatT, error) {
fi, err := os.Lstat(path)
if err != nil {
return nil, err
}
return fromStatT(&fi)
}

View file

@ -1,17 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
// MemInfo contains memory statistics of the host system.
type MemInfo struct {
// Total usable RAM (i.e. physical RAM minus a few reserved bits and the
// kernel binary code).
MemTotal int64
// Amount of free memory.
MemFree int64
// Total amount of swap space available.
SwapTotal int64
// Amount of swap space that is currently unused.
SwapFree int64
}

View file

@ -1,65 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"bufio"
"io"
"os"
"strconv"
"strings"
"github.com/docker/go-units"
)
// ReadMemInfo retrieves memory statistics of the host system and returns a
// MemInfo type.
func ReadMemInfo() (*MemInfo, error) {
file, err := os.Open("/proc/meminfo")
if err != nil {
return nil, err
}
defer file.Close()
return parseMemInfo(file)
}
// parseMemInfo parses the /proc/meminfo file into
// a MemInfo object given an io.Reader to the file.
// Throws error if there are problems reading from the file
func parseMemInfo(reader io.Reader) (*MemInfo, error) {
meminfo := &MemInfo{}
scanner := bufio.NewScanner(reader)
for scanner.Scan() {
// Expected format: ["MemTotal:", "1234", "kB"]
parts := strings.Fields(scanner.Text())
// Sanity checks: Skip malformed entries.
if len(parts) < 3 || parts[2] != "kB" {
continue
}
// Convert to bytes.
size, err := strconv.Atoi(parts[1])
if err != nil {
continue
}
bytes := int64(size) * units.KiB
switch parts[0] {
case "MemTotal:":
meminfo.MemTotal = bytes
case "MemFree:":
meminfo.MemFree = bytes
case "SwapTotal:":
meminfo.SwapTotal = bytes
case "SwapFree:":
meminfo.SwapFree = bytes
}
}
// Handle errors that may have occurred during the reading of the file.
if err := scanner.Err(); err != nil {
return nil, err
}
return meminfo, nil
}

View file

@ -1,8 +0,0 @@
// +build !linux,!windows
package system // import "github.com/docker/docker/pkg/system"
// ReadMemInfo is not supported on platforms other than linux and windows.
func ReadMemInfo() (*MemInfo, error) {
return nil, ErrNotSupportedPlatform
}

View file

@ -1,45 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"unsafe"
"golang.org/x/sys/windows"
)
var (
modkernel32 = windows.NewLazySystemDLL("kernel32.dll")
procGlobalMemoryStatusEx = modkernel32.NewProc("GlobalMemoryStatusEx")
)
// https://msdn.microsoft.com/en-us/library/windows/desktop/aa366589(v=vs.85).aspx
// https://msdn.microsoft.com/en-us/library/windows/desktop/aa366770(v=vs.85).aspx
type memorystatusex struct {
dwLength uint32
dwMemoryLoad uint32
ullTotalPhys uint64
ullAvailPhys uint64
ullTotalPageFile uint64
ullAvailPageFile uint64
ullTotalVirtual uint64
ullAvailVirtual uint64
ullAvailExtendedVirtual uint64
}
// ReadMemInfo retrieves memory statistics of the host system and returns a
// MemInfo type.
func ReadMemInfo() (*MemInfo, error) {
msi := &memorystatusex{
dwLength: 64,
}
r1, _, _ := procGlobalMemoryStatusEx.Call(uintptr(unsafe.Pointer(msi)))
if r1 == 0 {
return &MemInfo{}, nil
}
return &MemInfo{
MemTotal: int64(msi.ullTotalPhys),
MemFree: int64(msi.ullAvailPhys),
SwapTotal: int64(msi.ullTotalPageFile),
SwapFree: int64(msi.ullAvailPageFile),
}, nil
}

View file

@ -1,22 +0,0 @@
// +build !windows
package system // import "github.com/docker/docker/pkg/system"
import (
"golang.org/x/sys/unix"
)
// Mknod creates a filesystem node (file, device special file or named pipe) named path
// with attributes specified by mode and dev.
func Mknod(path string, mode uint32, dev int) error {
return unix.Mknod(path, mode, dev)
}
// Mkdev is used to build the value of linux devices (in /dev/) which specifies major
// and minor number of the newly created device special file.
// Linux device nodes are a bit weird due to backwards compat with 16 bit device nodes.
// They are, from low to high: the lower 8 bits of the minor, then 12 bits of the major,
// then the top 12 bits of the minor.
func Mkdev(major int64, minor int64) uint32 {
return uint32(unix.Mkdev(uint32(major), uint32(minor)))
}

View file

@ -1,11 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
// Mknod is not implemented on Windows.
func Mknod(path string, mode uint32, dev int) error {
return ErrNotSupportedPlatform
}
// Mkdev is not implemented on Windows.
func Mkdev(major int64, minor int64) uint32 {
panic("Mkdev not implemented on Windows.")
}

View file

@ -1,60 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"fmt"
"path/filepath"
"runtime"
"strings"
"github.com/containerd/continuity/pathdriver"
)
const defaultUnixPathEnv = "/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
// DefaultPathEnv is unix style list of directories to search for
// executables. Each directory is separated from the next by a colon
// ':' character .
func DefaultPathEnv(os string) string {
if runtime.GOOS == "windows" {
if os != runtime.GOOS {
return defaultUnixPathEnv
}
// Deliberately empty on Windows containers on Windows as the default path will be set by
// the container. Docker has no context of what the default path should be.
return ""
}
return defaultUnixPathEnv
}
// CheckSystemDriveAndRemoveDriveLetter verifies that a path, if it includes a drive letter,
// is the system drive.
// On Linux: this is a no-op.
// On Windows: this does the following>
// CheckSystemDriveAndRemoveDriveLetter verifies and manipulates a Windows path.
// This is used, for example, when validating a user provided path in docker cp.
// If a drive letter is supplied, it must be the system drive. The drive letter
// is always removed. Also, it translates it to OS semantics (IOW / to \). We
// need the path in this syntax so that it can ultimately be concatenated with
// a Windows long-path which doesn't support drive-letters. Examples:
// C: --> Fail
// C:\ --> \
// a --> a
// /a --> \a
// d:\ --> Fail
func CheckSystemDriveAndRemoveDriveLetter(path string, driver pathdriver.PathDriver) (string, error) {
if runtime.GOOS != "windows" || LCOWSupported() {
return path, nil
}
if len(path) == 2 && string(path[1]) == ":" {
return "", fmt.Errorf("No relative path specified in %q", path)
}
if !driver.IsAbs(path) || len(path) < 2 {
return filepath.FromSlash(path), nil
}
if string(path[1]) == ":" && !strings.EqualFold(string(path[0]), "c") {
return "", fmt.Errorf("The specified path is not on the system drive (C:)")
}
return filepath.FromSlash(path[2:]), nil
}

View file

@ -1,10 +0,0 @@
// +build !windows
package system // import "github.com/docker/docker/pkg/system"
// GetLongPathName converts Windows short pathnames to full pathnames.
// For example C:\Users\ADMIN~1 --> C:\Users\Administrator.
// It is a no-op on non-Windows platforms
func GetLongPathName(path string) (string, error) {
return path, nil
}

View file

@ -1,24 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import "syscall"
// GetLongPathName converts Windows short pathnames to full pathnames.
// For example C:\Users\ADMIN~1 --> C:\Users\Administrator.
// It is a no-op on non-Windows platforms
func GetLongPathName(path string) (string, error) {
// See https://groups.google.com/forum/#!topic/golang-dev/1tufzkruoTg
p := syscall.StringToUTF16(path)
b := p // GetLongPathName says we can reuse buffer
n, err := syscall.GetLongPathName(&p[0], &b[0], uint32(len(b)))
if err != nil {
return "", err
}
if n > uint32(len(b)) {
b = make([]uint16, n)
_, err = syscall.GetLongPathName(&p[0], &b[0], uint32(len(b)))
if err != nil {
return "", err
}
}
return syscall.UTF16ToString(b), nil
}

View file

@ -1,24 +0,0 @@
// +build linux freebsd darwin
package system // import "github.com/docker/docker/pkg/system"
import (
"syscall"
"golang.org/x/sys/unix"
)
// IsProcessAlive returns true if process with a given pid is running.
func IsProcessAlive(pid int) bool {
err := unix.Kill(pid, syscall.Signal(0))
if err == nil || err == unix.EPERM {
return true
}
return false
}
// KillProcess force-stops a process.
func KillProcess(pid int) {
unix.Kill(pid, unix.SIGKILL)
}

View file

@ -1,18 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import "os"
// IsProcessAlive returns true if process with a given pid is running.
func IsProcessAlive(pid int) bool {
_, err := os.FindProcess(pid)
return err == nil
}
// KillProcess force-stops a process.
func KillProcess(pid int) {
p, err := os.FindProcess(pid)
if err == nil {
p.Kill()
}
}

View file

@ -1,80 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"os"
"syscall"
"time"
"github.com/docker/docker/pkg/mount"
"github.com/pkg/errors"
)
// EnsureRemoveAll wraps `os.RemoveAll` to check for specific errors that can
// often be remedied.
// Only use `EnsureRemoveAll` if you really want to make every effort to remove
// a directory.
//
// Because of the way `os.Remove` (and by extension `os.RemoveAll`) works, there
// can be a race between reading directory entries and then actually attempting
// to remove everything in the directory.
// These types of errors do not need to be returned since it's ok for the dir to
// be gone we can just retry the remove operation.
//
// This should not return a `os.ErrNotExist` kind of error under any circumstances
func EnsureRemoveAll(dir string) error {
notExistErr := make(map[string]bool)
// track retries
exitOnErr := make(map[string]int)
maxRetry := 50
// Attempt to unmount anything beneath this dir first
mount.RecursiveUnmount(dir)
for {
err := os.RemoveAll(dir)
if err == nil {
return nil
}
pe, ok := err.(*os.PathError)
if !ok {
return err
}
if os.IsNotExist(err) {
if notExistErr[pe.Path] {
return err
}
notExistErr[pe.Path] = true
// There is a race where some subdir can be removed but after the parent
// dir entries have been read.
// So the path could be from `os.Remove(subdir)`
// If the reported non-existent path is not the passed in `dir` we
// should just retry, but otherwise return with no error.
if pe.Path == dir {
return nil
}
continue
}
if pe.Err != syscall.EBUSY {
return err
}
if mounted, _ := mount.Mounted(pe.Path); mounted {
if e := mount.Unmount(pe.Path); e != nil {
if mounted, _ := mount.Mounted(pe.Path); mounted {
return errors.Wrapf(e, "error while removing %s", dir)
}
}
}
if exitOnErr[pe.Path] == maxRetry {
return err
}
exitOnErr[pe.Path]++
time.Sleep(100 * time.Millisecond)
}
}

View file

@ -1,13 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import "syscall"
// fromStatT converts a syscall.Stat_t type to a system.Stat_t type
func fromStatT(s *syscall.Stat_t) (*StatT, error) {
return &StatT{size: s.Size,
mode: uint32(s.Mode),
uid: s.Uid,
gid: s.Gid,
rdev: uint64(s.Rdev),
mtim: s.Mtimespec}, nil
}

View file

@ -1,13 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import "syscall"
// fromStatT converts a syscall.Stat_t type to a system.Stat_t type
func fromStatT(s *syscall.Stat_t) (*StatT, error) {
return &StatT{size: s.Size,
mode: uint32(s.Mode),
uid: s.Uid,
gid: s.Gid,
rdev: uint64(s.Rdev),
mtim: s.Mtimespec}, nil
}

