mirror of
https://github.com/k3s-io/k3s.git
synced 2024-06-07 19:41:36 +00:00
346 lines
12 KiB
Go
346 lines
12 KiB
Go
/*
|
|
Copyright 2014 The Kubernetes Authors.
|
|
|
|
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 app implements a Server object for running the scheduler.
|
|
package app
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"io"
|
|
"net/http"
|
|
"os"
|
|
goruntime "runtime"
|
|
|
|
"github.com/spf13/cobra"
|
|
|
|
utilerrors "k8s.io/apimachinery/pkg/util/errors"
|
|
"k8s.io/apiserver/pkg/authentication/authenticator"
|
|
"k8s.io/apiserver/pkg/authorization/authorizer"
|
|
genericapifilters "k8s.io/apiserver/pkg/endpoints/filters"
|
|
apirequest "k8s.io/apiserver/pkg/endpoints/request"
|
|
genericfilters "k8s.io/apiserver/pkg/server/filters"
|
|
"k8s.io/apiserver/pkg/server/healthz"
|
|
"k8s.io/apiserver/pkg/server/mux"
|
|
"k8s.io/apiserver/pkg/server/routes"
|
|
"k8s.io/apiserver/pkg/util/term"
|
|
corev1 "k8s.io/client-go/kubernetes/typed/core/v1"
|
|
"k8s.io/client-go/tools/leaderelection"
|
|
cliflag "k8s.io/component-base/cli/flag"
|
|
"k8s.io/component-base/cli/globalflag"
|
|
"k8s.io/component-base/metrics/legacyregistry"
|
|
"k8s.io/klog"
|
|
schedulerserverconfig "k8s.io/kubernetes/cmd/kube-scheduler/app/config"
|
|
"k8s.io/kubernetes/cmd/kube-scheduler/app/options"
|
|
"k8s.io/kubernetes/pkg/api/legacyscheme"
|
|
"k8s.io/kubernetes/pkg/scheduler"
|
|
"k8s.io/kubernetes/pkg/scheduler/algorithmprovider"
|
|
kubeschedulerconfig "k8s.io/kubernetes/pkg/scheduler/apis/config"
|
|
framework "k8s.io/kubernetes/pkg/scheduler/framework/v1alpha1"
|
|
"k8s.io/kubernetes/pkg/scheduler/metrics"
|
|
"k8s.io/kubernetes/pkg/util/configz"
|
|
utilflag "k8s.io/kubernetes/pkg/util/flag"
|
|
"k8s.io/kubernetes/pkg/version"
|
|
"k8s.io/kubernetes/pkg/version/verflag"
|
|
)
|
|
|
|
// Option configures a framework.Registry.
|
|
type Option func(framework.Registry) error
|
|
|
|
// NewSchedulerCommand creates a *cobra.Command object with default parameters and registryOptions
|
|
func NewSchedulerCommand(registryOptions ...Option) *cobra.Command {
|
|
opts, err := options.NewOptions()
|
|
if err != nil {
|
|
klog.Fatalf("unable to initialize command options: %v", err)
|
|
}
|
|
|
|
cmd := &cobra.Command{
|
|
Use: "kube-scheduler",
|
|
Long: `The Kubernetes scheduler is a policy-rich, topology-aware,
|
|
workload-specific function that significantly impacts availability, performance,
|
|
and capacity. The scheduler needs to take into account individual and collective
|
|
resource requirements, quality of service requirements, hardware/software/policy
|
|
constraints, affinity and anti-affinity specifications, data locality, inter-workload
|
|
interference, deadlines, and so on. Workload-specific requirements will be exposed
|
|
through the API as necessary.`,
|
|
Run: func(cmd *cobra.Command, args []string) {
|
|
if err := runCommand(cmd, args, opts, registryOptions...); err != nil {
|
|
fmt.Fprintf(os.Stderr, "%v\n", err)
|
|
os.Exit(1)
|
|
}
|
|
},
|
|
}
|
|
fs := cmd.Flags()
|
|
namedFlagSets := opts.Flags()
|
|
verflag.AddFlags(namedFlagSets.FlagSet("global"))
|
|
globalflag.AddGlobalFlags(namedFlagSets.FlagSet("global"), cmd.Name())
|
|
for _, f := range namedFlagSets.FlagSets {
|
|
fs.AddFlagSet(f)
|
|
}
|
|
|
|
usageFmt := "Usage:\n %s\n"
|
|
cols, _, _ := term.TerminalSize(cmd.OutOrStdout())
|
|
cmd.SetUsageFunc(func(cmd *cobra.Command) error {
|
|
fmt.Fprintf(cmd.OutOrStderr(), usageFmt, cmd.UseLine())
|
|
cliflag.PrintSections(cmd.OutOrStderr(), namedFlagSets, cols)
|
|
return nil
|
|
})
|
|
cmd.SetHelpFunc(func(cmd *cobra.Command, args []string) {
|
|
fmt.Fprintf(cmd.OutOrStdout(), "%s\n\n"+usageFmt, cmd.Long, cmd.UseLine())
|
|
cliflag.PrintSections(cmd.OutOrStdout(), namedFlagSets, cols)
|
|
})
|
|
cmd.MarkFlagFilename("config", "yaml", "yml", "json")
|
|
|
|
return cmd
|
|
}
|
|
|
|
// runCommand runs the scheduler.
