mirror of https://github.com/helm/helm
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
687 lines
21 KiB
687 lines
21 KiB
/*
|
|
Copyright 2016 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 kube // import "k8s.io/helm/pkg/kube"
|
|
|
|
import (
|
|
"bytes"
|
|
"encoding/json"
|
|
goerrors "errors"
|
|
"fmt"
|
|
"io"
|
|
"log"
|
|
"strings"
|
|
"time"
|
|
|
|
jsonpatch "github.com/evanphx/json-patch"
|
|
"k8s.io/kubernetes/pkg/api"
|
|
"k8s.io/kubernetes/pkg/api/errors"
|
|
"k8s.io/kubernetes/pkg/api/meta"
|
|
"k8s.io/kubernetes/pkg/api/v1"
|
|
apps "k8s.io/kubernetes/pkg/apis/apps/v1beta1"
|
|
batchinternal "k8s.io/kubernetes/pkg/apis/batch"
|
|
batch "k8s.io/kubernetes/pkg/apis/batch/v1"
|
|
extensionsinternal "k8s.io/kubernetes/pkg/apis/extensions"
|
|
extensions "k8s.io/kubernetes/pkg/apis/extensions/v1beta1"
|
|
"k8s.io/kubernetes/pkg/client/clientset_generated/internalclientset"
|
|
conditions "k8s.io/kubernetes/pkg/client/unversioned"
|
|
"k8s.io/kubernetes/pkg/client/unversioned/clientcmd"
|
|
"k8s.io/kubernetes/pkg/fields"
|
|
"k8s.io/kubernetes/pkg/kubectl"
|
|
cmdutil "k8s.io/kubernetes/pkg/kubectl/cmd/util"
|
|
"k8s.io/kubernetes/pkg/kubectl/resource"
|
|
"k8s.io/kubernetes/pkg/labels"
|
|
"k8s.io/kubernetes/pkg/runtime"
|
|
"k8s.io/kubernetes/pkg/util/strategicpatch"
|
|
"k8s.io/kubernetes/pkg/util/wait"
|
|
"k8s.io/kubernetes/pkg/watch"
|
|
)
|
|
|
|
// ErrNoObjectsVisited indicates that during a visit operation, no matching objects were found.
|
|
var ErrNoObjectsVisited = goerrors.New("no objects visited")
|
|
|
|
// Client represents a client capable of communicating with the Kubernetes API.
|
|
type Client struct {
|
|
cmdutil.Factory
|
|
// SchemaCacheDir is the path for loading cached schema.
|
|
SchemaCacheDir string
|
|
}
|
|
|
|
// New create a new Client
|
|
func New(config clientcmd.ClientConfig) *Client {
|
|
return &Client{
|
|
Factory: cmdutil.NewFactory(config),
|
|
SchemaCacheDir: clientcmd.RecommendedSchemaFile,
|
|
}
|
|
}
|
|
|
|
// ResourceActorFunc performs an action on a single resource.
|
|
type ResourceActorFunc func(*resource.Info) error
|
|
|
|
// Create creates kubernetes resources from an io.reader
|
|
//
|
|
// Namespace will set the namespace
|
|
func (c *Client) Create(namespace string, reader io.Reader, timeout int64, shouldWait bool) error {
|
|
client, err := c.ClientSet()
|
|
if err != nil {
|
|
return err
|
|
}
|
|
if err := ensureNamespace(client, namespace); err != nil {
|
|
return err
|
|
}
|
|
infos, buildErr := c.BuildUnstructured(namespace, reader)
|
|
if buildErr != nil {
|
|
return buildErr
|
|
}
|
|
if err := perform(c, namespace, infos, createResource); err != nil {
|
|
return err
|
|
}
|
|
if shouldWait {
|
|
return c.waitForResources(time.Duration(timeout)*time.Second, infos)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
func (c *Client) newBuilder(namespace string, reader io.Reader) *resource.Result {
|
|
schema, err := c.Validator(true, c.SchemaCacheDir)
|
|
if err != nil {
|
|
log.Printf("warning: failed to load schema: %s", err)
|
|
}
|
|
return c.NewBuilder().
