moby/daemon/daemon_test.go
Cory Snider d222bf097c daemon: reload runtimes w/o breaking containers
The existing runtimes reload logic went to great lengths to replace the
directory containing runtime wrapper scripts as atomically as possible
within the limitations of the Linux filesystem ABI. Trouble is,
atomically swapping the wrapper scripts directory solves the wrong
problem! The runtime configuration is "locked in" when a container is
started, including the path to the runC binary. If a container is
started with a runtime which requires a daemon-managed wrapper script
and then the daemon is reloaded with a config which no longer requires
the wrapper script (i.e. some args -> no args, or the runtime is dropped
from the config), that container would become unmanageable. Any attempts
to stop, exec or otherwise perform lifecycle management operations on
the container are likely to fail due to the wrapper script no longer
existing at its original path.

Atomically swapping the wrapper scripts is also incompatible with the
read-copy-update paradigm for reloading configuration. A handler in the
daemon could retain a reference to the pre-reload configuration for an
indeterminate amount of time after the daemon configuration has been
reloaded and updated. It is possible for the daemon to attempt to start
a container using a deleted wrapper script if a request to run a
container races a reload.

Solve the problem of deleting referenced wrapper scripts by ensuring
that all wrapper scripts are *immutable* for the lifetime of the daemon
process. Any given runtime wrapper script must always exist with the
same contents, no matter how many times the daemon config is reloaded,
or what changes are made to the config. This is accomplished by using
everyone's favourite design pattern: content-addressable storage. Each
wrapper script file name is suffixed with the SHA-256 digest of its
contents to (probabilistically) guarantee immutability without needing
any concurrency control. Stale runtime wrapper scripts are only cleaned
up on the next daemon restart.

Split the derived runtimes configuration from the user-supplied
configuration to have a place to store derived state without mutating
the user-supplied configuration or exposing daemon internals in API
struct types. Hold the derived state and the user-supplied configuration
in a single struct value so that they can be updated as an atomic unit.

Signed-off-by: Cory Snider <csnider@mirantis.com>
2023-06-01 14:45:25 -04:00

315 lines
11 KiB
Go

package daemon // import "github.com/docker/docker/daemon"
import (
"os"
"path/filepath"
"runtime"
"testing"
containertypes "github.com/docker/docker/api/types/container"
"github.com/docker/docker/container"
"github.com/docker/docker/errdefs"
"github.com/docker/docker/libnetwork"
"github.com/docker/docker/pkg/idtools"
volumesservice "github.com/docker/docker/volume/service"
"github.com/docker/go-connections/nat"
"github.com/pkg/errors"
"gotest.tools/v3/assert"
is "gotest.tools/v3/assert/cmp"
)
//
// https://github.com/docker/docker/issues/8069
//
func TestGetContainer(t *testing.T) {
c1 := &container.Container{
ID: "5a4ff6a163ad4533d22d69a2b8960bf7fafdcba06e72d2febdba229008b0bf57",
Name: "tender_bardeen",
}
