operator.go

// Copyright 2016 The prometheus-operator 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 prometheus

import (
	"bytes"
	"compress/gzip"
	"fmt"
	"reflect"
	"strings"
	"time"

	monitoringv1 "github.com/coreos/prometheus-operator/pkg/apis/monitoring/v1"
	monitoringclient "github.com/coreos/prometheus-operator/pkg/client/versioned"
	"github.com/coreos/prometheus-operator/pkg/k8sutil"
	"github.com/coreos/prometheus-operator/pkg/listwatch"

	"github.com/go-kit/kit/log"
	"github.com/go-kit/kit/log/level"
	"github.com/mitchellh/hashstructure"
	"github.com/pkg/errors"
	"github.com/prometheus/client_golang/prometheus"
	appsv1 "k8s.io/api/apps/v1beta2"
	"k8s.io/api/core/v1"
	extensionsobj "k8s.io/apiextensions-apiserver/pkg/apis/apiextensions/v1beta1"
	apiextensionsclient "k8s.io/apiextensions-apiserver/pkg/client/clientset/clientset"
	apierrors "k8s.io/apimachinery/pkg/api/errors"
	"k8s.io/apimachinery/pkg/api/meta"
	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
	"k8s.io/apimachinery/pkg/fields"
	"k8s.io/apimachinery/pkg/labels"
	"k8s.io/apimachinery/pkg/runtime"
	utilruntime "k8s.io/apimachinery/pkg/util/runtime"
	"k8s.io/client-go/kubernetes"
	corev1client "k8s.io/client-go/kubernetes/typed/core/v1"
	"k8s.io/client-go/rest"
	"k8s.io/client-go/tools/cache"
	"k8s.io/client-go/util/workqueue"
)

const (
	resyncPeriod = 5 * time.Minute
)

// Operator manages life cycle of Prometheus deployments and
// monitoring configurations.
type Operator struct {
	kclient   kubernetes.Interface
	mclient   monitoringclient.Interface
	crdclient apiextensionsclient.Interface
	logger    log.Logger

	promInf cache.SharedIndexInformer
	smonInf cache.SharedIndexInformer
	ruleInf cache.SharedIndexInformer
	cmapInf cache.SharedIndexInformer
	secrInf cache.SharedIndexInformer
	ssetInf cache.SharedIndexInformer
	nsInf   cache.SharedIndexInformer

	queue workqueue.RateLimitingInterface

	reconcileErrorsCounter *prometheus.CounterVec

	// triggerByCounter is a set of counters keeping track of the amount
	// of times Prometheus Operator was triggered to reconcile its created
	// objects. It is split in the dimensions of Kubernetes objects and
	// corresponding actions (add, delete, update).
	triggerByCounter        *prometheus.CounterVec
	nodeAddressLookupErrors prometheus.Counter

	host                   string
	kubeletObjectName      string
	kubeletObjectNamespace string
	kubeletSyncEnabled     bool
	config                 Config

	configGenerator *configGenerator
}

type Labels struct {
	LabelsString string
	LabelsMap    map[string]string
}

// Implement the flag.Value interface
func (labels *Labels) String() string {
	return labels.LabelsString
}

// Merge labels create a new map with labels merged.
func (labels *Labels) Merge(otherLabels map[string]string) map[string]string {
	mergedLabels := map[string]string{}

	for key, value := range otherLabels {
		mergedLabels[key] = value
	}

	for key, value := range labels.LabelsMap {
		mergedLabels[key] = value
	}
	return mergedLabels
}

// Set implements the flag.Set interface.
func (labels *Labels) Set(value string) error {
	m := map[string]string{}
	if value != "" {
		splited := strings.Split(value, ",")
		for _, pair := range splited {
			sp := strings.Split(pair, "=")
			m[sp[0]] = sp[1]
		}
	}
	(*labels).LabelsMap = m
	(*labels).LabelsString = value
	return nil
}