View file

@ -1,19 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import "syscall"
// fromStatT converts a syscall.Stat_t type to a system.Stat_t type
func fromStatT(s *syscall.Stat_t) (*StatT, error) {
return &StatT{size: s.Size,
mode: s.Mode,
uid: s.Uid,
gid: s.Gid,
rdev: s.Rdev,
mtim: s.Mtim}, nil
}
// FromStatT converts a syscall.Stat_t type to a system.Stat_t type
// This is exposed on Linux as pkg/archive/changes uses it.
func FromStatT(s *syscall.Stat_t) (*StatT, error) {
return fromStatT(s)
}

View file

@ -1,13 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import "syscall"
// fromStatT converts a syscall.Stat_t type to a system.Stat_t type
func fromStatT(s *syscall.Stat_t) (*StatT, error) {
return &StatT{size: s.Size,
mode: uint32(s.Mode),
uid: s.Uid,
gid: s.Gid,
rdev: uint64(s.Rdev),
mtim: s.Mtim}, nil
}

View file

@ -1,13 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import "syscall"
// fromStatT converts a syscall.Stat_t type to a system.Stat_t type
func fromStatT(s *syscall.Stat_t) (*StatT, error) {
return &StatT{size: s.Size,
mode: uint32(s.Mode),
uid: s.Uid,
gid: s.Gid,
rdev: uint64(s.Rdev),
mtim: s.Mtim}, nil
}

View file

@ -1,66 +0,0 @@
// +build !windows
package system // import "github.com/docker/docker/pkg/system"
import (
"os"
"syscall"
)
// StatT type contains status of a file. It contains metadata
// like permission, owner, group, size, etc about a file.
type StatT struct {
mode uint32
uid uint32
gid uint32
rdev uint64
size int64
mtim syscall.Timespec
}
// Mode returns file's permission mode.
func (s StatT) Mode() uint32 {
return s.mode
}
// UID returns file's user id of owner.
func (s StatT) UID() uint32 {
return s.uid
}
// GID returns file's group id of owner.
func (s StatT) GID() uint32 {
return s.gid
}
// Rdev returns file's device ID (if it's special file).
func (s StatT) Rdev() uint64 {
return s.rdev
}
// Size returns file's size.
func (s StatT) Size() int64 {
return s.size
}
// Mtim returns file's last modification time.
func (s StatT) Mtim() syscall.Timespec {
return s.mtim
}
// IsDir reports whether s describes a directory.
func (s StatT) IsDir() bool {
return s.mode&syscall.S_IFDIR != 0
}
// Stat takes a path to a file and returns
// a system.StatT type pertaining to that file.
//
// Throws an error if the file does not exist
func Stat(path string) (*StatT, error) {
s := &syscall.Stat_t{}
if err := syscall.Stat(path, s); err != nil {
return nil, &os.PathError{Op: "Stat", Path: path, Err: err}
}
return fromStatT(s)
}

View file

@ -1,49 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"os"
"time"
)
// StatT type contains status of a file. It contains metadata
// like permission, size, etc about a file.
type StatT struct {
mode os.FileMode
size int64
mtim time.Time
}
// Size returns file's size.
func (s StatT) Size() int64 {
return s.size
}
// Mode returns file's permission mode.
func (s StatT) Mode() os.FileMode {
return os.FileMode(s.mode)
}
// Mtim returns file's last modification time.
func (s StatT) Mtim() time.Time {
return time.Time(s.mtim)
}
// Stat takes a path to a file and returns
// a system.StatT type pertaining to that file.
//
// Throws an error if the file does not exist
func Stat(path string) (*StatT, error) {
fi, err := os.Stat(path)
if err != nil {
return nil, err
}
return fromStatT(&fi)
}
// fromStatT converts a os.FileInfo type to a system.StatT type
func fromStatT(fi *os.FileInfo) (*StatT, error) {
return &StatT{
size: (*fi).Size(),
mode: (*fi).Mode(),
mtim: (*fi).ModTime()}, nil
}

View file

@ -1,17 +0,0 @@
// +build linux freebsd
package system // import "github.com/docker/docker/pkg/system"
import "golang.org/x/sys/unix"
// Unmount is a platform-specific helper function to call
// the unmount syscall.
func Unmount(dest string) error {
return unix.Unmount(dest, 0)
}
// CommandLineToArgv should not be used on Unix.
// It simply returns commandLine in the only element in the returned array.
func CommandLineToArgv(commandLine string) ([]string, error) {
return []string{commandLine}, nil
}

View file

@ -1,193 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"fmt"
"syscall"
"unsafe"
"github.com/sirupsen/logrus"
"golang.org/x/sys/windows"
)
const (
OWNER_SECURITY_INFORMATION = 0x00000001
GROUP_SECURITY_INFORMATION = 0x00000002
DACL_SECURITY_INFORMATION = 0x00000004
SACL_SECURITY_INFORMATION = 0x00000008
LABEL_SECURITY_INFORMATION = 0x00000010
ATTRIBUTE_SECURITY_INFORMATION = 0x00000020
SCOPE_SECURITY_INFORMATION = 0x00000040
PROCESS_TRUST_LABEL_SECURITY_INFORMATION = 0x00000080
ACCESS_FILTER_SECURITY_INFORMATION = 0x00000100
BACKUP_SECURITY_INFORMATION = 0x00010000
PROTECTED_DACL_SECURITY_INFORMATION = 0x80000000
PROTECTED_SACL_SECURITY_INFORMATION = 0x40000000
UNPROTECTED_DACL_SECURITY_INFORMATION = 0x20000000
UNPROTECTED_SACL_SECURITY_INFORMATION = 0x10000000
)
const (
SE_UNKNOWN_OBJECT_TYPE = iota
SE_FILE_OBJECT
SE_SERVICE
SE_PRINTER
SE_REGISTRY_KEY
SE_LMSHARE
SE_KERNEL_OBJECT
SE_WINDOW_OBJECT
SE_DS_OBJECT
SE_DS_OBJECT_ALL
SE_PROVIDER_DEFINED_OBJECT
SE_WMIGUID_OBJECT
SE_REGISTRY_WOW64_32KEY
)
const (
SeTakeOwnershipPrivilege = "SeTakeOwnershipPrivilege"
)
const (
ContainerAdministratorSidString = "S-1-5-93-2-1"
ContainerUserSidString = "S-1-5-93-2-2"
)
var (
ntuserApiset = windows.NewLazyDLL("ext-ms-win-ntuser-window-l1-1-0")
modadvapi32 = windows.NewLazySystemDLL("advapi32.dll")
procGetVersionExW = modkernel32.NewProc("GetVersionExW")
procGetProductInfo = modkernel32.NewProc("GetProductInfo")
procSetNamedSecurityInfo = modadvapi32.NewProc("SetNamedSecurityInfoW")
procGetSecurityDescriptorDacl = modadvapi32.NewProc("GetSecurityDescriptorDacl")
)
// OSVersion is a wrapper for Windows version information
// https://msdn.microsoft.com/en-us/library/windows/desktop/ms724439(v=vs.85).aspx
type OSVersion struct {
Version uint32
MajorVersion uint8
MinorVersion uint8
Build uint16
}
// https://msdn.microsoft.com/en-us/library/windows/desktop/ms724833(v=vs.85).aspx
type osVersionInfoEx struct {
OSVersionInfoSize uint32
MajorVersion uint32
MinorVersion uint32
BuildNumber uint32
PlatformID uint32
CSDVersion [128]uint16
ServicePackMajor uint16
ServicePackMinor uint16
SuiteMask uint16
ProductType byte
Reserve byte
}
// GetOSVersion gets the operating system version on Windows. Note that
// docker.exe must be manifested to get the correct version information.
func GetOSVersion() OSVersion {
var err error
osv := OSVersion{}
osv.Version, err = windows.GetVersion()
if err != nil {
// GetVersion never fails.
panic(err)
}
osv.MajorVersion = uint8(osv.Version & 0xFF)
osv.MinorVersion = uint8(osv.Version >> 8 & 0xFF)
osv.Build = uint16(osv.Version >> 16)
return osv
}
func (osv OSVersion) ToString() string {
return fmt.Sprintf("%d.%d.%d", osv.MajorVersion, osv.MinorVersion, osv.Build)
}
// IsWindowsClient returns true if the SKU is client
// @engine maintainers - this function should not be removed or modified as it
// is used to enforce licensing restrictions on Windows.
func IsWindowsClient() bool {
osviex := &osVersionInfoEx{OSVersionInfoSize: 284}
r1, _, err := procGetVersionExW.Call(uintptr(unsafe.Pointer(osviex)))
if r1 == 0 {
logrus.Warnf("GetVersionExW failed - assuming server SKU: %v", err)
return false
}
const verNTWorkstation = 0x00000001
return osviex.ProductType == verNTWorkstation
}
// IsIoTCore returns true if the currently running image is based off of
// Windows 10 IoT Core.
// @engine maintainers - this function should not be removed or modified as it
// is used to enforce licensing restrictions on Windows.
func IsIoTCore() bool {
var returnedProductType uint32
r1, _, err := procGetProductInfo.Call(6, 1, 0, 0, uintptr(unsafe.Pointer(&returnedProductType)))
if r1 == 0 {
logrus.Warnf("GetProductInfo failed - assuming this is not IoT: %v", err)
return false
}
const productIoTUAP = 0x0000007B
const productIoTUAPCommercial = 0x00000083
return returnedProductType == productIoTUAP || returnedProductType == productIoTUAPCommercial
}
// Unmount is a platform-specific helper function to call
// the unmount syscall. Not supported on Windows
func Unmount(dest string) error {
return nil
}
// CommandLineToArgv wraps the Windows syscall to turn a commandline into an argument array.
func CommandLineToArgv(commandLine string) ([]string, error) {
var argc int32
argsPtr, err := windows.UTF16PtrFromString(commandLine)
if err != nil {
return nil, err
}
argv, err := windows.CommandLineToArgv(argsPtr, &argc)
if err != nil {
return nil, err
}
defer windows.LocalFree(windows.Handle(uintptr(unsafe.Pointer(argv))))
newArgs := make([]string, argc)
for i, v := range (*argv)[:argc] {
newArgs[i] = string(windows.UTF16ToString((*v)[:]))
}
return newArgs, nil
}
// HasWin32KSupport determines whether containers that depend on win32k can
// run on this machine. Win32k is the driver used to implement windowing.
func HasWin32KSupport() bool {
// For now, check for ntuser API support on the host. In the future, a host
// may support win32k in containers even if the host does not support ntuser
// APIs.
return ntuserApiset.Load() == nil
}
func SetNamedSecurityInfo(objectName *uint16, objectType uint32, securityInformation uint32, sidOwner *windows.SID, sidGroup *windows.SID, dacl *byte, sacl *byte) (result error) {
r0, _, _ := syscall.Syscall9(procSetNamedSecurityInfo.Addr(), 7, uintptr(unsafe.Pointer(objectName)), uintptr(objectType), uintptr(securityInformation), uintptr(unsafe.Pointer(sidOwner)), uintptr(unsafe.Pointer(sidGroup)), uintptr(unsafe.Pointer(dacl)), uintptr(unsafe.Pointer(sacl)), 0, 0)
if r0 != 0 {
result = syscall.Errno(r0)
}
return
}
func GetSecurityDescriptorDacl(securityDescriptor *byte, daclPresent *uint32, dacl **byte, daclDefaulted *uint32) (result error) {
r1, _, e1 := syscall.Syscall6(procGetSecurityDescriptorDacl.Addr(), 4, uintptr(unsafe.Pointer(securityDescriptor)), uintptr(unsafe.Pointer(daclPresent)), uintptr(unsafe.Pointer(dacl)), uintptr(unsafe.Pointer(daclDefaulted)), 0, 0)
if r1 == 0 {
if e1 != 0 {
result = syscall.Errno(e1)
} else {
result = syscall.EINVAL
}
}
return
}