|
|
ContinueOnError().
|
|
Schema(schema).
|
|
NamespaceParam(namespace).
|
|
DefaultNamespace().
|
|
Stream(reader, "").
|
|
Flatten().
|
|
Do()
|
|
}
|
|
|
|
// BuildUnstructured validates for Kubernetes objects and returns unstructured infos.
|
|
func (c *Client) BuildUnstructured(namespace string, reader io.Reader) (Result, error) {
|
|
schema, err := c.Validator(true, c.SchemaCacheDir)
|
|
if err != nil {
|
|
log.Printf("warning: failed to load schema: %s", err)
|
|
}
|
|
|
|
mapper, typer, err := c.UnstructuredObject()
|
|
if err != nil {
|
|
log.Printf("failed to load mapper: %s", err)
|
|
return nil, err
|
|
}
|
|
var result Result
|
|
result, err = resource.NewBuilder(mapper, typer, resource.ClientMapperFunc(c.UnstructuredClientForMapping), runtime.UnstructuredJSONScheme).
|
|
ContinueOnError().
|
|
Schema(schema).
|
|
NamespaceParam(namespace).
|
|
DefaultNamespace().
|
|
Stream(reader, "").
|
|
Flatten().
|
|
Do().Infos()
|
|
return result, scrubValidationError(err)
|
|
}
|
|
|
|
// Build validates for Kubernetes objects and returns resource Infos from a io.Reader.
|
|
func (c *Client) Build(namespace string, reader io.Reader) (Result, error) {
|
|
var result Result
|
|
result, err := c.newBuilder(namespace, reader).Infos()
|
|
return result, scrubValidationError(err)
|
|
}
|
|
|
|
// Get gets kubernetes resources as pretty printed string
|
|
//
|
|
// Namespace will set the namespace
|
|
func (c *Client) Get(namespace string, reader io.Reader) (string, error) {
|
|
// Since we don't know what order the objects come in, let's group them by the types, so
|
|
// that when we print them, they come looking good (headers apply to subgroups, etc.)
|
|
objs := make(map[string][]runtime.Object)
|
|
infos, err := c.BuildUnstructured(namespace, reader)
|
|
if err != nil {
|
|
return "", err
|
|
}
|
|
err = perform(c, namespace, infos, func(info *resource.Info) error {
|
|
log.Printf("Doing get for: '%s'", info.Name)
|
|
obj, err := resource.NewHelper(info.Client, info.Mapping).Get(info.Namespace, info.Name, info.Export)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
// We need to grab the ObjectReference so we can correctly group the objects.
|
|
or, err := api.GetReference(obj)
|
|
if err != nil {
|
|
log.Printf("FAILED GetReference for: %#v\n%v", obj, err)
|
|
return err
|
|
}
|
|
|
|
// Use APIVersion/Kind as grouping mechanism. I'm not sure if you can have multiple
|
|
// versions per cluster, but this certainly won't hurt anything, so let's be safe.
