zishuo/runc
1
0
Fork 0
mirror of https://github.com/opencontainers/runc.git synced 2025-09-26 19:41:35 +08:00
Code Issues Packages Projects Releases Wiki Activity

Initial commit of runc binary

Browse Source 
Signed-off-by: Michael Crosby <crosbymichael@gmail.com>
This commit is contained in:
Michael Crosby
2015-06-21 19:31:12 -07:00
parent 080df7ab88
commit 9fac183294
14 changed files with 1655 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
.gitignore vendored
View File

@@ -1,3 +1,4 @@
bundles bundles
nsinit/nsinit nsinit/nsinit
vendor/pkg vendor/pkg
runc

191
LICENSE Normal file
View File

@@ -0,0 +1,191 @@
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.

9
Makefile Normal file
View File

@@ -0,0 +1,9 @@
all:
go build -o runc .
install:
cp runc /usr/local/bin/runc
rm runc
clean:
rm runc

145
README.md Normal file
View File

@@ -0,0 +1,145 @@
## runc
`runc` is a CLI tool for spawning and running containers according to the OCF specification.
### Building:
```bash
git clone https://github.com/opencontainers/runc
make
sudo make install
```
### Using:
To run a container that you received just execute `runc run` with the JSON format at the argument or have a
`container.json` file in the current working directory.
```bash
runc
/ $ ps
PID USER COMMAND
1 daemon sh
5 daemon sh
/ $
```
### OCF Container JSON Format;
```json
{
"version": "0.1",
"os": "linux",
"arch": "amd64",
"processes": [
{
"tty": true,
"user": "daemon",
"args": [
"sh"
],
"env": [
"PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin",
"TERM=xterm"
],
"cwd": ""
}
],
"root": {
"path": "rootfs",
"readonly": true
},
"cpus": 1.1,
"memory": 1024,
"hostname": "shell",
"namespaces": [
{
"type": "process"
},
{
"type": "network"
},
{
"type": "mount"
},
{
"type": "ipc"
},
{
"type": "uts"
}
],
"capabilities": [
"AUDIT_WRITE",
"KILL",
"NET_BIND_SERVICE"
],
"devices": [
"null",
"random",
"full",
"tty",
"zero",
"urandom"
],
"mounts": [
{
"type": "proc",
"source": "proc",
"destination": "/proc",
"options": ""
},
{
"type": "tmpfs",
"source": "tmpfs",
"destination": "/dev",
"options": "nosuid,strictatime,mode=755,size=65536k"
},
{
"type": "devpts",
"source": "devpts",
"destination": "/dev/pts",
"options": "nosuid,noexec,newinstance,ptmxmode=0666,mode=0620,gid=5"
},
{
"type": "tmpfs",
"source": "shm",
"destination": "/dev/shm",
"options": "nosuid,noexec,nodev,mode=1777,size=65536k"
},
{
"type": "mqueue",
"source": "mqueue",
"destination": "/dev/mqueue",
"options": "nosuid,noexec,nodev"
},
{
"type": "sysfs",
"source": "sysfs",
"destination": "/sys",
"options": "nosuid,noexec,nodev"
}
]
}
```
### Examples:
#### Using runc with systemd
```service
[Unit]
Description=Minecraft Build Server
Documentation=http://minecraft.net
After=network.target
[Service]
CPUQuota=200%
MemoryLimit=1536M
ExecStart=/usr/local/bin/runc
Restart=on-failure
WorkingDirectory=/containers/minecraftbuild
[Install]
WantedBy=multi-user.target
```

63
checkpoint.go Normal file
View File

@@ -0,0 +1,63 @@
package main
import (
"fmt"
"strconv"
"strings"
"github.com/codegangsta/cli"
"github.com/opencontainers/runc/libcontainer"
)
var checkpointCommand = cli.Command{
Name: "checkpoint",
Usage: "checkpoint a running container",
Flags: []cli.Flag{
cli.StringFlag{Name: "image-path", Value: "", Usage: "path for saving criu image files"},
cli.StringFlag{Name: "work-path", Value: "", Usage: "path for saving work files and logs"},
cli.BoolFlag{Name: "leave-running", Usage: "leave the process running after checkpointing"},
cli.BoolFlag{Name: "tcp-established", Usage: "allow open tcp connections"},
cli.BoolFlag{Name: "ext-unix-sk", Usage: "allow external unix sockets"},
cli.BoolFlag{Name: "shell-job", Usage: "allow shell jobs"},
cli.StringFlag{Name: "page-server", Value: "", Usage: "ADDRESS:PORT of the page server"},
},
Action: func(context *cli.Context) {
container, err := getContainer(context)
if err != nil {
fatal(err)
}
options := criuOptions(context)
// these are the mandatory criu options for a container
setPageServer(context, options)
if err := container.Checkpoint(options); err != nil {
fatal(err)
}
},
}
func getCheckpointImagePath(context *cli.Context) string {
imagePath := context.String("image-path")
if imagePath == "" {
imagePath = getDefaultImagePath(context)
}
return imagePath
}
func setPageServer(context *cli.Context, options *libcontainer.CriuOpts) {
// xxx following criu opts are optional
// The dump image can be sent to a criu page server
if psOpt := context.String("page-server"); psOpt != "" {
addressPort := strings.Split(psOpt, ":")
if len(addressPort) != 2 {
fatal(fmt.Errorf("Use --page-server ADDRESS:PORT to specify page server"))
}
port_int, err := strconv.Atoi(addressPort[1])
if err != nil {
fatal(fmt.Errorf("Invalid port number"))
}
options.PageServer = libcontainer.CriuPageServerInfo{
Address: addressPort[0],
Port: int32(port_int),
}
}
}