// Config defines configuration parameters for the Operator.
type Config struct {
	Host                         string
	KubeletObject                string
	TLSInsecure                  bool
	TLSConfig                    rest.TLSClientConfig
	ConfigReloaderImage          string
	PrometheusConfigReloader     string
	AlertmanagerDefaultBaseImage string
	PrometheusDefaultBaseImage   string
	ThanosDefaultBaseImage       string
	Namespaces                   []string
	Labels                       Labels
	CrdGroup                     string
	CrdKinds                     monitoringv1.CrdKinds
	EnableValidation             bool
	LocalHost                    string
	LogLevel                     string
	LogFormat                    string
	ManageCRDs                   bool
}

type BasicAuthCredentials struct {
	username string
	password string
}

// New creates a new controller.
func New(conf Config, logger log.Logger) (*Operator, error) {
	cfg, err := k8sutil.NewClusterConfig(conf.Host, conf.TLSInsecure, &conf.TLSConfig)
	if err != nil {
		return nil, errors.Wrap(err, "instantiating cluster config failed")
	}

	client, err := kubernetes.NewForConfig(cfg)
	if err != nil {
		return nil, errors.Wrap(err, "instantiating kubernetes client failed")
	}

	crdclient, err := apiextensionsclient.NewForConfig(cfg)
	if err != nil {
		return nil, errors.Wrap(err, "instantiating apiextensions client failed")
	}

	mclient, err := monitoringclient.NewForConfig(cfg)
	if err != nil {
		return nil, errors.Wrap(err, "instantiating monitoring client failed")
	}

	kubeletObjectName := ""
	kubeletObjectNamespace := ""
	kubeletSyncEnabled := false

	if conf.KubeletObject != "" {
		parts := strings.Split(conf.KubeletObject, "/")
		if len(parts) != 2 {
			return nil, fmt.Errorf("malformatted kubelet object string, must be in format \"namespace/name\"")
		}
		kubeletObjectNamespace = parts[0]
		kubeletObjectName = parts[1]
		kubeletSyncEnabled = true
	}

	c := &Operator{
		kclient:                client,
		mclient:                mclient,
		crdclient:              crdclient,
		logger:                 logger,
		queue:                  workqueue.NewNamedRateLimitingQueue(workqueue.DefaultControllerRateLimiter(), "prometheus"),
		host:                   cfg.Host,
		kubeletObjectName:      kubeletObjectName,
		kubeletObjectNamespace: kubeletObjectNamespace,
		kubeletSyncEnabled:     kubeletSyncEnabled,
		config:                 conf,
		configGenerator:        NewConfigGenerator(logger),
	}

	c.promInf = cache.NewSharedIndexInformer(
		listwatch.MultiNamespaceListerWatcher(c.config.Namespaces, func(namespace string) cache.ListerWatcher {
			return &cache.ListWatch{
				ListFunc: func(options metav1.ListOptions) (runtime.Object, error) {
					return mclient.MonitoringV1().Prometheuses(namespace).List(options)
				},
				WatchFunc: mclient.MonitoringV1().Prometheuses(namespace).Watch,
			}
		}),
		&monitoringv1.Prometheus{}, resyncPeriod, cache.Indexers{cache.NamespaceIndex: cache.MetaNamespaceIndexFunc},
	)

	c.smonInf = cache.NewSharedIndexInformer(
		listwatch.MultiNamespaceListerWatcher(c.config.Namespaces, func(namespace string) cache.ListerWatcher {
			return &cache.ListWatch{
				ListFunc: func(options metav1.ListOptions) (runtime.Object, error) {
					return mclient.MonitoringV1().ServiceMonitors(namespace).List(options)
				},
				WatchFunc: mclient.MonitoringV1().ServiceMonitors(namespace).Watch,
			}
		}),
		&monitoringv1.ServiceMonitor{}, resyncPeriod, cache.Indexers{cache.NamespaceIndex: cache.MetaNamespaceIndexFunc},
	)

	c.ruleInf = cache.NewSharedIndexInformer(
		listwatch.MultiNamespaceListerWatcher(c.config.Namespaces, func(namespace string) cache.ListerWatcher {
			return &cache.ListWatch{
				ListFunc: func(options metav1.ListOptions) (runtime.Object, error) {
					return mclient.MonitoringV1().PrometheusRules(namespace).List(options)
				},
				WatchFunc: mclient.MonitoringV1().PrometheusRules(namespace).Watch,
			}
		}),
		&monitoringv1.PrometheusRule{}, resyncPeriod, cache.Indexers{cache.NamespaceIndex: cache.MetaNamespaceIndexFunc},
	)

	c.cmapInf = cache.NewSharedIndexInformer(
		listwatch.MultiNamespaceListerWatcher(c.config.Namespaces, func(namespace string) cache.ListerWatcher {
			return cache.NewListWatchFromClient(c.kclient.Core().RESTClient(), "configmaps", namespace, fields.Everything())
		}),
		&v1.ConfigMap{}, resyncPeriod, cache.Indexers{cache.NamespaceIndex: cache.MetaNamespaceIndexFunc},
	)

