diff options
-rw-r--r-- | libpod/container_inspect.go | 34 | ||||
-rw-r--r-- | libpod/define/container_inspect.go | 7 | ||||
-rw-r--r-- | libpod/options.go | 16 | ||||
-rw-r--r-- | libpod/pod.go | 15 | ||||
-rw-r--r-- | libpod/pod_api.go | 20 |
5 files changed, 82 insertions, 10 deletions
diff --git a/libpod/container_inspect.go b/libpod/container_inspect.go index ae28dde94..b26dcddf6 100644 --- a/libpod/container_inspect.go +++ b/libpod/container_inspect.go @@ -580,7 +580,10 @@ func (c *Container) generateInspectContainerHostConfig(ctrSpec *spec.Spec, named networkMode := "" switch { case c.config.CreateNetNS: - networkMode = "default" + // We actually store the network + // mode for Slirp and Bridge, so + // we can just use that + networkMode = string(c.config.NetMode) case c.config.NetNsCtr != "": networkMode = fmt.Sprintf("container:%s", c.config.NetNsCtr) default: @@ -594,7 +597,10 @@ func (c *Container) generateInspectContainerHostConfig(ctrSpec *spec.Spec, named if ns.Path != "" { networkMode = fmt.Sprintf("ns:%s", ns.Path) } else { - networkMode = "private" + // We're making a network ns, but not + // configuring with Slirp or CNI. That + // means it's --net=none + networkMode = "none" } break } @@ -698,6 +704,30 @@ func (c *Container) generateInspectContainerHostConfig(ctrSpec *spec.Spec, named } hostConfig.IpcMode = ipcMode + // Cgroup namespace mode + cgroupMode := "" + if c.config.CgroupNsCtr != "" { + cgroupMode = fmt.Sprintf("container:%s", c.config.CgroupNsCtr) + } else if ctrSpec.Linux != nil { + // Locate the spec's cgroup namespace + // If there is none, it's cgroup=host. + // If there is one and it has a path, it's "ns:". + // If there is no path, it's private. + for _, ns := range ctrSpec.Linux.Namespaces { + if ns.Type == spec.CgroupNamespace { + if ns.Path != "" { + cgroupMode = fmt.Sprintf("ns:%s", ns.Path) + } else { + cgroupMode = "private" + } + } + } + if cgroupMode == "" { + cgroupMode = "host" + } + } + hostConfig.CgroupMode = cgroupMode + // CGroup parent // Need to check if it's the default, and not print if so. defaultCgroupParent := "" diff --git a/libpod/define/container_inspect.go b/libpod/define/container_inspect.go index e6a19e5b4..27ada8706 100644 --- a/libpod/define/container_inspect.go +++ b/libpod/define/container_inspect.go @@ -228,6 +228,13 @@ type InspectContainerHostConfig struct { // include a Mounts field in inspect. // Format: <src>:<destination>[:<comma-separated options>] Binds []string `json:"Binds"` + // CgroupMode is the configuration of the container's cgroup namespace. + // Populated as follows: + // private - a cgroup namespace has been created + // host - No cgroup namespace created + // container:<id> - Using another container's cgroup namespace + // ns:<path> - A path to a cgroup namespace has been specified + CgroupMode string `json:"CgroupMode"` // ContainerIDFile is a file created during container creation to hold // the ID of the created container. // This is not handled within libpod and is stored in an annotation. diff --git a/libpod/options.go b/libpod/options.go index 33b423bce..05241baf3 100644 --- a/libpod/options.go +++ b/libpod/options.go @@ -1692,6 +1692,22 @@ func WithPodUTS() PodCreateOption { } } +// WithPodCgroup tells containers in this pod to use the cgroup namespace +// created for this pod. +// Containers in a pod will inherit the kernel namespaces from the first +// container added. +func WithPodCgroup() PodCreateOption { + return func(pod *Pod) error { + if pod.valid { + return define.ErrPodFinalized + } + + pod.config.UsePodCgroupNS = true + + return nil + } +} + // WithInfraContainer tells the pod to create a pause container func WithInfraContainer() PodCreateOption { return func(pod *Pod) error { diff --git a/libpod/pod.go b/libpod/pod.go index 8eb06ae2f..34ceef5ef 100644 --- a/libpod/pod.go +++ b/libpod/pod.go @@ -51,12 +51,13 @@ type PodConfig struct { // The following UsePod{kernelNamespace} indicate whether the containers // in the pod will inherit the namespace from the first container in the pod. - UsePodPID bool `json:"sharesPid,omitempty"` - UsePodIPC bool `json:"sharesIpc,omitempty"` - UsePodNet bool `json:"sharesNet,omitempty"` - UsePodMount bool `json:"sharesMnt,omitempty"` - UsePodUser bool `json:"sharesUser,omitempty"` - UsePodUTS bool `json:"sharesUts,omitempty"` + UsePodPID bool `json:"sharesPid,omitempty"` + UsePodIPC bool `json:"sharesIpc,omitempty"` + UsePodNet bool `json:"sharesNet,omitempty"` + UsePodMount bool `json:"sharesMnt,omitempty"` + UsePodUser bool `json:"sharesUser,omitempty"` + UsePodUTS bool `json:"sharesUts,omitempty"` + UsePodCgroupNS bool `json:"sharesCgroupNS,omitempty"` InfraContainer *InfraContainerConfig `json:"infraConfig"` @@ -167,7 +168,7 @@ func (p *Pod) SharesUTS() bool { // SharesCgroup returns whether containers in the pod will default to this pod's // cgroup instead of the default libpod parent func (p *Pod) SharesCgroup() bool { - return p.config.UsePodCgroup + return p.config.UsePodCgroupNS } // CgroupPath returns the path to the pod's CGroup diff --git a/libpod/pod_api.go b/libpod/pod_api.go index 45aa5cb8d..0be9f2573 100644 --- a/libpod/pod_api.go +++ b/libpod/pod_api.go @@ -466,6 +466,24 @@ func (p *Pod) Inspect() (*define.InspectPodData, error) { if err != nil { return nil, err } + + namespaces := map[string]bool{ + "pid": p.config.UsePodPID, + "ipc": p.config.UsePodIPC, + "net": p.config.UsePodNet, + "mount": p.config.UsePodMount, + "user": p.config.UsePodUser, + "uts": p.config.UsePodUTS, + "cgroup": p.config.UsePodCgroupNS, + } + + sharesNS := []string{} + for nsStr, include := range namespaces { + if include { + sharesNS = append(sharesNS, nsStr) + } + } + inspectData := define.InspectPodData{ ID: p.ID(), Name: p.Name(), @@ -480,7 +498,7 @@ func (p *Pod) Inspect() (*define.InspectPodData, error) { CreateInfra: false, InfraContainerID: p.state.InfraContainerID, InfraConfig: nil, - SharedNamespaces: nil, + SharedNamespaces: sharesNS, NumContainers: uint(len(containers)), Containers: ctrs, } |