93
events.go Normal file
View File

@@ -0,0 +1,93 @@
package main
import (
"encoding/json"
"os"
"sync"
"time"
"github.com/Sirupsen/logrus"
"github.com/codegangsta/cli"
"github.com/opencontainers/runc/libcontainer"
)
// event struct for encoding the event data to json.
type event struct {
Type string `json:"type"`
ID string `json:"id"`
Data interface{} `json:"data,omitempty"`
}
var eventsCommand = cli.Command{
Name: "events",
Usage: "display container events such as OOM notifications and cpu, memeory, IO, and network stats",
Flags: []cli.Flag{
cli.DurationFlag{Name: "interval", Value: 5 * time.Second, Usage: "set the stats collection interval"},
cli.BoolFlag{Name: "stats", Usage: "display the container's stats then exit"},
},
Action: func(context *cli.Context) {
container, err := getContainer(context)
if err != nil {
logrus.Fatal(err)
}
var (
stats = make(chan *libcontainer.Stats, 1)
events = make(chan *event, 1024)
group = &sync.WaitGroup{}
)
group.Add(1)
go func() {
defer group.Done()
enc := json.NewEncoder(os.Stdout)
for e := range events {
if err := enc.Encode(e); err != nil {
logrus.Error(err)
}
}
}()
if context.Bool("stats") {
s, err := container.Stats()
if err != nil {
fatal(err)
}
events <- &event{Type: "stats", ID: container.ID(), Data: s}
close(events)
group.Wait()
return
}
go func() {
for _ = range time.Tick(context.Duration("interval")) {
s, err := container.Stats()
if err != nil {
logrus.Error(err)
continue
}
stats <- s
}
}()
n, err := container.NotifyOOM()
if err != nil {
logrus.Fatal(err)
}
for {
select {
case _, ok := <-n:
if ok {
// this means an oom event was received, if it is !ok then
// the channel was closed because the container stopped and
// the cgroups no longer exist.
events <- &event{Type: "oom", ID: container.ID()}
} else {
n = nil
}
case s := <-stats:
events <- &event{Type: "stats", ID: container.ID(), Data: s}
}
if n == nil {
close(events)
break
}
}
group.Wait()
},
}

99
main.go Normal file
View File

@@ -0,0 +1,99 @@
package main
import (
"os"
"runtime"
"github.com/Sirupsen/logrus"
"github.com/codegangsta/cli"
"github.com/opencontainers/runc/libcontainer"
_ "github.com/opencontainers/runc/libcontainer/nsenter"
)
const (
version = "0.1"
usage = `open container runtime
runc integrates well with existing process supervisors to provide a production container runtime environment for
applications. It can be used with your existing process monitoring tools and the container will be spawned as direct
child of the process supervisor. nsinit can be used to manage the lifetime of a single container.
Execute a simple container in your shell by running:
cd /mycontainer
runc
`
)
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 {
fatal(err)
}
panic("--this line should never been executed, congradulations--")
}
}
func main() {
app := cli.NewApp()
app.Name = "runc"
app.Usage = usage
app.Version = version
app.Flags = []cli.Flag{
cli.StringFlag{
Name: "id",
Value: getDefaultID(),
Usage: "specify the ID to be used for the container",
},
cli.BoolFlag{
Name: "debug",
Usage: "enable debug output for logging",
},
cli.StringFlag{
Name: "root",
Value: "/var/run/ocf",
Usage: "root directory for storage of container state (this should be located in tmpfs)",
},
cli.StringFlag{
Name: "criu",
Value: "criu",
Usage: "path to the criu binary used for checkpoint and restore",
},
}
app.Commands = []cli.Command{
checkpointCommand,
eventsCommand,
restoreCommand,
specCommand,
}
app.Before = func(context *cli.Context) error {
if context.GlobalBool("debug") {
logrus.SetLevel(logrus.DebugLevel)
}
return nil
}
// default action is to execute a container
app.Action = func(context *cli.Context) {
if os.Geteuid() != 0 {
cli.ShowAppHelp(context)
logrus.Fatal("runc should be run as root")
}
spec, err := loadSpec(context.Args().First())
if err != nil {
fatal(err)
}
status, err := execContainer(context, spec)
if err != nil {
fatal(err)
}
// exit with the container's exit status so any external supervisor is
// notified of the exit with the correct exit status.
os.Exit(status)
}
if err := app.Run(os.Args); err != nil {
logrus.Fatal(err)
}
}