c2 := &container.Container{
ID: "3cdbd1aa394fd68559fd1441d6eff2ab7c1e6363582c82febfaa8045df3bd8de",
Name: "drunk_hawking",
}
c3 := &container.Container{
ID: "3cdbd1aa394fd68559fd1441d6eff2abfafdcba06e72d2febdba229008b0bf57",
Name: "3cdbd1aa",
}
c4 := &container.Container{
ID: "75fb0b800922abdbef2d27e60abcdfaf7fb0698b2a96d22d3354da361a6ff4a5",
Name: "5a4ff6a163ad4533d22d69a2b8960bf7fafdcba06e72d2febdba229008b0bf57",
}
c5 := &container.Container{
ID: "d22d69a2b8960bf7fafdcba06e72d2febdba960bf7fafdcba06e72d2f9008b060b",
Name: "d22d69a2b896",
}
store := container.NewMemoryStore()
store.Add(c1.ID, c1)
store.Add(c2.ID, c2)
store.Add(c3.ID, c3)
store.Add(c4.ID, c4)
store.Add(c5.ID, c5)
containersReplica, err := container.NewViewDB()
if err != nil {
t.Fatalf("could not create ViewDB: %v", err)
}
containersReplica.Save(c1)
containersReplica.Save(c2)
containersReplica.Save(c3)
containersReplica.Save(c4)
containersReplica.Save(c5)
daemon := &Daemon{
containers: store,
containersReplica: containersReplica,
}
daemon.reserveName(c1.ID, c1.Name)
daemon.reserveName(c2.ID, c2.Name)
daemon.reserveName(c3.ID, c3.Name)
daemon.reserveName(c4.ID, c4.Name)
daemon.reserveName(c5.ID, c5.Name)
if ctr, _ := daemon.GetContainer("3cdbd1aa394fd68559fd1441d6eff2ab7c1e6363582c82febfaa8045df3bd8de"); ctr != c2 {
t.Fatal("Should explicitly match full container IDs")
}
if ctr, _ := daemon.GetContainer("75fb0b8009"); ctr != c4 {
t.Fatal("Should match a partial ID")
}
if ctr, _ := daemon.GetContainer("drunk_hawking"); ctr != c2 {
t.Fatal("Should match a full name")
}
// c3.Name is a partial match for both c3.ID and c2.ID
if c, _ := daemon.GetContainer("3cdbd1aa"); c != c3 {
t.Fatal("Should match a full name even though it collides with another container's ID")
}
if ctr, _ := daemon.GetContainer("d22d69a2b896"); ctr != c5 {
t.Fatal("Should match a container where the provided prefix is an exact match to the its name, and is also a prefix for its ID")
}
if _, err := daemon.GetContainer("3cdbd1"); err == nil {
t.Fatal("Should return an error when provided a prefix that partially matches multiple container ID's")
}
if _, err := daemon.GetContainer("nothing"); err == nil {
t.Fatal("Should return an error when provided a prefix that is neither a name or a partial match to an ID")
}
}
func initDaemonWithVolumeStore(tmp string) (*Daemon, error) {
var err error
daemon := &Daemon{
repository: tmp,
root: tmp,
}
daemon.volumes, err = volumesservice.NewVolumeService(tmp, nil, idtools.Identity{UID: 0, GID: 0}, daemon)
if err != nil {
return nil, err
}
return daemon, nil
}
func TestValidContainerNames(t *testing.T) {
invalidNames := []string{"-rm", "&sdfsfd", "safd%sd"}
validNames := []string{"word-word", "word_word", "1weoid"}
for _, name := range invalidNames {
if validContainerNamePattern.MatchString(name) {
t.Fatalf("%q is not a valid container name and was returned as valid.", name)
}
}
for _, name := range validNames {
if !validContainerNamePattern.MatchString(name) {
t.Fatalf("%q is a valid container name and was returned as invalid.", name)
}
}
}
func TestContainerInitDNS(t *testing.T) {
if os.Getuid() != 0 {
t.Skip("root required") // for chown
}
tmp, err := os.MkdirTemp("", "docker-container-test-")
if err != nil {
t.Fatal(err)
}