	c.secrInf = cache.NewSharedIndexInformer(
		listwatch.MultiNamespaceListerWatcher(c.config.Namespaces, func(namespace string) cache.ListerWatcher {
			return cache.NewListWatchFromClient(c.kclient.Core().RESTClient(), "secrets", namespace, fields.Everything())
		}),
		&v1.Secret{}, resyncPeriod, cache.Indexers{cache.NamespaceIndex: cache.MetaNamespaceIndexFunc},
	)

	c.ssetInf = cache.NewSharedIndexInformer(
		listwatch.MultiNamespaceListerWatcher(c.config.Namespaces, func(namespace string) cache.ListerWatcher {
			return cache.NewListWatchFromClient(c.kclient.AppsV1beta2().RESTClient(), "statefulsets", namespace, fields.Everything())
		}),
		&appsv1.StatefulSet{}, resyncPeriod, cache.Indexers{cache.NamespaceIndex: cache.MetaNamespaceIndexFunc},
	)

	// nsResyncPeriod is used to control how often the namespace informer
	// should resync. If the unprivileged ListerWatcher is used, then the
	// informer must resync more often because it cannot watch for
	// namespace changes.
	nsResyncPeriod := 15 * time.Second
	// If the only namespace is v1.NamespaceAll, then the client must be
	// privileged and a regular cache.ListWatch will be used. In this case
	// watching works and we do not need to resync so frequently.
	if listwatch.IsAllNamespaces(c.config.Namespaces) {
		nsResyncPeriod = resyncPeriod
	}
	c.nsInf = cache.NewSharedIndexInformer(
		listwatch.NewUnprivilegedNamespaceListWatchFromClient(c.kclient.Core().RESTClient(), c.config.Namespaces, fields.Everything()),
		&v1.Namespace{}, nsResyncPeriod, cache.Indexers{},
	)

	return c, nil
}

// RegisterMetrics registers Prometheus metrics on the given Prometheus
// registerer.
func (c *Operator) RegisterMetrics(r prometheus.Registerer, reconcileErrorsCounter *prometheus.CounterVec, triggerByCounter *prometheus.CounterVec) {
	c.reconcileErrorsCounter = reconcileErrorsCounter
	c.triggerByCounter = triggerByCounter

	c.reconcileErrorsCounter.With(prometheus.Labels{}).Add(0)

	c.nodeAddressLookupErrors = prometheus.NewCounter(prometheus.CounterOpts{
		Name: "prometheus_operator_node_address_lookup_errors_total",
		Help: "Number of times a node IP address could not be determined",
	})

	r.MustRegister(
		c.nodeAddressLookupErrors,
		NewPrometheusCollector(c.promInf.GetStore()),
	)
}

// waitForCacheSync waits for the informers' caches to be synced.
func (c *Operator) waitForCacheSync(stopc <-chan struct{}) error {
	ok := true
	informers := []struct {
		name     string
		informer cache.SharedIndexInformer
	}{
		{"Prometheus", c.promInf},
		{"ServiceMonitor", c.smonInf},
		{"PrometheusRule", c.ruleInf},
		{"ConfigMap", c.cmapInf},
		{"Secret", c.secrInf},
		{"StatefulSet", c.ssetInf},
		{"Namespace", c.nsInf},
	}
	for _, inf := range informers {
		if !cache.WaitForCacheSync(stopc, inf.informer.HasSynced) {
			level.Error(c.logger).Log("msg", fmt.Sprintf("failed to sync %s cache", inf.name))
			ok = false
		} else {
			level.Debug(c.logger).Log("msg", fmt.Sprintf("successfully synced %s cache", inf.name))
		}
	}
	if !ok {
		return errors.New("failed to sync caches")
	}
	level.Info(c.logger).Log("msg", "successfully synced all caches")
	return nil
}

// addHandlers adds the eventhandlers to the informers.
func (c *Operator) addHandlers() {
	c.promInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
		AddFunc:    c.handlePrometheusAdd,
		DeleteFunc: c.handlePrometheusDelete,
		UpdateFunc: c.handlePrometheusUpdate,
	})
	c.smonInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
		AddFunc:    c.handleSmonAdd,
		DeleteFunc: c.handleSmonDelete,
		UpdateFunc: c.handleSmonUpdate,
	})
	c.ruleInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
		AddFunc:    c.handleRuleAdd,
		DeleteFunc: c.handleRuleDelete,
		UpdateFunc: c.handleRuleUpdate,
	})
	c.cmapInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
		AddFunc:    c.handleConfigMapAdd,
		DeleteFunc: c.handleConfigMapDelete,
		UpdateFunc: c.handleConfigMapUpdate,
	})
	c.secrInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
		AddFunc:    c.handleSecretAdd,
		DeleteFunc: c.handleSecretDelete,
		UpdateFunc: c.handleSecretUpdate,
	})
	c.ssetInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
		AddFunc:    c.handleStatefulSetAdd,
		DeleteFunc: c.handleStatefulSetDelete,
		UpdateFunc: c.handleStatefulSetUpdate,
	})
}