129
restore.go Normal file
View File

@@ -0,0 +1,129 @@
package main
import (
"os"
"os/signal"
"syscall"
"github.com/Sirupsen/logrus"
"github.com/codegangsta/cli"
"github.com/opencontainers/runc/libcontainer"
"github.com/opencontainers/runc/libcontainer/configs"
"github.com/opencontainers/runc/libcontainer/utils"
)
var restoreCommand = cli.Command{
Name: "restore",
Usage: "restore a container from a previous checkpoint",
Flags: []cli.Flag{
cli.StringFlag{Name: "image-path", Value: "", Usage: "path to criu image files for restoring"},
cli.StringFlag{Name: "work-path", Value: "", Usage: "path for saving work files and logs"},
cli.BoolFlag{Name: "tcp-established", Usage: "allow open tcp connections"},
cli.BoolFlag{Name: "ext-unix-sk", Usage: "allow external unix sockets"},
cli.BoolFlag{Name: "shell-job", Usage: "allow shell jobs"},
},
Action: func(context *cli.Context) {
imagePath := context.String("image-path")
if imagePath == "" {
imagePath = getDefaultImagePath(context)
}
spec, err := loadSpec(context.Args().First())
if err != nil {
fatal(err)
}
config, err := createLibcontainerConfig(spec)
if err != nil {
fatal(err)
}
status, err := restoreContainer(context, spec, config, imagePath)
if err != nil {
fatal(err)
}
os.Exit(status)
},
}
func restoreContainer(context *cli.Context, spec *LinuxSpec, config *configs.Config, imagePath string) (code int, err error) {
rootuid := 0
factory, err := loadFactory(context)
if err != nil {
return -1, err
}
container, err := factory.Load(context.GlobalString("id"))
if err != nil {
container, err = factory.Create(context.GlobalString("id"), config)
if err != nil {
return -1, err
}
}
options := criuOptions(context)
// ensure that the container is always removed if we were the process
// that created it.
defer func() {
if err != nil {
return
}
status, err := container.Status()
if err != nil {
logrus.Error(err)
}
if status != libcontainer.Checkpointed {
if err := container.Destroy(); err != nil {
logrus.Error(err)
}
if err := os.RemoveAll(options.ImagesDirectory); err != nil {
logrus.Error(err)
}
}
}()
process := &libcontainer.Process{
Stdin: os.Stdin,
Stdout: os.Stdout,
Stderr: os.Stderr,
}
tty, err := newTty(spec.Processes[0].TTY, process, rootuid)
if err != nil {
return -1, err
}
defer tty.Close()
go handleSignals(process, tty)
if err := container.Restore(process, options); err != nil {
return -1, err
}
status, err := process.Wait()
if err != nil {
return -1, err
}
return utils.ExitStatus(status.Sys().(syscall.WaitStatus)), nil
}
func criuOptions(context *cli.Context) *libcontainer.CriuOpts {
imagePath := getCheckpointImagePath(context)
if err := os.MkdirAll(imagePath, 0655); err != nil {
fatal(err)
}
return &libcontainer.CriuOpts{
ImagesDirectory: imagePath,
WorkDirectory: context.String("work-path"),
LeaveRunning: context.Bool("leave-running"),
TcpEstablished: true, // context.Bool("tcp-established"),
ExternalUnixConnections: context.Bool("ext-unix-sk"),
ShellJob: context.Bool("shell-job"),
}
}
// we have to use this type of signal handler because there is a memory leak if we
// wait and reap with SICHLD.
func handleSignals(process *libcontainer.Process, tty *tty) {
sigc := make(chan os.Signal, 10)
signal.Notify(sigc)
tty.resize()
for sig := range sigc {
switch sig {
case syscall.SIGWINCH:
tty.resize()
default:
process.Signal(sig)
}
}
}

57
run.go Normal file
View File

@@ -0,0 +1,57 @@
package main
import (
"fmt"
"github.com/Sirupsen/logrus"
"github.com/codegangsta/cli"
"github.com/opencontainers/runc/libcontainer"
)
func execContainer(context *cli.Context, spec *LinuxSpec) (int, error) {
if len(spec.Processes) != 1 {
return -1, fmt.Errorf("runc only supports one(1) process for the container")
}
config, err := createLibcontainerConfig(spec)
if err != nil {
return -1, err
}
rootuid, err := config.HostUID()
if err != nil {
return -1, err
}
factory, err := loadFactory(context)
if err != nil {
return -1, err
}
container, err := factory.Create(context.GlobalString("id"), config)
if err != nil {
return -1, err
}
// ensure that the container is always removed if we were the process
// that created it.
defer destroy(container)
process := newProcess(spec.Processes[0])
tty, err := newTty(spec.Processes[0].TTY, process, rootuid)
if err != nil {
return -1, err
}
handler := newSignalHandler(tty)
defer handler.Close()
if err := container.Start(process); err != nil {
return -1, err
}
return handler.forward(process)
}
func destroy(container libcontainer.Container) {
status, err := container.Status()
if err != nil {
logrus.Error(err)
}
if status != libcontainer.Checkpointed {
if err := container.Destroy(); err != nil {
logrus.Error(err)
}
}
}