View file

@ -1,13 +0,0 @@
// +build !windows
package system // import "github.com/docker/docker/pkg/system"
import (
"golang.org/x/sys/unix"
)
// Umask sets current process's file mode creation mask to newmask
// and returns oldmask.
func Umask(newmask int) (oldmask int, err error) {
return unix.Umask(newmask), nil
}

View file

@ -1,7 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
// Umask is not supported on the windows platform.
func Umask(newmask int) (oldmask int, err error) {
// should not be called on cli code path
return 0, ErrNotSupportedPlatform
}

View file

@ -1,24 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"syscall"
"unsafe"
"golang.org/x/sys/unix"
)
// LUtimesNano is used to change access and modification time of the specified path.
// It's used for symbol link file because unix.UtimesNano doesn't support a NOFOLLOW flag atm.
func LUtimesNano(path string, ts []syscall.Timespec) error {
var _path *byte
_path, err := unix.BytePtrFromString(path)
if err != nil {
return err
}
if _, _, err := unix.Syscall(unix.SYS_LUTIMES, uintptr(unsafe.Pointer(_path)), uintptr(unsafe.Pointer(&ts[0])), 0); err != 0 && err != unix.ENOSYS {
return err
}
return nil
}

View file

@ -1,25 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import (
"syscall"
"unsafe"
"golang.org/x/sys/unix"
)
// LUtimesNano is used to change access and modification time of the specified path.
// It's used for symbol link file because unix.UtimesNano doesn't support a NOFOLLOW flag atm.
func LUtimesNano(path string, ts []syscall.Timespec) error {
atFdCwd := unix.AT_FDCWD
var _path *byte
_path, err := unix.BytePtrFromString(path)
if err != nil {
return err
}
if _, _, err := unix.Syscall6(unix.SYS_UTIMENSAT, uintptr(atFdCwd), uintptr(unsafe.Pointer(_path)), uintptr(unsafe.Pointer(&ts[0])), unix.AT_SYMLINK_NOFOLLOW, 0, 0); err != 0 && err != unix.ENOSYS {
return err
}
return nil
}

View file

@ -1,10 +0,0 @@
// +build !linux,!freebsd
package system // import "github.com/docker/docker/pkg/system"
import "syscall"
// LUtimesNano is only supported on linux and freebsd.
func LUtimesNano(path string, ts []syscall.Timespec) error {
return ErrNotSupportedPlatform
}

View file

@ -1,29 +0,0 @@
package system // import "github.com/docker/docker/pkg/system"
import "golang.org/x/sys/unix"
// Lgetxattr retrieves the value of the extended attribute identified by attr
// and associated with the given path in the file system.
// It will returns a nil slice and nil error if the xattr is not set.
func Lgetxattr(path string, attr string) ([]byte, error) {
dest := make([]byte, 128)
sz, errno := unix.Lgetxattr(path, attr, dest)
if errno == unix.ENODATA {
return nil, nil
}
if errno == unix.ERANGE {
dest = make([]byte, sz)
sz, errno = unix.Lgetxattr(path, attr, dest)
}
if errno != nil {
return nil, errno
}
return dest[:sz], nil
}
// Lsetxattr sets the value of the extended attribute identified by attr
// and associated with the given path in the file system.
func Lsetxattr(path string, attr string, data []byte, flags int) error {
return unix.Lsetxattr(path, attr, data, flags)
}

View file

@ -1,13 +0,0 @@
// +build !linux
package system // import "github.com/docker/docker/pkg/system"
// Lgetxattr is not supported on platforms other than linux.
func Lgetxattr(path string, attr string) ([]byte, error) {
return nil, ErrNotSupportedPlatform
}
// Lsetxattr is not supported on platforms other than linux.
func Lsetxattr(path string, attr string, data []byte, flags int) error {
return ErrNotSupportedPlatform
}

View file

@ -1,191 +0,0 @@
Apache License
Version 2.0, January 2004
https://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
Copyright 2015 Docker, Inc.
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
https://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.

View file

@ -1,16 +0,0 @@
[![GoDoc](https://godoc.org/github.com/docker/go-units?status.svg)](https://godoc.org/github.com/docker/go-units)
# Introduction
go-units is a library to transform human friendly measurements into machine friendly values.
## Usage
See the [docs in godoc](https://godoc.org/github.com/docker/go-units) for examples and documentation.
## Copyright and license
Copyright © 2015 Docker, Inc.
go-units is licensed under the Apache License, Version 2.0.
See [LICENSE](LICENSE) for the full text of the license.

View file

@ -1,35 +0,0 @@
// Package units provides helper function to parse and print size and time units
// in human-readable format.
package units
import (
"fmt"
"time"
)
// HumanDuration returns a human-readable approximation of a duration
// (eg. "About a minute", "4 hours ago", etc.).
func HumanDuration(d time.Duration) string {
if seconds := int(d.Seconds()); seconds < 1 {
return "Less than a second"
} else if seconds == 1 {
return "1 second"
} else if seconds < 60 {
return fmt.Sprintf("%d seconds", seconds)
} else if minutes := int(d.Minutes()); minutes == 1 {
return "About a minute"
} else if minutes < 46 {
return fmt.Sprintf("%d minutes", minutes)
} else if hours := int(d.Hours() + 0.5); hours == 1 {
return "About an hour"
} else if hours < 48 {
return fmt.Sprintf("%d hours", hours)
} else if hours < 24*7*2 {
return fmt.Sprintf("%d days", hours/24)
} else if hours < 24*30*2 {
return fmt.Sprintf("%d weeks", hours/24/7)
} else if hours < 24*365*2 {
return fmt.Sprintf("%d months", hours/24/30)
}
return fmt.Sprintf("%d years", int(d.Hours())/24/365)
}

View file

@ -1,108 +0,0 @@
package units
import (
"fmt"
"regexp"
"strconv"
"strings"
)
// See: http://en.wikipedia.org/wiki/Binary_prefix
const (
// Decimal
KB = 1000
MB = 1000 * KB
GB = 1000 * MB
TB = 1000 * GB
PB = 1000 * TB
// Binary
KiB = 1024
MiB = 1024 * KiB
GiB = 1024 * MiB
TiB = 1024 * GiB
PiB = 1024 * TiB
)
type unitMap map[string]int64
var (
decimalMap = unitMap{"k": KB, "m": MB, "g": GB, "t": TB, "p": PB}
binaryMap = unitMap{"k": KiB, "m": MiB, "g": GiB, "t": TiB, "p": PiB}
sizeRegex = regexp.MustCompile(`^(\d+(\.\d+)*) ?([kKmMgGtTpP])?[iI]?[bB]?$`)
)
var decimapAbbrs = []string{"B", "kB", "MB", "GB", "TB", "PB", "EB", "ZB", "YB"}
var binaryAbbrs = []string{"B", "KiB", "MiB", "GiB", "TiB", "PiB", "EiB", "ZiB", "YiB"}
func getSizeAndUnit(size float64, base float64, _map []string) (float64, string) {
i := 0
unitsLimit := len(_map) - 1
for size >= base && i < unitsLimit {
size = size / base
i++
}
return size, _map[i]
}
// CustomSize returns a human-readable approximation of a size
// using custom format.
func CustomSize(format string, size float64, base float64, _map []string) string {
size, unit := getSizeAndUnit(size, base, _map)
return fmt.Sprintf(format, size, unit)
}
// HumanSizeWithPrecision allows the size to be in any precision,
// instead of 4 digit precision used in units.HumanSize.
func HumanSizeWithPrecision(size float64, precision int) string {
size, unit := getSizeAndUnit(size, 1000.0, decimapAbbrs)
return fmt.Sprintf("%.*g%s", precision, size, unit)
}
// HumanSize returns a human-readable approximation of a size
// capped at 4 valid numbers (eg. "2.746 MB", "796 KB").
func HumanSize(size float64) string {
return HumanSizeWithPrecision(size, 4)
}
// BytesSize returns a human-readable size in bytes, kibibytes,
// mebibytes, gibibytes, or tebibytes (eg. "44kiB", "17MiB").
func BytesSize(size float64) string {
return CustomSize("%.4g%s", size, 1024.0, binaryAbbrs)
}
// FromHumanSize returns an integer from a human-readable specification of a
// size using SI standard (eg. "44kB", "17MB").
func FromHumanSize(size string) (int64, error) {
return parseSize(size, decimalMap)
}
// RAMInBytes parses a human-readable string representing an amount of RAM
// in bytes, kibibytes, mebibytes, gibibytes, or tebibytes and
// returns the number of bytes, or -1 if the string is unparseable.
// Units are case-insensitive, and the 'b' suffix is optional.
func RAMInBytes(size string) (int64, error) {
return parseSize(size, binaryMap)
}
// Parses the human-readable size string into the amount it represents.
func parseSize(sizeStr string, uMap unitMap) (int64, error) {
matches := sizeRegex.FindStringSubmatch(sizeStr)
if len(matches) != 4 {
return -1, fmt.Errorf("invalid size: '%s'", sizeStr)
}
size, err := strconv.ParseFloat(matches[1], 64)
if err != nil {
return -1, err
}
unitPrefix := strings.ToLower(matches[3])
if mul, ok := uMap[unitPrefix]; ok {
size *= float64(mul)
}
return int64(size), nil
}

View file

@ -1,118 +0,0 @@
package units
import (
"fmt"
"strconv"
"strings"
)
// Ulimit is a human friendly version of Rlimit.
type Ulimit struct {
Name string
Hard int64
Soft int64
}
// Rlimit specifies the resource limits, such as max open files.
type Rlimit struct {
Type int `json:"type,omitempty"`
Hard uint64 `json:"hard,omitempty"`
Soft uint64 `json:"soft,omitempty"`
}
const (
// magic numbers for making the syscall
// some of these are defined in the syscall package, but not all.
// Also since Windows client doesn't get access to the syscall package, need to
// define these here
rlimitAs = 9
rlimitCore = 4
rlimitCPU = 0
rlimitData = 2
rlimitFsize = 1
rlimitLocks = 10
rlimitMemlock = 8
rlimitMsgqueue = 12
rlimitNice = 13
rlimitNofile = 7
rlimitNproc = 6
rlimitRss = 5
rlimitRtprio = 14
rlimitRttime = 15
rlimitSigpending = 11
rlimitStack = 3
)
var ulimitNameMapping = map[string]int{
//"as": rlimitAs, // Disabled since this doesn't seem usable with the way Docker inits a container.
"core": rlimitCore,
"cpu": rlimitCPU,
"data": rlimitData,
"fsize": rlimitFsize,
"locks": rlimitLocks,
"memlock": rlimitMemlock,
"msgqueue": rlimitMsgqueue,
"nice": rlimitNice,
"nofile": rlimitNofile,
"nproc": rlimitNproc,
"rss": rlimitRss,
"rtprio": rlimitRtprio,
"rttime": rlimitRttime,
"sigpending": rlimitSigpending,
"stack": rlimitStack,
}
// ParseUlimit parses and returns a Ulimit from the specified string.
func ParseUlimit(val string) (*Ulimit, error) {
parts := strings.SplitN(val, "=", 2)
if len(parts) != 2 {
return nil, fmt.Errorf("invalid ulimit argument: %s", val)
}
if _, exists := ulimitNameMapping[parts[0]]; !exists {
return nil, fmt.Errorf("invalid ulimit type: %s", parts[0])
}
var (
soft int64
hard = &soft // default to soft in case no hard was set
temp int64
err error
)
switch limitVals := strings.Split(parts[1], ":"); len(limitVals) {
case 2:
temp, err = strconv.ParseInt(limitVals[1], 10, 64)
if err != nil {
return nil, err
}
hard = &temp
fallthrough
case 1:
soft, err = strconv.ParseInt(limitVals[0], 10, 64)
if err != nil {
return nil, err
}
default:
return nil, fmt.Errorf("too many limit value arguments - %s, can only have up to two, `soft[:hard]`", parts[1])
}
if soft > *hard {
return nil, fmt.Errorf("ulimit soft limit must be less than or equal to hard limit: %d > %d", soft, *hard)
}
return &Ulimit{Name: parts[0], Soft: soft, Hard: *hard}, nil
}
// GetRlimit returns the RLimit corresponding to Ulimit.
func (u *Ulimit) GetRlimit() (*Rlimit, error) {
t, exists := ulimitNameMapping[u.Name]
if !exists {
return nil, fmt.Errorf("invalid ulimit name %s", u.Name)
}
return &Rlimit{Type: t, Soft: uint64(u.Soft), Hard: uint64(u.Hard)}, nil
}
func (u *Ulimit) String() string {
return fmt.Sprintf("%s=%d:%d", u.Name, u.Soft, u.Hard)
}

View file

@ -1,191 +0,0 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
Copyright 2014 Docker, Inc.
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.