|
|
objType := or.APIVersion + "/" + or.Kind
|
|
objs[objType] = append(objs[objType], obj)
|
|
return nil
|
|
})
|
|
if err != nil {
|
|
return "", err
|
|
}
|
|
|
|
// Ok, now we have all the objects grouped by types (say, by v1/Pod, v1/Service, etc.), so
|
|
// spin through them and print them. Printer is cool since it prints the header only when
|
|
// an object type changes, so we can just rely on that. Problem is it doesn't seem to keep
|
|
// track of tab widths
|
|
buf := new(bytes.Buffer)
|
|
p := kubectl.NewHumanReadablePrinter(kubectl.PrintOptions{})
|
|
for t, ot := range objs {
|
|
if _, err = buf.WriteString("==> " + t + "\n"); err != nil {
|
|
return "", err
|
|
}
|
|
for _, o := range ot {
|
|
if err := p.PrintObj(o, buf); err != nil {
|
|
log.Printf("failed to print object type '%s', object: '%s' :\n %v", t, o, err)
|
|
return "", err
|
|
}
|
|
}
|
|
if _, err := buf.WriteString("\n"); err != nil {
|
|
return "", err
|
|
}
|
|
}
|
|
return buf.String(), nil
|
|
}
|
|
|
|
// Update reads in the current configuration and a target configuration from io.reader
|
|
// and creates resources that don't already exists, updates resources that have been modified
|
|
// in the target configuration and deletes resources from the current configuration that are
|
|
// not present in the target configuration
|
|
//
|
|
// Namespace will set the namespaces
|
|
func (c *Client) Update(namespace string, originalReader, targetReader io.Reader, recreate bool, timeout int64, shouldWait bool) error {
|
|
original, err := c.BuildUnstructured(namespace, originalReader)
|
|
if err != nil {
|
|
return fmt.Errorf("failed decoding reader into objects: %s", err)
|
|
}
|
|
|
|
target, err := c.BuildUnstructured(namespace, targetReader)
|
|
if err != nil {
|
|
return fmt.Errorf("failed decoding reader into objects: %s", err)
|
|
}
|
|
|
|
updateErrors := []string{}
|
|
|
|
err = target.Visit(func(info *resource.Info, err error) error {
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
helper := resource.NewHelper(info.Client, info.Mapping)
|
|
if _, err := helper.Get(info.Namespace, info.Name, info.Export); err != nil {
|
|
if !errors.IsNotFound(err) {
|
|
return fmt.Errorf("Could not get information about the resource: err: %s", err)
|
|
}
|
|
|
|
// Since the resource does not exist, create it.
|
|
if err := createResource(info); err != nil {
|
|
return fmt.Errorf("failed to create resource: %s", err)
|
|
}
|
|
|
|
kind := info.Mapping.GroupVersionKind.Kind
|
|
log.Printf("Created a new %s called %s\n", kind, info.Name)
|
|
return nil
|
|
}
|
|
|
|
originalInfo := original.Get(info)
|
|
if originalInfo == nil {
|
|
return fmt.Errorf("no resource with the name %s found", info.Name)
|
|
}
|
|
|
|
if err := updateResource(c, info, originalInfo.Object, recreate); err != nil {
|
|
log.Printf("error updating the resource %s:\n\t %v", info.Name, err)
|
|
updateErrors = append(updateErrors, err.Error())
|
|
}
|
|
|
|
return nil
|
|
})
|
|
|
|
switch {
|
|
case err != nil:
|
|
return err
|
|
case len(updateErrors) != 0:
|
|
return fmt.Errorf(strings.Join(updateErrors, " && "))
|
|
}
|
|
|
|
for _, info := range original.Difference(target) {
|
|
log.Printf("Deleting %s in %s...", info.Name, info.Namespace)
|
|
if err := deleteResource(c, info); err != nil {
|
|
log.Printf("Failed to delete %s, err: %s", info.Name, err)
|
|
}
|
|
}
|
|
if shouldWait {
|
|
return c.waitForResources(time.Duration(timeout)*time.Second, target)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// Delete deletes kubernetes resources from an io.reader
|
|
//
|
|
// Namespace will set the namespace
|
|
func (c *Client) Delete(namespace string, reader io.Reader) error {
|
|
infos, err := c.BuildUnstructured(namespace, reader)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
return perform(c, namespace, infos, func(info *resource.Info) error {
|
|
log.Printf("Starting delete for %s %s", info.Name, info.Mapping.GroupVersionKind.Kind)
|
|
err := deleteResource(c, info)
|
|
return skipIfNotFound(err)
|
|
})
|
|
}
|
|
|
|
func skipIfNotFound(err error) error {
|
|
if errors.IsNotFound(err) {
|
|
log.Printf("%v", err)
|
|
return nil
|
|
}
|
|
return err
|
|
}
|
|
|
|
func watchTimeout(t time.Duration) ResourceActorFunc {
|
|
return func(info *resource.Info) error {
|
|
return watchUntilReady(t, info)
|
|
}
|
|
}
|
|
|
|
// WatchUntilReady watches the resource given in the reader, and waits until it is ready.