108
signals.go Normal file
View File

@@ -0,0 +1,108 @@
package main
import (
"os"
"os/signal"
"syscall"
"github.com/Sirupsen/logrus"
"github.com/opencontainers/runc/libcontainer"
"github.com/opencontainers/runc/libcontainer/utils"
)
const signalBufferSize = 2048
// newSignalHandler returns a signal handler for processing SIGCHLD and SIGWINCH signals
// while still forwarding all other signals to the process.
func newSignalHandler(tty *tty) *signalHandler {
// ensure that we have a large buffer size so that we do not miss any signals
// incase we are not processing them fast enough.
s := make(chan os.Signal, signalBufferSize)
// handle all signals for the process.
signal.Notify(s)
return &signalHandler{
tty: tty,
signals: s,
}
}
// exit models a process exit status with the pid and
// exit status.
type exit struct {
pid int
status int
}
type signalHandler struct {
signals chan os.Signal
tty *tty
}
// forward handles the main signal event loop forwarding, resizing, or reaping depeding
// on the signal received.
func (h *signalHandler) forward(process *libcontainer.Process) (int, error) {
// make sure we know the pid of our main process so that we can return
// after it dies.
pid1, err := process.Pid()
if err != nil {
return -1, err
}
// perform the initial tty resize.
h.tty.resize()
for s := range h.signals {
switch s {
case syscall.SIGWINCH:
h.tty.resize()
case syscall.SIGCHLD:
exits, err := h.reap()
if err != nil {
logrus.Error(err)
}
for _, e := range exits {
logrus.WithFields(logrus.Fields{
"pid": e.pid,
"status": e.status,
}).Debug("process exited")
if e.pid == pid1 {
// call Wait() on the process even though we already have the exit
// status because we must ensure that any of the go specific process
// fun such as flushing pipes are complete before we return.
process.Wait()
return e.status, nil
}
}
default:
logrus.Debugf("sending signal to process %s", s)
if err := syscall.Kill(pid1, s.(syscall.Signal)); err != nil {
logrus.Error(err)
}
}
}
return -1, nil
}
// reap runs wait4 in a loop until we have finished processing any existing exits
// then returns all exits to the main event loop for further processing.
func (h *signalHandler) reap() (exits []exit, err error) {
var (
ws syscall.WaitStatus
rus syscall.Rusage
)
for {
pid, err := syscall.Wait4(-1, &ws, syscall.WNOHANG, &rus)
if err != nil {
if err == syscall.ECHILD {
return exits, nil
}
return nil, err
}
exits = append(exits, exit{
pid: pid,
status: utils.ExitStatus(ws),
})
}
}
func (h *signalHandler) Close() error {
return h.tty.Close()
}