View file

@ -1,17 +0,0 @@
runc
Copyright 2012-2015 Docker, Inc.
This product includes software developed at Docker, Inc. (http://www.docker.com).
The following is courtesy of our legal counsel:
Use and transfer of Docker may be subject to certain restrictions by the
United States and other governments.
It is your responsibility to ensure that your use and/or transfer does not
violate applicable laws.
For more information, please see http://www.bis.doc.gov
See also http://www.apache.org/dev/crypto.html and/or seek legal counsel.

View file

@ -1,269 +0,0 @@
# runc
[![Build Status](https://travis-ci.org/opencontainers/runc.svg?branch=master)](https://travis-ci.org/opencontainers/runc)
[![Go Report Card](https://goreportcard.com/badge/github.com/opencontainers/runc)](https://goreportcard.com/report/github.com/opencontainers/runc)
[![GoDoc](https://godoc.org/github.com/opencontainers/runc?status.svg)](https://godoc.org/github.com/opencontainers/runc)
## Introduction
`runc` is a CLI tool for spawning and running containers according to the OCI specification.
## Releases
`runc` depends on and tracks the [runtime-spec](https://github.com/opencontainers/runtime-spec) repository.
We will try to make sure that `runc` and the OCI specification major versions stay in lockstep.
This means that `runc` 1.0.0 should implement the 1.0 version of the specification.
You can find official releases of `runc` on the [release](https://github.com/opencontainers/runc/releases) page.
## Security
Reporting process and disclosure communications are outlined in [/org/security](https://github.com/opencontainers/org/blob/master/security/)
## Building
`runc` currently supports the Linux platform with various architecture support.
It must be built with Go version 1.6 or higher in order for some features to function properly.
In order to enable seccomp support you will need to install `libseccomp` on your platform.
> e.g. `libseccomp-devel` for CentOS, or `libseccomp-dev` for Ubuntu
Otherwise, if you do not want to build `runc` with seccomp support you can add `BUILDTAGS=""` when running make.
```bash
# create a 'github.com/opencontainers' in your GOPATH/src
cd github.com/opencontainers
git clone https://github.com/opencontainers/runc
cd runc
make
sudo make install
```
You can also use `go get` to install to your `GOPATH`, assuming that you have a `github.com` parent folder already created under `src`:
```bash
go get github.com/opencontainers/runc
cd $GOPATH/src/github.com/opencontainers/runc
make
sudo make install
```
`runc` will be installed to `/usr/local/sbin/runc` on your system.
#### Build Tags
`runc` supports optional build tags for compiling support of various features.
To add build tags to the make option the `BUILDTAGS` variable must be set.
```bash
make BUILDTAGS='seccomp apparmor'
```
| Build Tag | Feature | Dependency |
|-----------|------------------------------------|-------------|
| seccomp | Syscall filtering | libseccomp |
| selinux | selinux process and mount labeling | <none> |
| apparmor | apparmor profile support | <none> |
| ambient | ambient capability support | kernel 4.3 |
| nokmem | disable kernel memory account | <none> |
### Running the test suite
`runc` currently supports running its test suite via Docker.
To run the suite just type `make test`.
```bash
make test
```
There are additional make targets for running the tests outside of a container but this is not recommended as the tests are written with the expectation that they can write and remove anywhere.
You can run a specific test case by setting the `TESTFLAGS` variable.
```bash
# make test TESTFLAGS="-run=SomeTestFunction"
```
You can run a specific integration test by setting the `TESTPATH` variable.
```bash
# make test TESTPATH="/checkpoint.bats"
```
You can run a test in your proxy environment by setting `DOCKER_BUILD_PROXY` and `DOCKER_RUN_PROXY` variables.
```bash
# make test DOCKER_BUILD_PROXY="--build-arg HTTP_PROXY=http://yourproxy/" DOCKER_RUN_PROXY="-e HTTP_PROXY=http://yourproxy/"
```
### Dependencies Management
`runc` uses [vndr](https://github.com/LK4D4/vndr) for dependencies management.
Please refer to [vndr](https://github.com/LK4D4/vndr) for how to add or update
new dependencies.
## Using runc
### Creating an OCI Bundle
In order to use runc you must have your container in the format of an OCI bundle.
If you have Docker installed you can use its `export` method to acquire a root filesystem from an existing Docker container.
```bash
# create the top most bundle directory
mkdir /mycontainer
cd /mycontainer
# create the rootfs directory
mkdir rootfs
# export busybox via Docker into the rootfs directory
docker export $(docker create busybox) | tar -C rootfs -xvf -
```
After a root filesystem is populated you just generate a spec in the format of a `config.json` file inside your bundle.
`runc` provides a `spec` command to generate a base template spec that you are then able to edit.
To find features and documentation for fields in the spec please refer to the [specs](https://github.com/opencontainers/runtime-spec) repository.
```bash
runc spec
```
### Running Containers
Assuming you have an OCI bundle from the previous step you can execute the container in two different ways.
The first way is to use the convenience command `run` that will handle creating, starting, and deleting the container after it exits.
```bash
# run as root
cd /mycontainer
runc run mycontainerid
```
If you used the unmodified `runc spec` template this should give you a `sh` session inside the container.
The second way to start a container is using the specs lifecycle operations.
This gives you more power over how the container is created and managed while it is running.
This will also launch the container in the background so you will have to edit the `config.json` to remove the `terminal` setting for the simple examples here.
Your process field in the `config.json` should look like this below with `"terminal": false` and `"args": ["sleep", "5"]`.
```json
"process": {
"terminal": false,
"user": {
"uid": 0,
"gid": 0
},
"args": [
"sleep", "5"
],
"env": [
"PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin",
"TERM=xterm"
],
"cwd": "/",
"capabilities": {
"bounding": [
"CAP_AUDIT_WRITE",
"CAP_KILL",
"CAP_NET_BIND_SERVICE"
],
"effective": [
"CAP_AUDIT_WRITE",
"CAP_KILL",
"CAP_NET_BIND_SERVICE"
],
"inheritable": [
"CAP_AUDIT_WRITE",
"CAP_KILL",
"CAP_NET_BIND_SERVICE"
],
"permitted": [
"CAP_AUDIT_WRITE",
"CAP_KILL",
"CAP_NET_BIND_SERVICE"
],
"ambient": [
"CAP_AUDIT_WRITE",
"CAP_KILL",
"CAP_NET_BIND_SERVICE"
]
},
"rlimits": [
{
"type": "RLIMIT_NOFILE",
"hard": 1024,
"soft": 1024
}
],
"noNewPrivileges": true
},
```
Now we can go through the lifecycle operations in your shell.
```bash
# run as root
cd /mycontainer
runc create mycontainerid
# view the container is created and in the "created" state
runc list
# start the process inside the container
runc start mycontainerid
# after 5 seconds view that the container has exited and is now in the stopped state
runc list
# now delete the container
runc delete mycontainerid
```
This allows higher level systems to augment the containers creation logic with setup of various settings after the container is created and/or before it is deleted. For example, the container's network stack is commonly set up after `create` but before `start`.
#### Rootless containers
`runc` has the ability to run containers without root privileges. This is called `rootless`. You need to pass some parameters to `runc` in order to run rootless containers. See below and compare with the previous version. Run the following commands as an ordinary user:
```bash
# Same as the first example
mkdir ~/mycontainer
cd ~/mycontainer
mkdir rootfs
docker export $(docker create busybox) | tar -C rootfs -xvf -
# The --rootless parameter instructs runc spec to generate a configuration for a rootless container, which will allow you to run the container as a non-root user.
runc spec --rootless
# The --root parameter tells runc where to store the container state. It must be writable by the user.
runc --root /tmp/runc run mycontainerid
```
#### Supervisors
`runc` can be used with process supervisors and init systems to ensure that containers are restarted when they exit.
An example systemd unit file looks something like this.
```systemd
[Unit]
Description=Start My Container
[Service]
Type=forking
ExecStart=/usr/local/sbin/runc run -d --pid-file /run/mycontainerid.pid mycontainerid
ExecStopPost=/usr/local/sbin/runc delete mycontainerid
WorkingDirectory=/mycontainer
PIDFile=/run/mycontainerid.pid
[Install]
WantedBy=multi-user.target
```
## License
The code and docs are released under the [Apache 2.0 license](LICENSE).