|
|
func runCommand(cmd *cobra.Command, args []string, opts *options.Options, registryOptions ...Option) error {
|
|
verflag.PrintAndExitIfRequested()
|
|
utilflag.PrintFlags(cmd.Flags())
|
|
|
|
if len(args) != 0 {
|
|
fmt.Fprint(os.Stderr, "arguments are not supported\n")
|
|
}
|
|
|
|
if errs := opts.Validate(); len(errs) > 0 {
|
|
fmt.Fprintf(os.Stderr, "%v\n", utilerrors.NewAggregate(errs))
|
|
os.Exit(1)
|
|
}
|
|
|
|
if len(opts.WriteConfigTo) > 0 {
|
|
if err := options.WriteConfigFile(opts.WriteConfigTo, &opts.ComponentConfig); err != nil {
|
|
fmt.Fprintf(os.Stderr, "%v\n", err)
|
|
os.Exit(1)
|
|
}
|
|
klog.Infof("Wrote configuration to: %s\n", opts.WriteConfigTo)
|
|
}
|
|
|
|
c, err := opts.Config()
|
|
if err != nil {
|
|
fmt.Fprintf(os.Stderr, "%v\n", err)
|
|
os.Exit(1)
|
|
}
|
|
|
|
stopCh := make(chan struct{})
|
|
// Get the completed config
|
|
cc := c.Complete()
|
|
|
|
// To help debugging, immediately log version
|
|
klog.Infof("Version: %+v", version.Get())
|
|
|
|
// Apply algorithms based on feature gates.
|
|
// TODO: make configurable?
|
|
algorithmprovider.ApplyFeatureGates()
|
|
|
|
// Configz registration.
|
|
if cz, err := configz.New("componentconfig"); err == nil {
|
|
cz.Set(cc.ComponentConfig)
|
|
} else {
|
|
return fmt.Errorf("unable to register configz: %s", err)
|
|
}
|
|
|
|
return Run(cc, stopCh, registryOptions...)
|
|
}
|
|
|
|
// Run executes the scheduler based on the given configuration. It only return on error or when stopCh is closed.
|
|
func Run(cc schedulerserverconfig.CompletedConfig, stopCh <-chan struct{}, registryOptions ...Option) error {
|
|
// To help debugging, immediately log version
|
|
klog.V(1).Infof("Starting Kubernetes Scheduler version %+v", version.Get())
|
|
|
|
registry := framework.NewRegistry()
|
|
for _, option := range registryOptions {
|
|
if err := option(registry); err != nil {
|
|
return err
|
|
}
|
|
}
|
|
|
|
// Create the scheduler.