146
spec.go Normal file
View File

@@ -0,0 +1,146 @@
package main
import (
"encoding/json"
"fmt"
"runtime"
"github.com/codegangsta/cli"
)
const cpuQuotaMultiplyer = 100000
type Mount struct {
Type string `json:"type"`
Source string `json:"source"`
Destination string `json:"destination"`
Options string `json:"options"`
}
type Process struct {
TTY bool `json:"tty"`
User string `json:"user"`
Args []string `json:"args"`
Env []string `json:"env"`
Cwd string `json:"cwd"`
}
type Root struct {
Path string `json:"path"`
Readonly bool `json:"readonly"`
}
type Namespace struct {
Type string `json:"type"`
Path string `json:"path,omitempty"`
}
type PortableSpec struct {
Version string `json:"version"`
OS string `json:"os"`
Arch string `json:"arch"`
Processes []*Process `json:"processes"`
Root Root `json:"root"`
Cpus float64 `json:"cpus"` // in 1.1 for 110% cpus
Memory int64 `json:"memory"` // in mb; 1024m
Hostname string `json:"hostname"`
Namespaces []Namespace `json:"namespaces"`
Capabilities []string `json:"capabilities"`
Devices []string `json:"devices"`
Mounts []Mount `json:"mounts"`
}
var specCommand = cli.Command{
Name: "spec",
Usage: "create a new specification file",
Action: func(context *cli.Context) {
spec := PortableSpec{
Version: version,
OS: runtime.GOOS,
Arch: runtime.GOARCH,
Root: Root{
Path: "rootfs",
Readonly: true,
},
Processes: []*Process{
{
TTY: true,
User: "daemon",
Args: []string{
"sh",
},
Env: []string{
"PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin",
"TERM=xterm",
},
},
},
Cpus: 1.1,
Memory: 1024,
Hostname: "shell",
Capabilities: []string{
"AUDIT_WRITE",
"KILL",
"NET_BIND_SERVICE",
},
Devices: []string{
"null",
"random",
"full",
"tty",
"zero",
"urandom",
},
Namespaces: []Namespace{
{Type: "process"},
{Type: "network"},
{Type: "mount"},
{Type: "ipc"},
{Type: "uts"},
},
Mounts: []Mount{
{
Type: "proc",
Source: "proc",
Destination: "/proc",
Options: "",
},
{
Type: "tmpfs",
Source: "tmpfs",
Destination: "/dev",
Options: "nosuid,strictatime,mode=755,size=65536k",
},
{
Type: "devpts",
Source: "devpts",
Destination: "/dev/pts",
Options: "nosuid,noexec,newinstance,ptmxmode=0666,mode=0620,gid=5",
},
{
Type: "tmpfs",
Source: "shm",
Destination: "/dev/shm",
Options: "nosuid,noexec,nodev,mode=1777,size=65536k",
},
{
Type: "mqueue",
Source: "mqueue",
Destination: "/dev/mqueue",
Options: "nosuid,noexec,nodev",
},
{
Type: "sysfs",
Source: "sysfs",
Destination: "/sys",
Options: "nosuid,noexec,nodev",
},
},
}
data, err := json.MarshalIndent(&spec, "", "\t")
if err != nil {
fatal(err)
}
fmt.Printf("%s", data)
},
}