View file

@ -1,330 +0,0 @@
# libcontainer
[![GoDoc](https://godoc.org/github.com/opencontainers/runc/libcontainer?status.svg)](https://godoc.org/github.com/opencontainers/runc/libcontainer)
Libcontainer provides a native Go implementation for creating containers
with namespaces, cgroups, capabilities, and filesystem access controls.
It allows you to manage the lifecycle of the container performing additional operations
after the container is created.
#### Container
A container is a self contained execution environment that shares the kernel of the
host system and which is (optionally) isolated from other containers in the system.
#### Using libcontainer
Because containers are spawned in a two step process you will need a binary that
will be executed as the init process for the container. In libcontainer, we use
the current binary (/proc/self/exe) to be executed as the init process, and use
arg "init", we call the first step process "bootstrap", so you always need a "init"
function as the entry of "bootstrap".
In addition to the go init function the early stage bootstrap is handled by importing
[nsenter](https://github.com/opencontainers/runc/blob/master/libcontainer/nsenter/README.md).
```go
import (
_ "github.com/opencontainers/runc/libcontainer/nsenter"
)
func init() {
if len(os.Args) > 1 && os.Args[1] == "init" {
runtime.GOMAXPROCS(1)
runtime.LockOSThread()
factory, _ := libcontainer.New("")
if err := factory.StartInitialization(); err != nil {
logrus.Fatal(err)
}
panic("--this line should have never been executed, congratulations--")
}
}
```
Then to create a container you first have to initialize an instance of a factory
that will handle the creation and initialization for a container.
```go
factory, err := libcontainer.New("/var/lib/container", libcontainer.Cgroupfs, libcontainer.InitArgs(os.Args[0], "init"))
if err != nil {
logrus.Fatal(err)
return
}
```
Once you have an instance of the factory created we can create a configuration
struct describing how the container is to be created. A sample would look similar to this:
```go
defaultMountFlags := unix.MS_NOEXEC | unix.MS_NOSUID | unix.MS_NODEV
config := &configs.Config{
Rootfs: "/your/path/to/rootfs",
Capabilities: &configs.Capabilities{
Bounding: []string{
"CAP_CHOWN",
"CAP_DAC_OVERRIDE",
"CAP_FSETID",
"CAP_FOWNER",
"CAP_MKNOD",
"CAP_NET_RAW",
"CAP_SETGID",
"CAP_SETUID",
"CAP_SETFCAP",
"CAP_SETPCAP",
"CAP_NET_BIND_SERVICE",
"CAP_SYS_CHROOT",
"CAP_KILL",
"CAP_AUDIT_WRITE",
},
Effective: []string{
"CAP_CHOWN",
"CAP_DAC_OVERRIDE",
"CAP_FSETID",
"CAP_FOWNER",
"CAP_MKNOD",
"CAP_NET_RAW",
"CAP_SETGID",
"CAP_SETUID",
"CAP_SETFCAP",
"CAP_SETPCAP",
"CAP_NET_BIND_SERVICE",
"CAP_SYS_CHROOT",
"CAP_KILL",
"CAP_AUDIT_WRITE",
},
Inheritable: []string{
"CAP_CHOWN",
"CAP_DAC_OVERRIDE",
"CAP_FSETID",
"CAP_FOWNER",
"CAP_MKNOD",
"CAP_NET_RAW",
"CAP_SETGID",
"CAP_SETUID",
"CAP_SETFCAP",
"CAP_SETPCAP",
"CAP_NET_BIND_SERVICE",
"CAP_SYS_CHROOT",
"CAP_KILL",
"CAP_AUDIT_WRITE",
},
Permitted: []string{
"CAP_CHOWN",
"CAP_DAC_OVERRIDE",
"CAP_FSETID",
"CAP_FOWNER",
"CAP_MKNOD",
"CAP_NET_RAW",
"CAP_SETGID",
"CAP_SETUID",
"CAP_SETFCAP",
"CAP_SETPCAP",
"CAP_NET_BIND_SERVICE",
"CAP_SYS_CHROOT",
"CAP_KILL",
"CAP_AUDIT_WRITE",
},
Ambient: []string{
"CAP_CHOWN",
"CAP_DAC_OVERRIDE",
"CAP_FSETID",
"CAP_FOWNER",
"CAP_MKNOD",
"CAP_NET_RAW",
"CAP_SETGID",
"CAP_SETUID",
"CAP_SETFCAP",
"CAP_SETPCAP",
"CAP_NET_BIND_SERVICE",
"CAP_SYS_CHROOT",
"CAP_KILL",
"CAP_AUDIT_WRITE",
},
},
Namespaces: configs.Namespaces([]configs.Namespace{
{Type: configs.NEWNS},
{Type: configs.NEWUTS},
{Type: configs.NEWIPC},
{Type: configs.NEWPID},
{Type: configs.NEWUSER},
{Type: configs.NEWNET},
{Type: configs.NEWCGROUP},
}),
Cgroups: &configs.Cgroup{
Name: "test-container",
Parent: "system",
Resources: &configs.Resources{
MemorySwappiness: nil,
AllowAllDevices: nil,
AllowedDevices: configs.DefaultAllowedDevices,
},
},
MaskPaths: []string{
"/proc/kcore",
"/sys/firmware",
},
ReadonlyPaths: []string{
"/proc/sys", "/proc/sysrq-trigger", "/proc/irq", "/proc/bus",
},
Devices: configs.DefaultAutoCreatedDevices,
Hostname: "testing",
Mounts: []*configs.Mount{
{
Source: "proc",
Destination: "/proc",
Device: "proc",
Flags: defaultMountFlags,
},
{
Source: "tmpfs",
Destination: "/dev",
Device: "tmpfs",
Flags: unix.MS_NOSUID | unix.MS_STRICTATIME,
Data: "mode=755",
},
{
Source: "devpts",
Destination: "/dev/pts",
Device: "devpts",
Flags: unix.MS_NOSUID | unix.MS_NOEXEC,
Data: "newinstance,ptmxmode=0666,mode=0620,gid=5",
},
{
Device: "tmpfs",
Source: "shm",
Destination: "/dev/shm",
Data: "mode=1777,size=65536k",
Flags: defaultMountFlags,
},
{
Source: "mqueue",
Destination: "/dev/mqueue",
Device: "mqueue",
Flags: defaultMountFlags,
},
{
Source: "sysfs",
Destination: "/sys",
Device: "sysfs",
Flags: defaultMountFlags | unix.MS_RDONLY,
},
},
UidMappings: []configs.IDMap{
{
ContainerID: 0,
HostID: 1000,
Size: 65536,
},
},
GidMappings: []configs.IDMap{
{
ContainerID: 0,
HostID: 1000,
Size: 65536,
},
},
Networks: []*configs.Network{
{
Type: "loopback",
Address: "127.0.0.1/0",
Gateway: "localhost",
},
},
Rlimits: []configs.Rlimit{
{
Type: unix.RLIMIT_NOFILE,
Hard: uint64(1025),
Soft: uint64(1025),
},
},
}
```
Once you have the configuration populated you can create a container:
```go
container, err := factory.Create("container-id", config)
if err != nil {
logrus.Fatal(err)
return
}
```
To spawn bash as the initial process inside the container and have the
processes pid returned in order to wait, signal, or kill the process:
```go
process := &libcontainer.Process{
Args: []string{"/bin/bash"},
Env: []string{"PATH=/bin"},
User: "daemon",
Stdin: os.Stdin,
Stdout: os.Stdout,
Stderr: os.Stderr,
}
err := container.Run(process)
if err != nil {
container.Destroy()
logrus.Fatal(err)
return
}
// wait for the process to finish.
_, err := process.Wait()
if err != nil {
logrus.Fatal(err)
}
// destroy the container.
container.Destroy()
```
Additional ways to interact with a running container are:
```go
// return all the pids for all processes running inside the container.
processes, err := container.Processes()
// get detailed cpu, memory, io, and network statistics for the container and
// it's processes.
stats, err := container.Stats()
// pause all processes inside the container.
container.Pause()
// resume all paused processes.
container.Resume()
// send signal to container's init process.
container.Signal(signal)
// update container resource constraints.
container.Set(config)
// get current status of the container.
status, err := container.Status()
// get current container's state information.
state, err := container.State()
```
#### Checkpoint & Restore
libcontainer now integrates [CRIU](http://criu.org/) for checkpointing and restoring containers.
This let's you save the state of a process running inside a container to disk, and then restore
that state into a new process, on the same machine or on another machine.
`criu` version 1.5.2 or higher is required to use checkpoint and restore.
If you don't already have `criu` installed, you can build it from source, following the
[online instructions](http://criu.org/Installation). `criu` is also installed in the docker image
generated when building libcontainer with docker.
## Copyright and license
Code and documentation copyright 2014 Docker, inc.
The code and documentation are released under the [Apache 2.0 license](../LICENSE).
The documentation is also released under Creative Commons Attribution 4.0 International License.
You may obtain a copy of the license, titled CC-BY-4.0, at http://creativecommons.org/licenses/by/4.0/.

View file

@ -1,44 +0,0 @@
## nsenter
The `nsenter` package registers a special init constructor that is called before
the Go runtime has a chance to boot. This provides us the ability to `setns` on
existing namespaces and avoid the issues that the Go runtime has with multiple
threads. This constructor will be called if this package is registered,
imported, in your go application.
The `nsenter` package will `import "C"` and it uses [cgo](https://golang.org/cmd/cgo/)
package. In cgo, if the import of "C" is immediately preceded by a comment, that comment,
called the preamble, is used as a header when compiling the C parts of the package.
So every time we import package `nsenter`, the C code function `nsexec()` would be
called. And package `nsenter` is only imported in `init.go`, so every time the runc
`init` command is invoked, that C code is run.
Because `nsexec()` must be run before the Go runtime in order to use the
Linux kernel namespace, you must `import` this library into a package if
you plan to use `libcontainer` directly. Otherwise Go will not execute
the `nsexec()` constructor, which means that the re-exec will not cause
the namespaces to be joined. You can import it like this:
```go
import _ "github.com/opencontainers/runc/libcontainer/nsenter"
```
`nsexec()` will first get the file descriptor number for the init pipe
from the environment variable `_LIBCONTAINER_INITPIPE` (which was opened
by the parent and kept open across the fork-exec of the `nsexec()` init
process). The init pipe is used to read bootstrap data (namespace paths,
clone flags, uid and gid mappings, and the console path) from the parent
process. `nsexec()` will then call `setns(2)` to join the namespaces
provided in the bootstrap data (if available), `clone(2)` a child process
with the provided clone flags, update the user and group ID mappings, do
some further miscellaneous setup steps, and then send the PID of the
child process to the parent of the `nsexec()` "caller". Finally,
the parent `nsexec()` will exit and the child `nsexec()` process will
return to allow the Go runtime take over.
NOTE: We do both `setns(2)` and `clone(2)` even if we don't have any
`CLONE_NEW*` clone flags because we must fork a new process in order to
enter the PID namespace.