defer os.RemoveAll(tmp)
containerID := "d59df5276e7b219d510fe70565e0404bc06350e0d4b43fe961f22f339980170e"
containerPath := filepath.Join(tmp, containerID)
if err := os.MkdirAll(containerPath, 0755); err != nil {
t.Fatal(err)
}
config := `{"State":{"Running":true,"Paused":false,"Restarting":false,"OOMKilled":false,"Dead":false,"Pid":2464,"ExitCode":0,
"Error":"","StartedAt":"2015-05-26T16:48:53.869308965Z","FinishedAt":"0001-01-01T00:00:00Z"},
"ID":"d59df5276e7b219d510fe70565e0404bc06350e0d4b43fe961f22f339980170e","Created":"2015-05-26T16:48:53.7987917Z","Path":"top",
"Args":[],"Config":{"Hostname":"d59df5276e7b","Domainname":"","User":"","Memory":0,"MemorySwap":0,"CpuShares":0,"Cpuset":"",
"AttachStdin":false,"AttachStdout":false,"AttachStderr":false,"PortSpecs":null,"ExposedPorts":null,"Tty":true,"OpenStdin":true,
"StdinOnce":false,"Env":null,"Cmd":["top"],"Image":"ubuntu:latest","Volumes":null,"WorkingDir":"","Entrypoint":null,
"NetworkDisabled":false,"MacAddress":"","OnBuild":null,"Labels":{}},"Image":"07f8e8c5e66084bef8f848877857537ffe1c47edd01a93af27e7161672ad0e95",
"NetworkSettings":{"IPAddress":"172.17.0.1","IPPrefixLen":16,"MacAddress":"02:42:ac:11:00:01","LinkLocalIPv6Address":"fe80::42:acff:fe11:1",
"LinkLocalIPv6PrefixLen":64,"GlobalIPv6Address":"","GlobalIPv6PrefixLen":0,"Gateway":"172.17.42.1","IPv6Gateway":"","Bridge":"docker0","Ports":{}},
"ResolvConfPath":"/var/lib/docker/containers/d59df5276e7b219d510fe70565e0404bc06350e0d4b43fe961f22f339980170e/resolv.conf",
"HostnamePath":"/var/lib/docker/containers/d59df5276e7b219d510fe70565e0404bc06350e0d4b43fe961f22f339980170e/hostname",
"HostsPath":"/var/lib/docker/containers/d59df5276e7b219d510fe70565e0404bc06350e0d4b43fe961f22f339980170e/hosts",
"LogPath":"/var/lib/docker/containers/d59df5276e7b219d510fe70565e0404bc06350e0d4b43fe961f22f339980170e/d59df5276e7b219d510fe70565e0404bc06350e0d4b43fe961f22f339980170e-json.log",
"Name":"/ubuntu","Driver":"aufs","MountLabel":"","ProcessLabel":"","AppArmorProfile":"","RestartCount":0,
"UpdateDns":false,"Volumes":{},"VolumesRW":{},"AppliedVolumesFrom":null}`
// Container struct only used to retrieve path to config file
ctr := &container.Container{Root: containerPath}
configPath, err := ctr.ConfigPath()
if err != nil {
t.Fatal(err)
}
if err = os.WriteFile(configPath, []byte(config), 0644); err != nil {
t.Fatal(err)
}
hostConfig := `{"Binds":[],"ContainerIDFile":"","Memory":0,"MemorySwap":0,"CpuShares":0,"CpusetCpus":"",
"Privileged":false,"PortBindings":{},"Links":null,"PublishAllPorts":false,"Dns":null,"DnsOptions":null,"DnsSearch":null,"ExtraHosts":null,"VolumesFrom":null,
"Devices":[],"NetworkMode":"bridge","IpcMode":"","PidMode":"","CapAdd":null,"CapDrop":null,"RestartPolicy":{"Name":"no","MaximumRetryCount":0},
"SecurityOpt":null,"ReadonlyRootfs":false,"Ulimits":null,"LogConfig":{"Type":"","Config":null},"CgroupParent":""}`
hostConfigPath, err := ctr.HostConfigPath()
if err != nil {
t.Fatal(err)
}
if err = os.WriteFile(hostConfigPath, []byte(hostConfig), 0644); err != nil {
t.Fatal(err)
}
daemon, err := initDaemonWithVolumeStore(tmp)
if err != nil {
t.Fatal(err)
}
c, err := daemon.load(containerID)
if err != nil {
t.Fatal(err)
}
if c.HostConfig.DNS == nil {
t.Fatal("Expected container DNS to not be nil")