344
spec_linux.go Normal file
View File

@@ -0,0 +1,344 @@
// +build linux
package main
import (
"encoding/json"
"fmt"
"os"
"path/filepath"
"strings"
"syscall"
"github.com/opencontainers/runc/libcontainer/cgroups"
"github.com/opencontainers/runc/libcontainer/configs"
"github.com/opencontainers/runc/libcontainer/devices"
)
type UserMapping struct {
From int `json:"from"`
To int `json:"to"`
Count int `json:"count"`
}
type Rlimit struct {
Type int `json:"type"`
Hard uint64 `json:"hard"`
Soft uint64 `json:"soft"`
}
type HugepageLimit struct {
Pagesize string `json:"pageSize"`
Limit int `json:"limit"`
}
type IfPrioMap struct {
Interface string `json:"interface"`
Priority int64 `json:"priority"`
}
type Resources struct {
// Memory reservation or soft_limit (in bytes)
MemoryReservation int64 `json:"memoryReservation"`
// Total memory usage (memory + swap); set `-1' to disable swap
MemorySwap int64 `json:"memorySwap"`
// Kernel memory limit (in bytes)
KernelMemory int64 `json:"kernelMemory"`
// CPU shares (relative weight vs. other containers)
CpuShares int64 `json:"cpuShares"`
// CPU hardcap limit (in usecs). Allowed cpu time in a given period.
CpuQuota int64 `json:"cpuQuota"`
// CPU period to be used for hardcapping (in usecs). 0 to use system default.
CpuPeriod int64 `json:"cpuPeriod"`
// How many time CPU will use in realtime scheduling (in usecs).
CpuRtRuntime int64 `json:"cpuQuota"`
// CPU period to be used for realtime scheduling (in usecs).
CpuRtPeriod int64 `json:"cpuPeriod"`
// CPU to use
CpusetCpus string `json:"cpusetCpus"`
// MEM to use
CpusetMems string `json:"cpusetMems"`
// IO read rate limit per cgroup per device, bytes per second.
BlkioThrottleReadBpsDevice string `json:"blkioThrottleReadBpsDevice"`
// IO write rate limit per cgroup per divice, bytes per second.
BlkioThrottleWriteBpsDevice string `json:"blkioThrottleWriteBpsDevice"`
// IO read rate limit per cgroup per device, IO per second.
BlkioThrottleReadIOpsDevice string `json:"blkioThrottleReadIopsDevice"`
// IO write rate limit per cgroup per device, IO per second.
BlkioThrottleWriteIOpsDevice string `json:"blkioThrottleWriteIopsDevice"`
// Specifies per cgroup weight, range is from 10 to 1000.
BlkioWeight int64 `json:"blkioWeight"`
// Weight per cgroup per device, can override BlkioWeight.
BlkioWeightDevice string `json:"blkioWeightDevice"`
// Hugetlb limit (in bytes)
HugetlbLimit []*HugepageLimit `json:"hugetlbLimit"`
// Whether to disable OOM Killer
DisableOOMKiller bool `json:"disableOOMKiller"`
// Set priority of network traffic for container
NetPrioIfpriomap []*IfPrioMap `json:"netPrioIfpriomap"`
// Set class identifier for container's network packets
NetClsClassid string `json:"netClsClassid"`
}
type LinuxSpec struct {
PortableSpec
UserMapping map[string]UserMapping `json:"userMapping"`
Rlimits []Rlimit `json:"rlimits"`
SystemProperties map[string]string `json:"systemProperties"`
Resources *Resources `json:"resources"`
}
var namespaceMapping = map[string]configs.NamespaceType{
"process": configs.NEWPID,
"network": configs.NEWNET,
"mount": configs.NEWNS,
"user": configs.NEWUSER,
"ipc": configs.NEWIPC,
"uts": configs.NEWUTS,
}
// loadSpec loads the specification from the provided path.
// If the path is empty then the default path will be "container.json"
func loadSpec(path string) (*LinuxSpec, error) {
if path == "" {
path = "container.json"
}
f, err := os.Open(path)
if err != nil {
if os.IsNotExist(err) {
return nil, fmt.Errorf("JSON specification file for %s not found", path)
}
return nil, err
}
defer f.Close()
var s *LinuxSpec
if err := json.NewDecoder(f).Decode(&s); err != nil {
return nil, err
}
return s, nil
}
func createLibcontainerConfig(spec *LinuxSpec) (*configs.Config, error) {
cwd, err := os.Getwd()
if err != nil {
return nil, err
}
config := &configs.Config{
Capabilities: spec.Capabilities,
Rootfs: filepath.Join(cwd, spec.Root.Path),
Readonlyfs: spec.Root.Readonly,
Hostname: spec.Hostname,
Privatefs: true,
}
for _, ns := range spec.Namespaces {
t, exists := namespaceMapping[ns.Type]
if !exists {
return nil, fmt.Errorf("namespace %q does not exist", ns)
}
config.Namespaces.Add(t, ns.Path)
}
for _, m := range spec.Mounts {
config.Mounts = append(config.Mounts, createLibcontianerMount(cwd, m))
}
if err := createDevices(spec, config); err != nil {
return nil, err
}
if err := setupUserNamespace(spec, config); err != nil {
return nil, err
}
c, err := createCgroupConfig(spec, config.Devices)
if err != nil {
return nil, err
}
config.Cgroups = c
if config.Readonlyfs {
setReadonly(config)
config.MaskPaths = []string{
"/proc/kcore",
}
config.ReadonlyPaths = []string{
"/proc/sys", "/proc/sysrq-trigger", "/proc/irq", "/proc/bus",
}
}
return config, nil
}
func createLibcontianerMount(cwd string, m Mount) *configs.Mount {
flags, data := parseMountOptions(m.Options)
source := m.Source
if m.Type == "bind" {
if !filepath.IsAbs(source) {
source = filepath.Join(cwd, m.Source)
}
}
return &configs.Mount{
Device: m.Type,
Source: source,
Destination: m.Destination,
Data: data,
Flags: flags,
}
}
func createCgroupConfig(spec *LinuxSpec, devices []*configs.Device) (*configs.Cgroup, error) {
myCgroupPath, err := cgroups.GetThisCgroupDir("devices")
if err != nil {
return nil, err
}
c := &configs.Cgroup{
Name: getDefaultID(),
Parent: myCgroupPath,
AllowedDevices: append(devices, allowedDevices...),
CpuQuota: getCPUQuota(spec.Cpus),
Memory: spec.Memory * 1024 * 1024,
MemorySwap: -1,
}
if r := spec.Resources; r != nil {
c.MemoryReservation = r.MemoryReservation
c.MemorySwap = r.MemorySwap
c.KernelMemory = r.KernelMemory
c.CpuShares = r.CpuShares
c.CpuQuota = r.CpuQuota
c.CpuPeriod = r.CpuPeriod
c.CpuRtRuntime = r.CpuRtRuntime
c.CpuRtPeriod = r.CpuRtPeriod
c.CpusetCpus = r.CpusetCpus
c.CpusetMems = r.CpusetMems
c.BlkioThrottleReadBpsDevice = r.BlkioThrottleReadBpsDevice
c.BlkioThrottleWriteBpsDevice = r.BlkioThrottleWriteBpsDevice
c.BlkioThrottleReadIOpsDevice = r.BlkioThrottleReadIOpsDevice
c.BlkioThrottleWriteIOpsDevice = r.BlkioThrottleWriteIOpsDevice
c.BlkioWeight = r.BlkioWeight
c.BlkioWeightDevice = r.BlkioWeightDevice
for _, l := range r.HugetlbLimit {
c.HugetlbLimit = append(c.HugetlbLimit, &configs.HugepageLimit{
Pagesize: l.Pagesize,
Limit: l.Limit,
})
}
c.OomKillDisable = r.DisableOOMKiller
for _, m := range r.NetPrioIfpriomap {
c.NetPrioIfpriomap = append(c.NetPrioIfpriomap, &configs.IfPrioMap{
Interface: m.Interface,
Priority: m.Priority,
})
}
c.NetClsClassid = r.NetClsClassid
}
return c, nil
}
func createDevices(spec *LinuxSpec, config *configs.Config) error {
for _, name := range spec.Devices {
d, err := devices.DeviceFromPath(filepath.Join("/dev", name), "rwm")
if err != nil {
return err
}
config.Devices = append(config.Devices, d)
}
return nil
}
func setReadonly(config *configs.Config) {
for _, m := range config.Mounts {
if m.Device == "sysfs" {
m.Flags |= syscall.MS_RDONLY
}
}
}
func getCPUQuota(cpus float64) int64 {
return int64(cpus * cpuQuotaMultiplyer)
}
func setupUserNamespace(spec *LinuxSpec, config *configs.Config) error {
if len(spec.UserMapping) == 0 {
return nil
}
config.Namespaces.Add(configs.NEWUSER, "")
mappings := make(map[string][]configs.IDMap)
for k, v := range spec.UserMapping {
mappings[k] = append(mappings[k], configs.IDMap{
ContainerID: v.From,
HostID: v.To,
Size: v.Count,
})
}
config.UidMappings = mappings["uid"]
config.GidMappings = mappings["gid"]
rootUid, err := config.HostUID()
if err != nil {
return err
}
rootGid, err := config.HostGID()
if err != nil {
return err
}
for _, node := range config.Devices {
node.Uid = uint32(rootUid)
node.Gid = uint32(rootGid)
}
return nil
}
// parseMountOptions parses the string and returns the flags and any mount data that
// it contains.
func parseMountOptions(options string) (int, string) {
var (
flag int
data []string
)
flags := map[string]struct {
clear bool
flag int
}{
"defaults": {false, 0},
"ro": {false, syscall.MS_RDONLY},
"rw": {true, syscall.MS_RDONLY},
"suid": {true, syscall.MS_NOSUID},
"nosuid": {false, syscall.MS_NOSUID},
"dev": {true, syscall.MS_NODEV},
"nodev": {false, syscall.MS_NODEV},
"exec": {true, syscall.MS_NOEXEC},
"noexec": {false, syscall.MS_NOEXEC},
"sync": {false, syscall.MS_SYNCHRONOUS},
"async": {true, syscall.MS_SYNCHRONOUS},
"dirsync": {false, syscall.MS_DIRSYNC},
"remount": {false, syscall.MS_REMOUNT},
"mand": {false, syscall.MS_MANDLOCK},
"nomand": {true, syscall.MS_MANDLOCK},
"atime": {true, syscall.MS_NOATIME},
"noatime": {false, syscall.MS_NOATIME},
"diratime": {true, syscall.MS_NODIRATIME},
"nodiratime": {false, syscall.MS_NODIRATIME},
"bind": {false, syscall.MS_BIND},
"rbind": {false, syscall.MS_BIND | syscall.MS_REC},
"unbindable": {false, syscall.MS_UNBINDABLE},
"runbindable": {false, syscall.MS_UNBINDABLE | syscall.MS_REC},
"private": {false, syscall.MS_PRIVATE},
"rprivate": {false, syscall.MS_PRIVATE | syscall.MS_REC},
"shared": {false, syscall.MS_SHARED},
"rshared": {false, syscall.MS_SHARED | syscall.MS_REC},
"slave": {false, syscall.MS_SLAVE},
"rslave": {false, syscall.MS_SLAVE | syscall.MS_REC},
"relatime": {false, syscall.MS_RELATIME},
"norelatime": {true, syscall.MS_RELATIME},
"strictatime": {false, syscall.MS_STRICTATIME},
"nostrictatime": {true, syscall.MS_STRICTATIME},
}
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, ",")
}

