/* Copyright 2014 The Kubernetes Authors All rights reserved. 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. */ package nfs import ( "fmt" "os" "github.com/GoogleCloudPlatform/kubernetes/pkg/api" "github.com/GoogleCloudPlatform/kubernetes/pkg/types" "github.com/GoogleCloudPlatform/kubernetes/pkg/util" "github.com/GoogleCloudPlatform/kubernetes/pkg/util/mount" "github.com/GoogleCloudPlatform/kubernetes/pkg/volume" "github.com/golang/glog" ) // This is the primary entrypoint for volume plugins. func ProbeVolumePlugins() []volume.VolumePlugin { return []volume.VolumePlugin{&nfsPlugin{nil, mount.New()}} } type nfsPlugin struct { host volume.VolumeHost mounter mount.Interface } var _ volume.VolumePlugin = &nfsPlugin{} const ( nfsPluginName = "kubernetes.io/nfs" ) func (plugin *nfsPlugin) Init(host volume.VolumeHost) { plugin.host = host } func (plugin *nfsPlugin) Name() string { return nfsPluginName } func (plugin *nfsPlugin) CanSupport(spec *volume.Spec) bool { return spec.VolumeSource.NFS != nil } func (plugin *nfsPlugin) GetAccessModes() []api.AccessModeType { return []api.AccessModeType{ api.ReadWriteOnce, api.ReadOnlyMany, api.ReadWriteMany, } } func (plugin *nfsPlugin) NewBuilder(spec *volume.Spec, pod *api.Pod, _ volume.VolumeOptions, _ mount.Interface) (volume.Builder, error) { return plugin.newBuilderInternal(spec, pod, plugin.mounter) } func (plugin *nfsPlugin) newBuilderInternal(spec *volume.Spec, pod *api.Pod, mounter mount.Interface) (volume.Builder, error) { return &nfs{ volName: spec.Name, server: spec.VolumeSource.NFS.Server, exportPath: spec.VolumeSource.NFS.Path, readOnly: spec.VolumeSource.NFS.ReadOnly, mounter: mounter, pod: pod, plugin: plugin, }, nil } func (plugin *nfsPlugin) NewCleaner(volName string, podUID types.UID, _ mount.Interface) (volume.Cleaner, error) { return plugin.newCleanerInternal(volName, podUID, plugin.mounter) } func (plugin *nfsPlugin) newCleanerInternal(volName string, podUID types.UID, mounter mount.Interface) (volume.Cleaner, error) { return &nfs{ volName: volName, server: "", exportPath: "", readOnly: false, mounter: mounter, pod: &api.Pod{ObjectMeta: api.ObjectMeta{UID: podUID}}, plugin: plugin, }, nil } // NFS volumes represent a bare host file or directory mount of an NFS export. type nfs struct { volName string pod *api.Pod server string exportPath string readOnly bool mounter mount.Interface plugin *nfsPlugin } // SetUp attaches the disk and bind mounts to the volume path. func (nfsVolume *nfs) SetUp() error { return nfsVolume.SetUpAt(nfsVolume.GetPath()) } func (nfsVolume *nfs) SetUpAt(dir string) error { mountpoint, err := nfsVolume.mounter.IsMountPoint(dir) glog.V(4).Infof("NFS mount set up: %s %v %v", dir, mountpoint, err) if err != nil && !os.IsNotExist(err) { return err } if mountpoint { return nil } os.MkdirAll(dir, 0750) source := fmt.Sprintf("%s:%s", nfsVolume.server, nfsVolume.exportPath) options := []string{} if nfsVolume.readOnly { options = append(options, "ro") } err = nfsVolume.mounter.Mount(source, dir, "nfs", options) if err != nil { mountpoint, mntErr := nfsVolume.mounter.IsMountPoint(dir) if mntErr != nil { glog.Errorf("IsMountpoint check failed: %v", mntErr) return err } if mountpoint { if mntErr = nfsVolume.mounter.Unmount(dir); mntErr != nil { glog.Errorf("Failed to unmount: %v", mntErr) return err } mountpoint, mntErr := nfsVolume.mounter.IsMountPoint(dir) if mntErr != nil { glog.Errorf("IsMountpoint check failed: %v", mntErr) return err } if mountpoint { // This is very odd, we don't expect it. We'll try again next sync loop. glog.Errorf("%s is still mounted, despite call to unmount(). Will try again next sync loop.", dir) return err } } os.Remove(dir) return err } return nil } func (nfsVolume *nfs) GetPath() string { name := nfsPluginName return nfsVolume.plugin.host.GetPodVolumeDir(nfsVolume.pod.UID, util.EscapeQualifiedNameForDisk(name), nfsVolume.volName) } func (nfsVolume *nfs) TearDown() error { return nfsVolume.TearDownAt(nfsVolume.GetPath()) } func (nfsVolume *nfs) TearDownAt(dir string) error { mountpoint, err := nfsVolume.mounter.IsMountPoint(dir) if err != nil { glog.Errorf("Error checking IsMountPoint: %v", err) return err } if !mountpoint { return os.Remove(dir) } if err := nfsVolume.mounter.Unmount(dir); err != nil { glog.Errorf("Unmounting failed: %v", err) return err } mountpoint, mntErr := nfsVolume.mounter.IsMountPoint(dir) if mntErr != nil { glog.Errorf("IsMountpoint check failed: %v", mntErr) return mntErr } if !mountpoint { if err := os.Remove(dir); err != nil { return err } } return nil }