View file

@ -1,516 +0,0 @@
/*
* Copyright (C) 2019 Aleksa Sarai <cyphar@cyphar.com>
* Copyright (C) 2019 SUSE LLC
*
* 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.
*/
#define _GNU_SOURCE
#include <unistd.h>
#include <stdio.h>
#include <stdlib.h>
#include <stdbool.h>
#include <string.h>
#include <limits.h>
#include <fcntl.h>
#include <errno.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <sys/statfs.h>
#include <sys/vfs.h>
#include <sys/mman.h>
#include <sys/mount.h>
#include <sys/sendfile.h>
#include <sys/syscall.h>
/* Use our own wrapper for memfd_create. */
#if !defined(SYS_memfd_create) && defined(__NR_memfd_create)
# define SYS_memfd_create __NR_memfd_create
#endif
/* memfd_create(2) flags -- copied from <linux/memfd.h>. */
#ifndef MFD_CLOEXEC
# define MFD_CLOEXEC 0x0001U
# define MFD_ALLOW_SEALING 0x0002U
#endif
int memfd_create(const char *name, unsigned int flags)
{
#ifdef SYS_memfd_create
return syscall(SYS_memfd_create, name, flags);
#else
errno = ENOSYS;
return -1;
#endif
}
/* This comes directly from <linux/fcntl.h>. */
#ifndef F_LINUX_SPECIFIC_BASE
# define F_LINUX_SPECIFIC_BASE 1024
#endif
#ifndef F_ADD_SEALS
# define F_ADD_SEALS (F_LINUX_SPECIFIC_BASE + 9)
# define F_GET_SEALS (F_LINUX_SPECIFIC_BASE + 10)
#endif
#ifndef F_SEAL_SEAL
# define F_SEAL_SEAL 0x0001 /* prevent further seals from being set */
# define F_SEAL_SHRINK 0x0002 /* prevent file from shrinking */
# define F_SEAL_GROW 0x0004 /* prevent file from growing */
# define F_SEAL_WRITE 0x0008 /* prevent writes */
#endif
#define CLONED_BINARY_ENV "_LIBCONTAINER_CLONED_BINARY"
#define RUNC_MEMFD_COMMENT "runc_cloned:/proc/self/exe"
#define RUNC_MEMFD_SEALS \
(F_SEAL_SEAL | F_SEAL_SHRINK | F_SEAL_GROW | F_SEAL_WRITE)
static void *must_realloc(void *ptr, size_t size)
{
void *old = ptr;
do {
ptr = realloc(old, size);
} while(!ptr);
return ptr;
}
/*
* Verify whether we are currently in a self-cloned program (namely, is
* /proc/self/exe a memfd). F_GET_SEALS will only succeed for memfds (or rather
* for shmem files), and we want to be sure it's actually sealed.
*/
static int is_self_cloned(void)
{
int fd, ret, is_cloned = 0;
struct stat statbuf = {};
struct statfs fsbuf = {};
fd = open("/proc/self/exe", O_RDONLY|O_CLOEXEC);
if (fd < 0)
return -ENOTRECOVERABLE;
/*
* Is the binary a fully-sealed memfd? We don't need CLONED_BINARY_ENV for
* this, because you cannot write to a sealed memfd no matter what (so
* sharing it isn't a bad thing -- and an admin could bind-mount a sealed
* memfd to /usr/bin/runc to allow re-use).
*/
ret = fcntl(fd, F_GET_SEALS);
if (ret >= 0) {
is_cloned = (ret == RUNC_MEMFD_SEALS);
goto out;
}
/*
* All other forms require CLONED_BINARY_ENV, since they are potentially
* writeable (or we can't tell if they're fully safe) and thus we must
* check the environment as an extra layer of defence.
*/
if (!getenv(CLONED_BINARY_ENV)) {
is_cloned = false;
goto out;
}
/*
* Is the binary on a read-only filesystem? We can't detect bind-mounts in
* particular (in-kernel they are identical to regular mounts) but we can
* at least be sure that it's read-only. In addition, to make sure that
* it's *our* bind-mount we check CLONED_BINARY_ENV.
*/
if (fstatfs(fd, &fsbuf) >= 0)
is_cloned |= (fsbuf.f_flags & MS_RDONLY);
/*
* Okay, we're a tmpfile -- or we're currently running on RHEL <=7.6
* which appears to have a borked backport of F_GET_SEALS. Either way,
* having a file which has no hardlinks indicates that we aren't using
* a host-side "runc" binary and this is something that a container
* cannot fake (because unlinking requires being able to resolve the
* path that you want to unlink).
*/
if (fstat(fd, &statbuf) >= 0)
is_cloned |= (statbuf.st_nlink == 0);
out:
close(fd);
return is_cloned;
}
/* Read a given file into a new buffer, and providing the length. */
static char *read_file(char *path, size_t *length)
{
int fd;
char buf[4096], *copy = NULL;
if (!length)
return NULL;
fd = open(path, O_RDONLY | O_CLOEXEC);
if (fd < 0)
return NULL;
*length = 0;
for (;;) {
ssize_t n;
n = read(fd, buf, sizeof(buf));
if (n < 0)
goto error;
if (!n)
break;
copy = must_realloc(copy, (*length + n) * sizeof(*copy));
memcpy(copy + *length, buf, n);
*length += n;
}
close(fd);
return copy;
error:
close(fd);
free(copy);
return NULL;
}
/*
* A poor-man's version of "xargs -0". Basically parses a given block of
* NUL-delimited data, within the given length and adds a pointer to each entry
* to the array of pointers.
*/
static int parse_xargs(char *data, int data_length, char ***output)
{
int num = 0;
char *cur = data;
if (!data || *output != NULL)
return -1;
while (cur < data + data_length) {
num++;
*output = must_realloc(*output, (num + 1) * sizeof(**output));
(*output)[num - 1] = cur;
cur += strlen(cur) + 1;
}
(*output)[num] = NULL;
return num;
}
/*
* "Parse" out argv from /proc/self/cmdline.
* This is necessary because we are running in a context where we don't have a
* main() that we can just get the arguments from.
*/
static int fetchve(char ***argv)
{
char *cmdline = NULL;
size_t cmdline_size;
cmdline = read_file("/proc/self/cmdline", &cmdline_size);
if (!cmdline)
goto error;
if (parse_xargs(cmdline, cmdline_size, argv) <= 0)
goto error;
return 0;
error:
free(cmdline);
return -EINVAL;
}
enum {
EFD_NONE = 0,
EFD_MEMFD,
EFD_FILE,
};
/*
* This comes from <linux/fcntl.h>. We can't hard-code __O_TMPFILE because it
* changes depending on the architecture. If we don't have O_TMPFILE we always
* have the mkostemp(3) fallback.
*/
#ifndef O_TMPFILE
# if defined(__O_TMPFILE) && defined(O_DIRECTORY)
# define O_TMPFILE (__O_TMPFILE | O_DIRECTORY)
# endif
#endif
static int make_execfd(int *fdtype)
{
int fd = -1;
char template[PATH_MAX] = {0};
char *prefix = secure_getenv("_LIBCONTAINER_STATEDIR");
if (!prefix || *prefix != '/')
prefix = "/tmp";
if (snprintf(template, sizeof(template), "%s/runc.XXXXXX", prefix) < 0)
return -1;
/*
* Now try memfd, it's much nicer than actually creating a file in STATEDIR
* since it's easily detected thanks to sealing and also doesn't require
* assumptions about STATEDIR.
*/
*fdtype = EFD_MEMFD;
fd = memfd_create(RUNC_MEMFD_COMMENT, MFD_CLOEXEC | MFD_ALLOW_SEALING);
if (fd >= 0)
return fd;
if (errno != ENOSYS && errno != EINVAL)
goto error;
#ifdef O_TMPFILE
/*
* Try O_TMPFILE to avoid races where someone might snatch our file. Note
* that O_EXCL isn't actually a security measure here (since you can just
* fd re-open it and clear O_EXCL).
*/
*fdtype = EFD_FILE;
fd = open(prefix, O_TMPFILE | O_EXCL | O_RDWR | O_CLOEXEC, 0700);
if (fd >= 0) {
struct stat statbuf = {};
bool working_otmpfile = false;
/*
* open(2) ignores unknown O_* flags -- yeah, I was surprised when I
* found this out too. As a result we can't check for EINVAL. However,
* if we get nlink != 0 (or EISDIR) then we know that this kernel
* doesn't support O_TMPFILE.
*/
if (fstat(fd, &statbuf) >= 0)
working_otmpfile = (statbuf.st_nlink == 0);
if (working_otmpfile)
return fd;
/* Pretend that we got EISDIR since O_TMPFILE failed. */
close(fd);
errno = EISDIR;
}
if (errno != EISDIR)
goto error;
#endif /* defined(O_TMPFILE) */
/*
* Our final option is to create a temporary file the old-school way, and
* then unlink it so that nothing else sees it by accident.
*/
*fdtype = EFD_FILE;
fd = mkostemp(template, O_CLOEXEC);
if (fd >= 0) {
if (unlink(template) >= 0)
return fd;
close(fd);
}
error:
*fdtype = EFD_NONE;
return -1;
}
static int seal_execfd(int *fd, int fdtype)
{
switch (fdtype) {
case EFD_MEMFD:
return fcntl(*fd, F_ADD_SEALS, RUNC_MEMFD_SEALS);
case EFD_FILE: {
/* Need to re-open our pseudo-memfd as an O_PATH to avoid execve(2) giving -ETXTBSY. */
int newfd;
char fdpath[PATH_MAX] = {0};
if (fchmod(*fd, 0100) < 0)
return -1;
if (snprintf(fdpath, sizeof(fdpath), "/proc/self/fd/%d", *fd) < 0)
return -1;
newfd = open(fdpath, O_PATH | O_CLOEXEC);
if (newfd < 0)
return -1;
close(*fd);
*fd = newfd;
return 0;
}
default:
break;
}
return -1;
}
static int try_bindfd(void)
{
int fd, ret = -1;
char template[PATH_MAX] = {0};
char *prefix = secure_getenv("_LIBCONTAINER_STATEDIR");
if (!prefix || *prefix != '/')
prefix = "/tmp";
if (snprintf(template, sizeof(template), "%s/runc.XXXXXX", prefix) < 0)
return ret;
/*
* We need somewhere to mount it, mounting anything over /proc/self is a
* BAD idea on the host -- even if we do it temporarily.
*/
fd = mkstemp(template);
if (fd < 0)
return ret;
close(fd);
/*
* For obvious reasons this won't work in rootless mode because we haven't
* created a userns+mntns -- but getting that to work will be a bit
* complicated and it's only worth doing if someone actually needs it.
*/
ret = -EPERM;
if (mount("/proc/self/exe", template, "", MS_BIND, "") < 0)
goto out;
if (mount("", template, "", MS_REMOUNT | MS_BIND | MS_RDONLY, "") < 0)
goto out_umount;
/* Get read-only handle that we're sure can't be made read-write. */
ret = open(template, O_PATH | O_CLOEXEC);
out_umount:
/*
* Make sure the MNT_DETACH works, otherwise we could get remounted
* read-write and that would be quite bad (the fd would be made read-write
* too, invalidating the protection).
*/
if (umount2(template, MNT_DETACH) < 0) {
if (ret >= 0)
close(ret);
ret = -ENOTRECOVERABLE;
}
out:
/*
* We don't care about unlink errors, the worst that happens is that
* there's an empty file left around in STATEDIR.
*/
unlink(template);
return ret;
}
static ssize_t fd_to_fd(int outfd, int infd)
{
ssize_t total = 0;
char buffer[4096];
for (;;) {
ssize_t nread, nwritten = 0;
nread = read(infd, buffer, sizeof(buffer));
if (nread < 0)
return -1;
if (!nread)
break;
do {
ssize_t n = write(outfd, buffer + nwritten, nread - nwritten);
if (n < 0)
return -1;
nwritten += n;
} while(nwritten < nread);
total += nwritten;
}
return total;
}
static int clone_binary(void)
{
int binfd, execfd;
struct stat statbuf = {};
size_t sent = 0;
int fdtype = EFD_NONE;
/*
* Before we resort to copying, let's try creating an ro-binfd in one shot
* by getting a handle for a read-only bind-mount of the execfd.
*/
execfd = try_bindfd();
if (execfd >= 0)
return execfd;
/*
* Dammit, that didn't work -- time to copy the binary to a safe place we
* can seal the contents.
*/
execfd = make_execfd(&fdtype);
if (execfd < 0 || fdtype == EFD_NONE)
return -ENOTRECOVERABLE;
binfd = open("/proc/self/exe", O_RDONLY | O_CLOEXEC);
if (binfd < 0)
goto error;
if (fstat(binfd, &statbuf) < 0)
goto error_binfd;
while (sent < statbuf.st_size) {
int n = sendfile(execfd, binfd, NULL, statbuf.st_size - sent);
if (n < 0) {
/* sendfile can fail so we fallback to a dumb user-space copy. */
n = fd_to_fd(execfd, binfd);
if (n < 0)
goto error_binfd;
}
sent += n;
}
close(binfd);
if (sent != statbuf.st_size)
goto error;
if (seal_execfd(&execfd, fdtype) < 0)
goto error;
return execfd;
error_binfd:
close(binfd);
error:
close(execfd);
return -EIO;
}
/* Get cheap access to the environment. */
extern char **environ;
int ensure_cloned_binary(void)
{
int execfd;
char **argv = NULL;
/* Check that we're not self-cloned, and if we are then bail. */
int cloned = is_self_cloned();
if (cloned > 0 || cloned == -ENOTRECOVERABLE)
return cloned;
if (fetchve(&argv) < 0)
return -EINVAL;
execfd = clone_binary();
if (execfd < 0)
return -EIO;
if (putenv(CLONED_BINARY_ENV "=1"))
goto error;
fexecve(execfd, argv, environ);
error:
close(execfd);
return -ENOEXEC;
}

View file

@ -1,32 +0,0 @@
#ifndef NSENTER_NAMESPACE_H
#define NSENTER_NAMESPACE_H
#ifndef _GNU_SOURCE
# define _GNU_SOURCE
#endif
#include <sched.h>
/* All of these are taken from include/uapi/linux/sched.h */
#ifndef CLONE_NEWNS
# define CLONE_NEWNS 0x00020000 /* New mount namespace group */
#endif
#ifndef CLONE_NEWCGROUP
# define CLONE_NEWCGROUP 0x02000000 /* New cgroup namespace */
#endif
#ifndef CLONE_NEWUTS
# define CLONE_NEWUTS 0x04000000 /* New utsname namespace */
#endif
#ifndef CLONE_NEWIPC
# define CLONE_NEWIPC 0x08000000 /* New ipc namespace */
#endif
#ifndef CLONE_NEWUSER
# define CLONE_NEWUSER 0x10000000 /* New user namespace */
#endif
#ifndef CLONE_NEWPID
# define CLONE_NEWPID 0x20000000 /* New pid namespace */
#endif
#ifndef CLONE_NEWNET
# define CLONE_NEWNET 0x40000000 /* New network namespace */
#endif
#endif /* NSENTER_NAMESPACE_H */