97
tty.go Normal file
View File

@@ -0,0 +1,97 @@
package main
import (
"io"
"os"
"github.com/docker/docker/pkg/term"
"github.com/opencontainers/runc/libcontainer"
)
// newTty creates a new pty for use with the container. If a tty is not to be
// created for the process, pipes are created so that the TTY of the parent
// process are not inherited by the container.
func newTty(create bool, p *libcontainer.Process, rootuid int) (*tty, error) {
if create {
return createTty(p, rootuid)
}
return createStdioPipes(p)
}
// setup standard pipes so that the TTY of the calling nsinit process
// is not inherited by the container.
func createStdioPipes(p *libcontainer.Process) (*tty, error) {
t := &tty{}
r, w, err := os.Pipe()
if err != nil {
return nil, err
}
go io.Copy(w, os.Stdin)
t.closers = append(t.closers, w)
p.Stdin = r
if r, w, err = os.Pipe(); err != nil {
return nil, err
}
go io.Copy(os.Stdout, r)
p.Stdout = w
t.closers = append(t.closers, r)
if r, w, err = os.Pipe(); err != nil {
return nil, err
}
go io.Copy(os.Stderr, r)
p.Stderr = w
t.closers = append(t.closers, r)
return t, nil
}
func createTty(p *libcontainer.Process, rootuid int) (*tty, error) {
console, err := p.NewConsole(rootuid)
if err != nil {
return nil, err
}
go io.Copy(console, os.Stdin)
go io.Copy(os.Stdout, console)
state, err := term.SetRawTerminal(os.Stdin.Fd())
if err != nil {
return nil, err
}
t := &tty{
console: console,
state: state,
closers: []io.Closer{
console,
},
}
p.Stderr = nil
p.Stdout = nil
p.Stdin = nil
return t, nil
}
type tty struct {
console libcontainer.Console
state *term.State
closers []io.Closer
}
func (t *tty) Close() error {
for _, c := range t.closers {
c.Close()
}
if t.state != nil {
term.RestoreTerminal(os.Stdin.Fd(), t.state)
}
return nil
}
func (t *tty) resize() error {
if t.console == nil {
return nil
}
ws, err := term.GetWinsize(os.Stdin.Fd())
if err != nil {
return err
}
return term.SetWinsize(t.console.Fd(), ws)
}