|
|
//
|
|
// This function is mainly for hook implementations. It watches for a resource to
|
|
// hit a particular milestone. The milestone depends on the Kind.
|
|
//
|
|
// For most kinds, it checks to see if the resource is marked as Added or Modified
|
|
// by the Kubernetes event stream. For some kinds, it does more:
|
|
//
|
|
// - Jobs: A job is marked "Ready" when it has successfully completed. This is
|
|
// ascertained by watching the Status fields in a job's output.
|
|
//
|
|
// Handling for other kinds will be added as necessary.
|
|
func (c *Client) WatchUntilReady(namespace string, reader io.Reader, timeout int64, shouldWait bool) error {
|
|
infos, err := c.Build(namespace, reader)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
// For jobs, there's also the option to do poll c.Jobs(namespace).Get():
|
|
// https://github.com/adamreese/kubernetes/blob/master/test/e2e/job.go#L291-L300
|
|
return perform(c, namespace, infos, watchTimeout(time.Duration(timeout)*time.Second))
|
|
}
|
|
|
|
func perform(c *Client, namespace string, infos Result, fn ResourceActorFunc) error {
|
|
if len(infos) == 0 {
|
|
return ErrNoObjectsVisited
|
|
}
|
|
|
|
for _, info := range infos {
|
|
if err := fn(info); err != nil {
|
|
return err
|
|
}
|
|
}
|
|
return nil
|
|
}
|
|
|
|
func createResource(info *resource.Info) error {
|
|
obj, err := resource.NewHelper(info.Client, info.Mapping).Create(info.Namespace, true, info.Object)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
return info.Refresh(obj, true)
|
|
}
|
|
|
|
func deleteResource(c *Client, info *resource.Info) error {
|
|
reaper, err := c.Reaper(info.Mapping)
|
|
if err != nil {
|
|
// If there is no reaper for this resources, delete it.
|
|
if kubectl.IsNoSuchReaperError(err) {
|
|
return resource.NewHelper(info.Client, info.Mapping).Delete(info.Namespace, info.Name)
|
|
}
|
|
return err
|
|
}
|
|
log.Printf("Using reaper for deleting %s", info.Name)
|
|
return reaper.Stop(info.Namespace, info.Name, 0, nil)
|
|
}
|
|
|
|
func createPatch(mapping *meta.RESTMapping, target, current runtime.Object) ([]byte, api.PatchType, error) {
|
|
oldData, err := json.Marshal(current)
|
|
if err != nil {
|
|
return nil, api.StrategicMergePatchType, fmt.Errorf("serializing current configuration: %s", err)
|
|
}
|
|
newData, err := json.Marshal(target)
|
|
if err != nil {
|
|
return nil, api.StrategicMergePatchType, fmt.Errorf("serializing target configuration: %s", err)
|
|
}
|
|
|
|
if api.Semantic.DeepEqual(oldData, newData) {
|
|
return nil, api.StrategicMergePatchType, nil
|
|
}
|
|
|
|
// Get a versioned object
|
|
versionedObject, err := api.Scheme.New(mapping.GroupVersionKind)
|
|
switch {
|
|
case runtime.IsNotRegisteredError(err):
|
|
// fall back to generic JSON merge patch
|
|
patch, err := jsonpatch.CreateMergePatch(oldData, newData)
|
|
return patch, api.MergePatchType, err
|
|
case err != nil:
|
|
return nil, api.StrategicMergePatchType, fmt.Errorf("failed to get versionedObject: %s", err)
|
|
default:
|
|
log.Printf("generating strategic merge patch for %T", target)
|
|
patch, err := strategicpatch.CreateTwoWayMergePatch(oldData, newData, versionedObject)
|
|
return patch, api.StrategicMergePatchType, err
|
|
}
|
|
}
|
|
|
|
func updateResource(c *Client, target *resource.Info, currentObj runtime.Object, recreate bool) error {
|
|
patch, patchType, err := createPatch(target.Mapping, target.Object, currentObj)