|
|
sched, err := scheduler.New(cc.Client,
|
|
cc.InformerFactory.Core().V1().Nodes(),
|
|
cc.PodInformer,
|
|
cc.InformerFactory.Core().V1().PersistentVolumes(),
|
|
cc.InformerFactory.Core().V1().PersistentVolumeClaims(),
|
|
cc.InformerFactory.Core().V1().ReplicationControllers(),
|
|
cc.InformerFactory.Apps().V1().ReplicaSets(),
|
|
cc.InformerFactory.Apps().V1().StatefulSets(),
|
|
cc.InformerFactory.Core().V1().Services(),
|
|
cc.InformerFactory.Policy().V1beta1().PodDisruptionBudgets(),
|
|
cc.InformerFactory.Storage().V1().StorageClasses(),
|
|
cc.InformerFactory.Storage().V1beta1().CSINodes(),
|
|
cc.Recorder,
|
|
cc.ComponentConfig.AlgorithmSource,
|
|
stopCh,
|
|
registry,
|
|
cc.ComponentConfig.Plugins,
|
|
cc.ComponentConfig.PluginConfig,
|
|
scheduler.WithName(cc.ComponentConfig.SchedulerName),
|
|
scheduler.WithHardPodAffinitySymmetricWeight(cc.ComponentConfig.HardPodAffinitySymmetricWeight),
|
|
scheduler.WithPreemptionDisabled(cc.ComponentConfig.DisablePreemption),
|
|
scheduler.WithPercentageOfNodesToScore(cc.ComponentConfig.PercentageOfNodesToScore),
|
|
scheduler.WithBindTimeoutSeconds(*cc.ComponentConfig.BindTimeoutSeconds))
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
// Prepare the event broadcaster.
|
|
if cc.Broadcaster != nil && cc.EventClient != nil {
|
|
cc.Broadcaster.StartRecordingToSink(stopCh)
|
|
}
|
|
if cc.LeaderElectionBroadcaster != nil && cc.CoreEventClient != nil {
|
|
cc.LeaderElectionBroadcaster.StartRecordingToSink(&corev1.EventSinkImpl{Interface: cc.CoreEventClient.Events("")})
|
|
}
|
|
// Setup healthz checks.
|
|
var checks []healthz.HealthChecker
|
|
if cc.ComponentConfig.LeaderElection.LeaderElect {
|
|
checks = append(checks, cc.LeaderElection.WatchDog)
|
|
}
|
|
|
|
// Start up the healthz server.
|
|
if cc.InsecureServing != nil {
|
|
separateMetrics := cc.InsecureMetricsServing != nil
|
|
handler := buildHandlerChain(newHealthzHandler(&cc.ComponentConfig, separateMetrics, checks...), nil, nil)
|
|
if err := cc.InsecureServing.Serve(handler, 0, stopCh); err != nil {
|
|
return fmt.Errorf("failed to start healthz server: %v", err)
|
|
}
|
|
}
|
|
if cc.InsecureMetricsServing != nil {
|
|
handler := buildHandlerChain(newMetricsHandler(&cc.ComponentConfig), nil, nil)
|
|
if err := cc.InsecureMetricsServing.Serve(handler, 0, stopCh); err != nil {
|
|
return fmt.Errorf("failed to start metrics server: %v", err)
|
|
}
|
|
}
|
|
if cc.SecureServing != nil {
|
|
handler := buildHandlerChain(newHealthzHandler(&cc.ComponentConfig, false, checks...), cc.Authentication.Authenticator, cc.Authorization.Authorizer)
|
|
// TODO: handle stoppedCh returned by c.SecureServing.Serve
|
|
if _, err := cc.SecureServing.Serve(handler, 0, stopCh); err != nil {
|
|
// fail early for secure handlers, removing the old error loop from above
|
|
return fmt.Errorf("failed to start secure server: %v", err)
|
|
}
|
|
}
|
|
|
|
// Start all informers.
|
|
go cc.PodInformer.Informer().Run(stopCh)
|
|
cc.InformerFactory.Start(stopCh)
|
|
|
|
// Wait for all caches to sync before scheduling.
|
|
cc.InformerFactory.WaitForCacheSync(stopCh)
|
|
|
|
// Prepare a reusable runCommand function.
|
|
run := func(ctx context.Context) {
|
|
sched.Run()
|
|
<-ctx.Done()
|
|
}
|
|
|
|
ctx, cancel := context.WithCancel(context.TODO()) // TODO once Run() accepts a context, it should be used here
|
|
defer cancel()
|
|
|
|
go func() {
|
|
select {
|
|
case <-stopCh:
|
|
cancel()
|
|
case <-ctx.Done():
|
|
}
|
|
}()
|
|
|
|
// If leader election is enabled, runCommand via LeaderElector until done and exit.