173
utils.go Normal file
View File

@@ -0,0 +1,173 @@
package main
import (
"fmt"
"os"
"path/filepath"
"github.com/codegangsta/cli"
"github.com/opencontainers/runc/libcontainer"
"github.com/opencontainers/runc/libcontainer/configs"
)
const wildcard = -1
var allowedDevices = []*configs.Device{
// allow mknod for any device
{
Type: 'c',
Major: wildcard,
Minor: wildcard,
Permissions: "m",
},
{
Type: 'b',
Major: wildcard,
Minor: wildcard,
Permissions: "m",
},
{
Path: "/dev/console",
Type: 'c',
Major: 5,
Minor: 1,
Permissions: "rwm",
},
{
Path: "/dev/tty0",
Type: 'c',
Major: 4,
Minor: 0,
Permissions: "rwm",
},
{
Path: "/dev/tty1",
Type: 'c',
Major: 4,
Minor: 1,
Permissions: "rwm",
},
// /dev/pts/ - pts namespaces are "coming soon"
{
Path: "",
Type: 'c',
Major: 136,
Minor: wildcard,
Permissions: "rwm",
},
{
Path: "",
Type: 'c',
Major: 5,
Minor: 2,
Permissions: "rwm",
},
// tuntap
{
Path: "",
Type: 'c',
Major: 10,
Minor: 200,
Permissions: "rwm",
},
}
var container libcontainer.Container
func containerPreload(context *cli.Context) error {
c, err := getContainer(context)
if err != nil {
return err
}
container = c
return nil
}
var factory libcontainer.Factory
func factoryPreload(context *cli.Context) error {
f, err := loadFactory(context)
if err != nil {
return err
}
factory = f
return nil
}
// loadFactory returns the configured factory instance for execing containers.
func loadFactory(context *cli.Context) (libcontainer.Factory, error) {
root := context.GlobalString("root")
abs, err := filepath.Abs(root)
if err != nil {
return nil, err
}
return libcontainer.New(abs, libcontainer.Cgroupfs, func(l *libcontainer.LinuxFactory) error {
l.CriuPath = context.GlobalString("criu")
return nil
})
}
// getContainer returns the specified container instance by loading it from state
// with the default factory.
func getContainer(context *cli.Context) (libcontainer.Container, error) {
factory, err := loadFactory(context)
if err != nil {
return nil, err
}
container, err := factory.Load(context.GlobalString("id"))
if err != nil {
return nil, err
}
return container, nil
}
// fatal prints the error's details if it is a libcontainer specific error type
// then exists the program with an exit status of 1.
func fatal(err error) {
if lerr, ok := err.(libcontainer.Error); ok {
lerr.Detail(os.Stderr)
os.Exit(1)
}
fmt.Fprintln(os.Stderr, err)
os.Exit(1)
}
// fatalf formats the errror string with the specified template then exits the
// program with an exit status of 1.
func fatalf(t string, v ...interface{}) {
fmt.Fprintf(os.Stderr, t, v...)
os.Exit(1)
}
// getDefaultID returns a string to be used as the container id based on the
// current working directory of the nsinit process. This function panics
// if the cwd is unable to be found based on a system error.
func getDefaultID() string {
cwd, err := os.Getwd()
if err != nil {
panic(err)
}
return filepath.Base(cwd)
}
func getDefaultImagePath(context *cli.Context) string {
cwd, err := os.Getwd()
if err != nil {
panic(err)
}
return filepath.Join(cwd, "checkpoint")
}
// newProcess returns a new libcontainer Process with the arguments from the
// spec and stdio from the current process.
func newProcess(p *Process) *libcontainer.Process {
return &libcontainer.Process{
Args: p.Args,
Env: p.Env,
User: p.User,
Cwd: p.Cwd,
Stdin: os.Stdin,
Stdout: os.Stdout,
Stderr: os.Stderr,
}
}