|
|
if err != nil {
|
|
return fmt.Errorf("failed to create patch: %s", err)
|
|
}
|
|
if patch == nil {
|
|
log.Printf("Looks like there are no changes for %s", target.Name)
|
|
return nil
|
|
}
|
|
|
|
// send patch to server
|
|
helper := resource.NewHelper(target.Client, target.Mapping)
|
|
obj, err := helper.Patch(target.Namespace, target.Name, patchType, patch)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
target.Refresh(obj, true)
|
|
|
|
if !recreate {
|
|
return nil
|
|
}
|
|
|
|
selector, err := getSelectorFromObject(currentObj)
|
|
if err != nil {
|
|
return nil
|
|
}
|
|
client, _ := c.ClientSet()
|
|
return recreatePods(client, target.Namespace, selector)
|
|
}
|
|
|
|
func getSelectorFromObject(obj runtime.Object) (map[string]string, error) {
|
|
switch typed := obj.(type) {
|
|
case *v1.ReplicationController:
|
|
return typed.Spec.Selector, nil
|
|
case *extensions.ReplicaSet:
|
|
return typed.Spec.Selector.MatchLabels, nil
|
|
case *extensions.Deployment:
|
|
return typed.Spec.Selector.MatchLabels, nil
|
|
case *extensions.DaemonSet:
|
|
return typed.Spec.Selector.MatchLabels, nil
|
|
case *batch.Job:
|
|
return typed.Spec.Selector.MatchLabels, nil
|
|
case *apps.StatefulSet:
|
|
return typed.Spec.Selector.MatchLabels, nil
|
|
default:
|
|
return nil, fmt.Errorf("Unsupported kind when getting selector: %v", obj)
|
|
}
|
|
}
|
|
|
|
func recreatePods(client *internalclientset.Clientset, namespace string, selector map[string]string) error {
|
|
pods, err := client.Pods(namespace).List(api.ListOptions{
|
|
FieldSelector: fields.Everything(),
|
|
LabelSelector: labels.Set(selector).AsSelector(),
|
|
})
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
// Restart pods
|
|
for _, pod := range pods.Items {
|
|
log.Printf("Restarting pod: %v/%v", pod.Namespace, pod.Name)
|
|
|
|
// Delete each pod for get them restarted with changed spec.
|
|
if err := client.Pods(pod.Namespace).Delete(pod.Name, api.NewPreconditionDeleteOptions(string(pod.UID))); err != nil {
|
|
return err
|
|
}
|
|
}
|
|
return nil
|
|
}
|
|
|
|
func watchUntilReady(timeout time.Duration, info *resource.Info) error {
|
|
w, err := resource.NewHelper(info.Client, info.Mapping).WatchSingle(info.Namespace, info.Name, info.ResourceVersion)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
kind := info.Mapping.GroupVersionKind.Kind
|
|
log.Printf("Watching for changes to %s %s with timeout of %v", kind, info.Name, timeout)
|
|
|
|
// What we watch for depends on the Kind.
|
|
// - For a Job, we watch for completion.
|
|
// - For all else, we watch until Ready.
|
|
// In the future, we might want to add some special logic for types
|
|
// like Ingress, Volume, etc.
|
|
|
|
_, err = watch.Until(timeout, w, func(e watch.Event) (bool, error) {
|
|
switch e.Type {
|
|
case watch.Added, watch.Modified:
|
|
// For things like a secret or a config map, this is the best indicator
|
|
// we get. We care mostly about jobs, where what we want to see is
|
|
// the status go into a good state. For other types, like ReplicaSet
|
|
// we don't really do anything to support these as hooks.
|
|
log.Printf("Add/Modify event for %s: %v", info.Name, e.Type)
|
|
if kind == "Job" {
|
|
return waitForJob(e, info.Name)
|
|
}
|
|
return true, nil
|
|
case watch.Deleted:
|
|
log.Printf("Deleted event for %s", info.Name)
|
|
return true, nil
|
|
case watch.Error:
|
|
// Handle error and return with an error.