View file

@ -1,12 +0,0 @@
// +build linux,!gccgo
package nsenter
/*
#cgo CFLAGS: -Wall
extern void nsexec();
void __attribute__((constructor)) init(void) {
nsexec();
}
*/
import "C"

View file

@ -1,25 +0,0 @@
// +build linux,gccgo
package nsenter
/*
#cgo CFLAGS: -Wall
extern void nsexec();
void __attribute__((constructor)) init(void) {
nsexec();
}
*/
import "C"
// AlwaysFalse is here to stay false
// (and be exported so the compiler doesn't optimize out its reference)
var AlwaysFalse bool
func init() {
if AlwaysFalse {
// by referencing this C init() in a noop test, it will ensure the compiler
// links in the C function.
// https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65134
C.init()
}
}

View file

@ -1,5 +0,0 @@
// +build !linux !cgo
package nsenter
import "C"

File diff suppressed because it is too large Load diff

View file

@ -1,41 +0,0 @@
package user
import (
"errors"
)
var (
// The current operating system does not provide the required data for user lookups.
ErrUnsupported = errors.New("user lookup: operating system does not provide passwd-formatted data")
// No matching entries found in file.
ErrNoPasswdEntries = errors.New("no matching entries in passwd file")
ErrNoGroupEntries = errors.New("no matching entries in group file")
)
// LookupUser looks up a user by their username in /etc/passwd. If the user
// cannot be found (or there is no /etc/passwd file on the filesystem), then
// LookupUser returns an error.
func LookupUser(username string) (User, error) {
return lookupUser(username)
}
// LookupUid looks up a user by their user id in /etc/passwd. If the user cannot
// be found (or there is no /etc/passwd file on the filesystem), then LookupId
// returns an error.
func LookupUid(uid int) (User, error) {
return lookupUid(uid)
}
// LookupGroup looks up a group by its name in /etc/group. If the group cannot
// be found (or there is no /etc/group file on the filesystem), then LookupGroup
// returns an error.
func LookupGroup(groupname string) (Group, error) {
return lookupGroup(groupname)
}
// LookupGid looks up a group by its group id in /etc/group. If the group cannot
// be found (or there is no /etc/group file on the filesystem), then LookupGid
// returns an error.
func LookupGid(gid int) (Group, error) {
return lookupGid(gid)
}

View file

@ -1,144 +0,0 @@
// +build darwin dragonfly freebsd linux netbsd openbsd solaris
package user
import (
"io"
"os"
"strconv"
"golang.org/x/sys/unix"
)
// Unix-specific path to the passwd and group formatted files.
const (
unixPasswdPath = "/etc/passwd"
unixGroupPath = "/etc/group"
)
func lookupUser(username string) (User, error) {
return lookupUserFunc(func(u User) bool {
return u.Name == username
})
}
func lookupUid(uid int) (User, error) {
return lookupUserFunc(func(u User) bool {
return u.Uid == uid
})
}
func lookupUserFunc(filter func(u User) bool) (User, error) {
// Get operating system-specific passwd reader-closer.
passwd, err := GetPasswd()
if err != nil {
return User{}, err
}
defer passwd.Close()
// Get the users.
users, err := ParsePasswdFilter(passwd, filter)
if err != nil {
return User{}, err
}
// No user entries found.
if len(users) == 0 {
return User{}, ErrNoPasswdEntries
}
// Assume the first entry is the "correct" one.
return users[0], nil
}
func lookupGroup(groupname string) (Group, error) {
return lookupGroupFunc(func(g Group) bool {
return g.Name == groupname
})
}
func lookupGid(gid int) (Group, error) {
return lookupGroupFunc(func(g Group) bool {
return g.Gid == gid
})
}
func lookupGroupFunc(filter func(g Group) bool) (Group, error) {
// Get operating system-specific group reader-closer.
group, err := GetGroup()
if err != nil {
return Group{}, err
}
defer group.Close()
// Get the users.
groups, err := ParseGroupFilter(group, filter)
if err != nil {
return Group{}, err
}
// No user entries found.
if len(groups) == 0 {
return Group{}, ErrNoGroupEntries
}
// Assume the first entry is the "correct" one.
return groups[0], nil
}
func GetPasswdPath() (string, error) {
return unixPasswdPath, nil
}
func GetPasswd() (io.ReadCloser, error) {
return os.Open(unixPasswdPath)
}
func GetGroupPath() (string, error) {
return unixGroupPath, nil
}
func GetGroup() (io.ReadCloser, error) {
return os.Open(unixGroupPath)
}
// CurrentUser looks up the current user by their user id in /etc/passwd. If the
// user cannot be found (or there is no /etc/passwd file on the filesystem),
// then CurrentUser returns an error.
func CurrentUser() (User, error) {
return LookupUid(unix.Getuid())
}
// CurrentGroup looks up the current user's group by their primary group id's
// entry in /etc/passwd. If the group cannot be found (or there is no
// /etc/group file on the filesystem), then CurrentGroup returns an error.
func CurrentGroup() (Group, error) {
return LookupGid(unix.Getgid())
}
func currentUserSubIDs(fileName string) ([]SubID, error) {
u, err := CurrentUser()
if err != nil {
return nil, err
}
filter := func(entry SubID) bool {
return entry.Name == u.Name || entry.Name == strconv.Itoa(u.Uid)
}
return ParseSubIDFileFilter(fileName, filter)
}
func CurrentUserSubUIDs() ([]SubID, error) {
return currentUserSubIDs("/etc/subuid")
}
func CurrentUserSubGIDs() ([]SubID, error) {
return currentUserSubIDs("/etc/subgid")
}
func CurrentProcessUIDMap() ([]IDMap, error) {
return ParseIDMapFile("/proc/self/uid_map")
}
func CurrentProcessGIDMap() ([]IDMap, error) {
return ParseIDMapFile("/proc/self/gid_map")
}

View file

@ -1,40 +0,0 @@
// +build windows
package user
import (
"fmt"
"os/user"
)
func lookupUser(username string) (User, error) {
u, err := user.Lookup(username)
if err != nil {
return User{}, err
}
return userFromOS(u)
}
func lookupUid(uid int) (User, error) {
u, err := user.LookupId(fmt.Sprintf("%d", uid))
if err != nil {
return User{}, err
}
return userFromOS(u)
}
func lookupGroup(groupname string) (Group, error) {
g, err := user.LookupGroup(groupname)
if err != nil {
return Group{}, err
}
return groupFromOS(g)
}
func lookupGid(gid int) (Group, error) {
g, err := user.LookupGroupId(fmt.Sprintf("%d", gid))
if err != nil {
return Group{}, err
}
return groupFromOS(g)
}