|
|
if cc.LeaderElection != nil {
|
|
cc.LeaderElection.Callbacks = leaderelection.LeaderCallbacks{
|
|
OnStartedLeading: run,
|
|
OnStoppedLeading: func() {
|
|
klog.Fatalf("leaderelection lost")
|
|
},
|
|
}
|
|
leaderElector, err := leaderelection.NewLeaderElector(*cc.LeaderElection)
|
|
if err != nil {
|
|
return fmt.Errorf("couldn't create leader elector: %v", err)
|
|
}
|
|
|
|
leaderElector.Run(ctx)
|
|
|
|
return fmt.Errorf("lost lease")
|
|
}
|
|
|
|
// Leader election is disabled, so runCommand inline until done.
|
|
run(ctx)
|
|
return fmt.Errorf("finished without leader elect")
|
|
}
|
|
|
|
// buildHandlerChain wraps the given handler with the standard filters.
|
|
func buildHandlerChain(handler http.Handler, authn authenticator.Request, authz authorizer.Authorizer) http.Handler {
|
|
requestInfoResolver := &apirequest.RequestInfoFactory{}
|
|
failedHandler := genericapifilters.Unauthorized(legacyscheme.Codecs, false)
|
|
|
|
handler = genericapifilters.WithAuthorization(handler, authz, legacyscheme.Codecs)
|
|
handler = genericapifilters.WithAuthentication(handler, authn, failedHandler, nil)
|
|
handler = genericapifilters.WithRequestInfo(handler, requestInfoResolver)
|
|
handler = genericapifilters.WithCacheControl(handler)
|
|
handler = genericfilters.WithPanicRecovery(handler)
|
|
|
|
return handler
|
|
}
|
|
|
|
func installMetricHandler(pathRecorderMux *mux.PathRecorderMux) {
|
|
configz.InstallHandler(pathRecorderMux)
|
|
defaultMetricsHandler := legacyregistry.Handler().ServeHTTP
|
|
pathRecorderMux.HandleFunc("/metrics", func(w http.ResponseWriter, req *http.Request) {
|
|
if req.Method == "DELETE" {
|
|
metrics.Reset()
|
|
io.WriteString(w, "metrics reset\n")
|
|
return
|
|
}
|
|
defaultMetricsHandler(w, req)
|
|
})
|
|
}
|
|
|
|
// newMetricsHandler builds a metrics server from the config.
|
|
func newMetricsHandler(config *kubeschedulerconfig.KubeSchedulerConfiguration) http.Handler {
|
|
pathRecorderMux := mux.NewPathRecorderMux("kube-scheduler")
|
|
installMetricHandler(pathRecorderMux)
|
|
if config.EnableProfiling {
|
|
routes.Profiling{}.Install(pathRecorderMux)
|
|
if config.EnableContentionProfiling {
|
|
goruntime.SetBlockProfileRate(1)
|
|
}
|
|
}
|
|
return pathRecorderMux
|
|
}
|
|
|
|
// newHealthzHandler creates a healthz server from the config, and will also
|
|
// embed the metrics handler if the healthz and metrics address configurations
|
|
// are the same.
|
|
func newHealthzHandler(config *kubeschedulerconfig.KubeSchedulerConfiguration, separateMetrics bool, checks ...healthz.HealthChecker) http.Handler {
|
|
pathRecorderMux := mux.NewPathRecorderMux("kube-scheduler")
|
|
healthz.InstallHandler(pathRecorderMux, checks...)
|
|
if !separateMetrics {
|
|
installMetricHandler(pathRecorderMux)
|
|
}
|
|
if config.EnableProfiling {
|
|
routes.Profiling{}.Install(pathRecorderMux)
|
|
if config.EnableContentionProfiling {
|
|
goruntime.SetBlockProfileRate(1)
|
|
}
|
|
}
|
|
return pathRecorderMux
|
|
}
|
|
|
|
// WithPlugin creates an Option based on plugin name and factory.
|
|
func WithPlugin(name string, factory framework.PluginFactory) Option {
|
|
return func(registry framework.Registry) error {
|
|
return registry.Register(name, factory)
|
|
}
|
|
}
|