|
|
log.Printf("Error event for %s", info.Name)
|
|
return true, fmt.Errorf("Failed to deploy %s", info.Name)
|
|
default:
|
|
return false, nil
|
|
}
|
|
})
|
|
return err
|
|
}
|
|
|
|
func podsReady(pods []api.Pod) bool {
|
|
for _, pod := range pods {
|
|
if !api.IsPodReady(&pod) {
|
|
return false
|
|
}
|
|
}
|
|
return true
|
|
}
|
|
|
|
func servicesReady(svc []api.Service) bool {
|
|
for _, s := range svc {
|
|
if !api.IsServiceIPSet(&s) {
|
|
return false
|
|
}
|
|
// This checks if the service has a LoadBalancer and that balancer has an Ingress defined
|
|
if s.Spec.Type == api.ServiceTypeLoadBalancer && s.Status.LoadBalancer.Ingress == nil {
|
|
return false
|
|
}
|
|
}
|
|
return true
|
|
}
|
|
|
|
func volumesReady(vols []api.PersistentVolumeClaim) bool {
|
|
for _, v := range vols {
|
|
if v.Status.Phase != api.ClaimBound {
|
|
return false
|
|
}
|
|
}
|
|
return true
|
|
}
|
|
|
|
func getPods(client *internalclientset.Clientset, namespace string, selector map[string]string) ([]api.Pod, error) {
|
|
list, err := client.Pods(namespace).List(api.ListOptions{
|
|
FieldSelector: fields.Everything(),
|
|
LabelSelector: labels.Set(selector).AsSelector(),
|
|
})
|
|
return list.Items, err
|
|
}
|
|
|
|
// waitForResources polls to get the current status of all pods, PVCs, and Services
|
|
// until all are ready or a timeout is reached
|
|
func (c *Client) waitForResources(timeout time.Duration, created Result) error {
|
|
log.Printf("beginning wait for resources with timeout of %v", timeout)
|
|
client, _ := c.ClientSet()
|
|
return wait.Poll(2*time.Second, timeout, func() (bool, error) {
|
|
pods := []api.Pod{}
|
|
services := []api.Service{}
|
|
pvc := []api.PersistentVolumeClaim{}
|
|
for _, v := range created {
|
|
switch value := v.Object.(type) {
|
|
case (*api.ReplicationController):
|
|
list, err := getPods(client, value.Namespace, value.Spec.Selector)
|
|
if err != nil {
|
|
return false, err
|
|
}
|
|
pods = append(pods, list...)
|
|
case (*api.Pod):
|
|
pod, err := client.Pods(value.Namespace).Get(value.Name)
|
|
if err != nil {
|
|
return false, err
|
|
}
|
|
pods = append(pods, *pod)
|
|
case (*extensionsinternal.Deployment):
|
|
// Get the RS children first
|
|
rs, err := client.ReplicaSets(value.Namespace).List(api.ListOptions{
|
|
FieldSelector: fields.Everything(),
|
|
LabelSelector: labels.Set(value.Spec.Selector.MatchLabels).AsSelector(),
|
|
})
|
|
if err != nil {
|
|
return false, err
|
|
}
|
|
for _, r := range rs.Items {
|
|
list, err := getPods(client, value.Namespace, r.Spec.Selector.MatchLabels)
|
|
if err != nil {
|
|
return false, err
|
|
}
|
|
pods = append(pods, list...)
|
|
}
|
|
case (*extensionsinternal.DaemonSet):
|
|
list, err := getPods(client, value.Namespace, value.Spec.Selector.MatchLabels)
|
|
if err != nil {
|
|
return false, err
|
|
}
|
|
pods = append(pods, list...)
|
|
case (*apps.StatefulSet):
|
|
list, err := getPods(client, value.Namespace, value.Spec.Selector.MatchLabels)
|
|
if err != nil {
|
|
return false, err
|
|
}
|
|
pods = append(pods, list...)