View file

@ -1,608 +0,0 @@
package user
import (
"bufio"
"fmt"
"io"
"os"
"os/user"
"strconv"
"strings"
)
const (
minId = 0
maxId = 1<<31 - 1 //for 32-bit systems compatibility
)
var (
ErrRange = fmt.Errorf("uids and gids must be in range %d-%d", minId, maxId)
)
type User struct {
Name string
Pass string
Uid int
Gid int
Gecos string
Home string
Shell string
}
// userFromOS converts an os/user.(*User) to local User
//
// (This does not include Pass, Shell or Gecos)
func userFromOS(u *user.User) (User, error) {
newUser := User{
Name: u.Username,
Home: u.HomeDir,
}
id, err := strconv.Atoi(u.Uid)
if err != nil {
return newUser, err
}
newUser.Uid = id
id, err = strconv.Atoi(u.Gid)
if err != nil {
return newUser, err
}
newUser.Gid = id
return newUser, nil
}
type Group struct {
Name string
Pass string
Gid int
List []string
}
// groupFromOS converts an os/user.(*Group) to local Group
//
// (This does not include Pass, Shell or Gecos)
func groupFromOS(g *user.Group) (Group, error) {
newGroup := Group{
Name: g.Name,
}
id, err := strconv.Atoi(g.Gid)
if err != nil {
return newGroup, err
}
newGroup.Gid = id
return newGroup, nil
}
// SubID represents an entry in /etc/sub{u,g}id
type SubID struct {
Name string
SubID int64
Count int64
}
// IDMap represents an entry in /proc/PID/{u,g}id_map
type IDMap struct {
ID int64
ParentID int64
Count int64
}
func parseLine(line string, v ...interface{}) {
parseParts(strings.Split(line, ":"), v...)
}
func parseParts(parts []string, v ...interface{}) {
if len(parts) == 0 {
return
}
for i, p := range parts {
// Ignore cases where we don't have enough fields to populate the arguments.
// Some configuration files like to misbehave.
if len(v) <= i {
break
}
// Use the type of the argument to figure out how to parse it, scanf() style.
// This is legit.
switch e := v[i].(type) {
case *string:
*e = p
case *int:
// "numbers", with conversion errors ignored because of some misbehaving configuration files.
*e, _ = strconv.Atoi(p)
case *int64:
*e, _ = strconv.ParseInt(p, 10, 64)
case *[]string:
// Comma-separated lists.
if p != "" {
*e = strings.Split(p, ",")
} else {
*e = []string{}
}
default:
// Someone goof'd when writing code using this function. Scream so they can hear us.
panic(fmt.Sprintf("parseLine only accepts {*string, *int, *int64, *[]string} as arguments! %#v is not a pointer!", e))
}
}
}
func ParsePasswdFile(path string) ([]User, error) {
passwd, err := os.Open(path)
if err != nil {
return nil, err
}
defer passwd.Close()
return ParsePasswd(passwd)
}
func ParsePasswd(passwd io.Reader) ([]User, error) {
return ParsePasswdFilter(passwd, nil)
}
func ParsePasswdFileFilter(path string, filter func(User) bool) ([]User, error) {
passwd, err := os.Open(path)
if err != nil {
return nil, err
}
defer passwd.Close()
return ParsePasswdFilter(passwd, filter)
}
func ParsePasswdFilter(r io.Reader, filter func(User) bool) ([]User, error) {
if r == nil {
return nil, fmt.Errorf("nil source for passwd-formatted data")
}
var (
s = bufio.NewScanner(r)
out = []User{}
)
for s.Scan() {
if err := s.Err(); err != nil {
return nil, err
}
line := strings.TrimSpace(s.Text())
if line == "" {
continue
}
// see: man 5 passwd
// name:password:UID:GID:GECOS:directory:shell
// Name:Pass:Uid:Gid:Gecos:Home:Shell
// root:x:0:0:root:/root:/bin/bash
// adm:x:3:4:adm:/var/adm:/bin/false
p := User{}
parseLine(line, &p.Name, &p.Pass, &p.Uid, &p.Gid, &p.Gecos, &p.Home, &p.Shell)
if filter == nil || filter(p) {
out = append(out, p)
}
}
return out, nil
}
func ParseGroupFile(path string) ([]Group, error) {
group, err := os.Open(path)
if err != nil {
return nil, err
}
defer group.Close()
return ParseGroup(group)
}
func ParseGroup(group io.Reader) ([]Group, error) {
return ParseGroupFilter(group, nil)
}
func ParseGroupFileFilter(path string, filter func(Group) bool) ([]Group, error) {
group, err := os.Open(path)
if err != nil {
return nil, err
}
defer group.Close()
return ParseGroupFilter(group, filter)
}
func ParseGroupFilter(r io.Reader, filter func(Group) bool) ([]Group, error) {
if r == nil {
return nil, fmt.Errorf("nil source for group-formatted data")
}
var (
s = bufio.NewScanner(r)
out = []Group{}
)
for s.Scan() {
if err := s.Err(); err != nil {
return nil, err
}
text := s.Text()
if text == "" {
continue
}
// see: man 5 group
// group_name:password:GID:user_list
// Name:Pass:Gid:List
// root:x:0:root
// adm:x:4:root,adm,daemon
p := Group{}
parseLine(text, &p.Name, &p.Pass, &p.Gid, &p.List)
if filter == nil || filter(p) {
out = append(out, p)
}
}
return out, nil
}
type ExecUser struct {
Uid int
Gid int
Sgids []int
Home string
}
// GetExecUserPath is a wrapper for GetExecUser. It reads data from each of the
// given file paths and uses that data as the arguments to GetExecUser. If the
// files cannot be opened for any reason, the error is ignored and a nil
// io.Reader is passed instead.
func GetExecUserPath(userSpec string, defaults *ExecUser, passwdPath, groupPath string) (*ExecUser, error) {
var passwd, group io.Reader
if passwdFile, err := os.Open(passwdPath); err == nil {
passwd = passwdFile
defer passwdFile.Close()
}
if groupFile, err := os.Open(groupPath); err == nil {
group = groupFile
defer groupFile.Close()
}
return GetExecUser(userSpec, defaults, passwd, group)
}
// GetExecUser parses a user specification string (using the passwd and group
// readers as sources for /etc/passwd and /etc/group data, respectively). In
// the case of blank fields or missing data from the sources, the values in
// defaults is used.
//
// GetExecUser will return an error if a user or group literal could not be
// found in any entry in passwd and group respectively.
//
// Examples of valid user specifications are:
// * ""
// * "user"
// * "uid"
// * "user:group"
// * "uid:gid
// * "user:gid"
// * "uid:group"
//
// It should be noted that if you specify a numeric user or group id, they will
// not be evaluated as usernames (only the metadata will be filled). So attempting
// to parse a user with user.Name = "1337" will produce the user with a UID of
// 1337.
func GetExecUser(userSpec string, defaults *ExecUser, passwd, group io.Reader) (*ExecUser, error) {
if defaults == nil {
defaults = new(ExecUser)
}
// Copy over defaults.
user := &ExecUser{
Uid: defaults.Uid,
Gid: defaults.Gid,
Sgids: defaults.Sgids,
Home: defaults.Home,
}
// Sgids slice *cannot* be nil.
if user.Sgids == nil {
user.Sgids = []int{}
}
// Allow for userArg to have either "user" syntax, or optionally "user:group" syntax
var userArg, groupArg string
parseLine(userSpec, &userArg, &groupArg)
// Convert userArg and groupArg to be numeric, so we don't have to execute
// Atoi *twice* for each iteration over lines.
uidArg, uidErr := strconv.Atoi(userArg)
gidArg, gidErr := strconv.Atoi(groupArg)
// Find the matching user.
users, err := ParsePasswdFilter(passwd, func(u User) bool {
if userArg == "" {
// Default to current state of the user.
return u.Uid == user.Uid
}
if uidErr == nil {
// If the userArg is numeric, always treat it as a UID.
return uidArg == u.Uid
}
return u.Name == userArg
})
// If we can't find the user, we have to bail.
if err != nil && passwd != nil {
if userArg == "" {
userArg = strconv.Itoa(user.Uid)
}
return nil, fmt.Errorf("unable to find user %s: %v", userArg, err)
}
var matchedUserName string
if len(users) > 0 {
// First match wins, even if there's more than one matching entry.
matchedUserName = users[0].Name
user.Uid = users[0].Uid
user.Gid = users[0].Gid
user.Home = users[0].Home
} else if userArg != "" {
// If we can't find a user with the given username, the only other valid
// option is if it's a numeric username with no associated entry in passwd.
if uidErr != nil {
// Not numeric.
return nil, fmt.Errorf("unable to find user %s: %v", userArg, ErrNoPasswdEntries)
}
user.Uid = uidArg
// Must be inside valid uid range.
if user.Uid < minId || user.Uid > maxId {
return nil, ErrRange
}
// Okay, so it's numeric. We can just roll with this.
}
// On to the groups. If we matched a username, we need to do this because of
// the supplementary group IDs.
if groupArg != "" || matchedUserName != "" {
groups, err := ParseGroupFilter(group, func(g Group) bool {
// If the group argument isn't explicit, we'll just search for it.
if groupArg == "" {
// Check if user is a member of this group.
for _, u := range g.List {
if u == matchedUserName {
return true
}
}
return false
}
if gidErr == nil {
// If the groupArg is numeric, always treat it as a GID.
return gidArg == g.Gid
}
return g.Name == groupArg
})
if err != nil && group != nil {
return nil, fmt.Errorf("unable to find groups for spec %v: %v", matchedUserName, err)
}
// Only start modifying user.Gid if it is in explicit form.
if groupArg != "" {
if len(groups) > 0 {
// First match wins, even if there's more than one matching entry.
user.Gid = groups[0].Gid
} else {
// If we can't find a group with the given name, the only other valid
// option is if it's a numeric group name with no associated entry in group.
if gidErr != nil {
// Not numeric.
return nil, fmt.Errorf("unable to find group %s: %v", groupArg, ErrNoGroupEntries)
}
user.Gid = gidArg
// Must be inside valid gid range.
if user.Gid < minId || user.Gid > maxId {
return nil, ErrRange
}
// Okay, so it's numeric. We can just roll with this.
}
} else if len(groups) > 0 {
// Supplementary group ids only make sense if in the implicit form.
user.Sgids = make([]int, len(groups))
for i, group := range groups {
user.Sgids[i] = group.Gid
}
}
}
return user, nil
}
// GetAdditionalGroups looks up a list of groups by name or group id
// against the given /etc/group formatted data. If a group name cannot
// be found, an error will be returned. If a group id cannot be found,
// or the given group data is nil, the id will be returned as-is
// provided it is in the legal range.
func GetAdditionalGroups(additionalGroups []string, group io.Reader) ([]int, error) {
var groups = []Group{}
if group != nil {
var err error
groups, err = ParseGroupFilter(group, func(g Group) bool {
for _, ag := range additionalGroups {
if g.Name == ag || strconv.Itoa(g.Gid) == ag {
return true
}
}
return false
})
if err != nil {
return nil, fmt.Errorf("Unable to find additional groups %v: %v", additionalGroups, err)
}
}
gidMap := make(map[int]struct{})
for _, ag := range additionalGroups {
var found bool
for _, g := range groups {
// if we found a matched group either by name or gid, take the
// first matched as correct
if g.Name == ag || strconv.Itoa(g.Gid) == ag {
if _, ok := gidMap[g.Gid]; !ok {
gidMap[g.Gid] = struct{}{}
found = true
break
}
}
}
// we asked for a group but didn't find it. let's check to see
// if we wanted a numeric group
if !found {
gid, err := strconv.Atoi(ag)
if err != nil {
return nil, fmt.Errorf("Unable to find group %s", ag)
}
// Ensure gid is inside gid range.
if gid < minId || gid > maxId {
return nil, ErrRange
}
gidMap[gid] = struct{}{}
}
}
gids := []int{}
for gid := range gidMap {
gids = append(gids, gid)
}
return gids, nil
}
// GetAdditionalGroupsPath is a wrapper around GetAdditionalGroups
// that opens the groupPath given and gives it as an argument to
// GetAdditionalGroups.
func GetAdditionalGroupsPath(additionalGroups []string, groupPath string) ([]int, error) {
var group io.Reader
if groupFile, err := os.Open(groupPath); err == nil {
group = groupFile
defer groupFile.Close()
}
return GetAdditionalGroups(additionalGroups, group)
}
func ParseSubIDFile(path string) ([]SubID, error) {
subid, err := os.Open(path)
if err != nil {
return nil, err
}
defer subid.Close()
return ParseSubID(subid)
}
func ParseSubID(subid io.Reader) ([]SubID, error) {
return ParseSubIDFilter(subid, nil)
}
func ParseSubIDFileFilter(path string, filter func(SubID) bool) ([]SubID, error) {
subid, err := os.Open(path)
if err != nil {
return nil, err
}
defer subid.Close()
return ParseSubIDFilter(subid, filter)
}
func ParseSubIDFilter(r io.Reader, filter func(SubID) bool) ([]SubID, error) {
if r == nil {
return nil, fmt.Errorf("nil source for subid-formatted data")
}
var (
s = bufio.NewScanner(r)
out = []SubID{}
)
for s.Scan() {
if err := s.Err(); err != nil {
return nil, err
}
line := strings.TrimSpace(s.Text())
if line == "" {
continue
}
// see: man 5 subuid
p := SubID{}
parseLine(line, &p.Name, &p.SubID, &p.Count)
if filter == nil || filter(p) {
out = append(out, p)
}
}
return out, nil
}
func ParseIDMapFile(path string) ([]IDMap, error) {
r, err := os.Open(path)
if err != nil {
return nil, err
}
defer r.Close()
return ParseIDMap(r)
}
func ParseIDMap(r io.Reader) ([]IDMap, error) {
return ParseIDMapFilter(r, nil)
}
func ParseIDMapFileFilter(path string, filter func(IDMap) bool) ([]IDMap, error) {
r, err := os.Open(path)
if err != nil {
return nil, err
}
defer r.Close()
return ParseIDMapFilter(r, filter)
}
func ParseIDMapFilter(r io.Reader, filter func(IDMap) bool) ([]IDMap, error) {
if r == nil {
return nil, fmt.Errorf("nil source for idmap-formatted data")
}
var (
s = bufio.NewScanner(r)
out = []IDMap{}
)
for s.Scan() {
if err := s.Err(); err != nil {
return nil, err
}
line := strings.TrimSpace(s.Text())
if line == "" {
continue
}
// see: man 7 user_namespaces
p := IDMap{}
parseParts(strings.Fields(line), &p.ID, &p.ParentID, &p.Count)
if filter == nil || filter(p) {
out = append(out, p)
}
}
return out, nil
}

View file

@ -1,26 +0,0 @@
# OCI runtime-spec. When updating this, make sure you use a version tag rather
# than a commit ID so it's much more obvious what version of the spec we are
# using.
github.com/opencontainers/runtime-spec 29686dbc5559d93fb1ef402eeda3e35c38d75af4
# Core libcontainer functionality.
github.com/checkpoint-restore/go-criu v3.11
github.com/mrunalp/fileutils ed869b029674c0e9ce4c0dfa781405c2d9946d08
github.com/opencontainers/selinux v1.0.0-rc1
github.com/seccomp/libseccomp-golang 84e90a91acea0f4e51e62bc1a75de18b1fc0790f
github.com/sirupsen/logrus a3f95b5c423586578a4e099b11a46c2479628cac
github.com/syndtr/gocapability db04d3cc01c8b54962a58ec7e491717d06cfcc16
github.com/vishvananda/netlink 1e2e08e8a2dcdacaae3f14ac44c5cfa31361f270
# systemd integration.
github.com/coreos/go-systemd v14
github.com/coreos/pkg v3
github.com/godbus/dbus v3
github.com/golang/protobuf 18c9bb3261723cd5401db4d0c9fbc5c3b6c70fe8
# Command-line interface.
github.com/cyphar/filepath-securejoin v0.2.1
github.com/docker/go-units v0.2.0
github.com/urfave/cli d53eb991652b1d438abdd34ce4bfa3ef1539108e
golang.org/x/sys 41f3e6584952bb034a481797859f6ab34b6803bd https://github.com/golang/sys
# console dependencies
github.com/containerd/console 2748ece16665b45a47f884001d5831ec79703880
github.com/pkg/errors v0.8.0