}
if c.HostConfig.DNSSearch == nil {
t.Fatal("Expected container DNSSearch to not be nil")
}
if c.HostConfig.DNSOptions == nil {
t.Fatal("Expected container DNSOptions to not be nil")
}
}
func newPortNoError(proto, port string) nat.Port {
p, _ := nat.NewPort(proto, port)
return p
}
func TestMerge(t *testing.T) {
volumesImage := make(map[string]struct{})
volumesImage["/test1"] = struct{}{}
volumesImage["/test2"] = struct{}{}
portsImage := make(nat.PortSet)
portsImage[newPortNoError("tcp", "1111")] = struct{}{}
portsImage[newPortNoError("tcp", "2222")] = struct{}{}
configImage := &containertypes.Config{
ExposedPorts: portsImage,
Env: []string{"VAR1=1", "VAR2=2"},
Volumes: volumesImage,
}
portsUser := make(nat.PortSet)
portsUser[newPortNoError("tcp", "2222")] = struct{}{}
portsUser[newPortNoError("tcp", "3333")] = struct{}{}
volumesUser := make(map[string]struct{})
volumesUser["/test3"] = struct{}{}
configUser := &containertypes.Config{
ExposedPorts: portsUser,
Env: []string{"VAR2=3", "VAR3=3"},
Volumes: volumesUser,
}
if err := merge(configUser, configImage); err != nil {
t.Error(err)
}
if len(configUser.ExposedPorts) != 3 {
t.Fatalf("Expected 3 ExposedPorts, 1111, 2222 and 3333, found %d", len(configUser.ExposedPorts))
}
for portSpecs := range configUser.ExposedPorts {
if portSpecs.Port() != "1111" && portSpecs.Port() != "2222" && portSpecs.Port() != "3333" {
t.Fatalf("Expected 1111 or 2222 or 3333, found %s", portSpecs)
}
}
if len(configUser.Env) != 3 {
t.Fatalf("Expected 3 env var, VAR1=1, VAR2=3 and VAR3=3, found %d", len(configUser.Env))
}
for _, env := range configUser.Env {
if env != "VAR1=1" && env != "VAR2=3" && env != "VAR3=3" {
t.Fatalf("Expected VAR1=1 or VAR2=3 or VAR3=3, found %s", env)
}
}
if len(configUser.Volumes) != 3 {
t.Fatalf("Expected 3 volumes, /test1, /test2 and /test3, found %d", len(configUser.Volumes))
}
for v := range configUser.Volumes {
if v != "/test1" && v != "/test2" && v != "/test3" {
t.Fatalf("Expected /test1 or /test2 or /test3, found %s", v)
}
}
ports, _, err := nat.ParsePortSpecs([]string{"0000"})
if err != nil {
t.Error(err)
}
configImage2 := &containertypes.Config{
ExposedPorts: ports,
}
if err := merge(configUser, configImage2); err != nil {
t.Error(err)
}
if len(configUser.ExposedPorts) != 4 {
t.Fatalf("Expected 4 ExposedPorts, 0000, 1111, 2222 and 3333, found %d", len(configUser.ExposedPorts))
}
for portSpecs := range configUser.ExposedPorts {
if portSpecs.Port() != "0" && portSpecs.Port() != "1111" && portSpecs.Port() != "2222" && portSpecs.Port() != "3333" {
t.Fatalf("Expected %q or %q or %q or %q, found %s", 0, 1111, 2222, 3333, portSpecs)
}
}
}
func TestValidateContainerIsolation(t *testing.T) {
d := Daemon{}
_, err := d.verifyContainerSettings(&configStore{}, &containertypes.HostConfig{Isolation: containertypes.Isolation("invalid")}, nil, false)
assert.Check(t, is.Error(err, "invalid isolation 'invalid' on "+runtime.GOOS))
}
func TestFindNetworkErrorType(t *testing.T) {
d := Daemon{}
_, err := d.FindNetwork("fakeNet")
var nsn libnetwork.ErrNoSuchNetwork
ok := errors.As(err, &nsn)
if !errdefs.IsNotFound(err) || !ok {
t.Error("The FindNetwork method MUST always return an error that implements the NotFound interface and is ErrNoSuchNetwork")
}
}