|
|
case (*extensionsinternal.ReplicaSet):
|
|
list, err := getPods(client, value.Namespace, value.Spec.Selector.MatchLabels)
|
|
if err != nil {
|
|
return false, err
|
|
}
|
|
pods = append(pods, list...)
|
|
case (*api.PersistentVolumeClaim):
|
|
claim, err := client.PersistentVolumeClaims(value.Namespace).Get(value.Name)
|
|
if err != nil {
|
|
return false, err
|
|
}
|
|
pvc = append(pvc, *claim)
|
|
case (*api.Service):
|
|
svc, err := client.Services(value.Namespace).Get(value.Name)
|
|
if err != nil {
|
|
return false, err
|
|
}
|
|
services = append(services, *svc)
|
|
}
|
|
}
|
|
return podsReady(pods) && servicesReady(services) && volumesReady(pvc), nil
|
|
})
|
|
}
|
|
|
|
// waitForJob is a helper that waits for a job to complete.
|
|
//
|
|
// This operates on an event returned from a watcher.
|
|
func waitForJob(e watch.Event, name string) (bool, error) {
|
|
o, ok := e.Object.(*batchinternal.Job)
|
|
if !ok {
|
|
return true, fmt.Errorf("Expected %s to be a *batch.Job, got %T", name, e.Object)
|
|
}
|
|
|
|
for _, c := range o.Status.Conditions {
|
|
if c.Type == batchinternal.JobComplete && c.Status == api.ConditionTrue {
|
|
return true, nil
|
|
} else if c.Type == batchinternal.JobFailed && c.Status == api.ConditionTrue {
|
|
return true, fmt.Errorf("Job failed: %s", c.Reason)
|
|
}
|
|
}
|
|
|
|
log.Printf("%s: Jobs active: %d, jobs failed: %d, jobs succeeded: %d", name, o.Status.Active, o.Status.Failed, o.Status.Succeeded)
|
|
return false, nil
|
|
}
|
|
|
|
// scrubValidationError removes kubectl info from the message
|
|
func scrubValidationError(err error) error {
|
|
if err == nil {
|
|
return nil
|
|
}
|
|
const stopValidateMessage = "if you choose to ignore these errors, turn validation off with --validate=false"
|
|
|
|
if strings.Contains(err.Error(), stopValidateMessage) {
|
|
return goerrors.New(strings.Replace(err.Error(), "; "+stopValidateMessage, "", -1))
|
|
}
|
|
return err
|
|
}
|
|
|
|
// WaitAndGetCompletedPodPhase waits up to a timeout until a pod enters a completed phase
|
|
// and returns said phase (PodSucceeded or PodFailed qualify)
|
|
func (c *Client) WaitAndGetCompletedPodPhase(namespace string, reader io.Reader, timeout time.Duration) (api.PodPhase, error) {
|
|
infos, err := c.Build(namespace, reader)
|
|
if err != nil {
|
|
return api.PodUnknown, err
|
|
}
|
|
info := infos[0]
|
|
|
|
kind := info.Mapping.GroupVersionKind.Kind
|
|
if kind != "Pod" {
|
|
return api.PodUnknown, fmt.Errorf("%s is not a Pod", info.Name)
|
|
}
|
|
|
|
if err := watchPodUntilComplete(timeout, info); err != nil {
|
|
return api.PodUnknown, err
|
|
}
|
|
|
|
if err := info.Get(); err != nil {
|
|
return api.PodUnknown, err
|
|
}
|
|
status := info.Object.(*api.Pod).Status.Phase
|
|
|
|
return status, nil
|
|
}
|
|
|
|
func watchPodUntilComplete(timeout time.Duration, info *resource.Info) error {
|
|
w, err := resource.NewHelper(info.Client, info.Mapping).WatchSingle(info.Namespace, info.Name, info.ResourceVersion)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
log.Printf("Watching pod %s for completion with timeout of %v", info.Name, timeout)
|
|
_, err = watch.Until(timeout, w, func(e watch.Event) (bool, error) {
|
|
return conditions.PodCompleted(e)
|
|
})
|
|
|
|
return err
|
|
}
|