// Run the controller.
func (c *Operator) Run(stopc <-chan struct{}) error {
	defer c.queue.ShutDown()

	errChan := make(chan error)
	go func() {
		v, err := c.kclient.Discovery().ServerVersion()
		if err != nil {
			errChan <- errors.Wrap(err, "communicating with server failed")
			return
		}
		level.Info(c.logger).Log("msg", "connection established", "cluster-version", v)

		if c.config.ManageCRDs {
			if err := c.createCRDs(); err != nil {
				errChan <- errors.Wrap(err, "creating CRDs failed")
				return
			}
		}
		errChan <- nil
	}()

	select {
	case err := <-errChan:
		if err != nil {
			return err
		}
		level.Info(c.logger).Log("msg", "CRD API endpoints ready")
	case <-stopc:
		return nil
	}

	go c.worker()

	go c.promInf.Run(stopc)
	go c.smonInf.Run(stopc)
	go c.ruleInf.Run(stopc)
	go c.cmapInf.Run(stopc)
	go c.secrInf.Run(stopc)
	go c.ssetInf.Run(stopc)
	go c.nsInf.Run(stopc)
	if err := c.waitForCacheSync(stopc); err != nil {
		return err
	}
	c.addHandlers()

	if c.kubeletSyncEnabled {
		go c.reconcileNodeEndpoints(stopc)
	}

	<-stopc
	return nil
}

func (c *Operator) keyFunc(obj interface{}) (string, bool) {
	k, err := cache.DeletionHandlingMetaNamespaceKeyFunc(obj)
	if err != nil {
		level.Error(c.logger).Log("msg", "creating key failed", "err", err)
		return k, false
	}
	return k, true
}

func (c *Operator) handlePrometheusAdd(obj interface{}) {
	key, ok := c.keyFunc(obj)
	if !ok {
		return
	}

	level.Debug(c.logger).Log("msg", "Prometheus added", "key", key)
	c.triggerByCounter.WithLabelValues(monitoringv1.PrometheusesKind, "add").Inc()
	c.enqueue(key)
}

func (c *Operator) handlePrometheusDelete(obj interface{}) {
	key, ok := c.keyFunc(obj)
	if !ok {
		return
	}

	level.Debug(c.logger).Log("msg", "Prometheus deleted", "key", key)
	c.triggerByCounter.WithLabelValues(monitoringv1.PrometheusesKind, "delete").Inc()
	c.enqueue(key)
}

func (c *Operator) handlePrometheusUpdate(old, cur interface{}) {
	if old.(*monitoringv1.Prometheus).ResourceVersion == cur.(*monitoringv1.Prometheus).ResourceVersion {
		return
	}

	key, ok := c.keyFunc(cur)
	if !ok {
		return
	}

	level.Debug(c.logger).Log("msg", "Prometheus updated", "key", key)
	c.triggerByCounter.WithLabelValues(monitoringv1.PrometheusesKind, "update").Inc()
	c.enqueue(key)
}

func (c *Operator) reconcileNodeEndpoints(stopc <-chan struct{}) {
	ticker := time.NewTicker(3 * time.Minute)
	defer ticker.Stop()
	for {
		select {
		case <-stopc:
			return
		case <-ticker.C:
			err := c.syncNodeEndpoints()
			if err != nil {
				level.Error(c.logger).Log("msg", "syncing nodes into Endpoints object failed", "err", err)
			}
		}
	}
}

// nodeAddresses returns the provided node's address, based on the priority:
// 1. NodeInternalIP
// 2. NodeExternalIP
//
// Copied from github.com/prometheus/prometheus/discovery/kubernetes/node.go
func nodeAddress(node v1.Node) (string, map[v1.NodeAddressType][]string, error) {
	m := map[v1.NodeAddressType][]string{}
	for _, a := range node.Status.Addresses {
		m[a.Type] = append(m[a.Type], a.Address)
	}

	if addresses, ok := m[v1.NodeInternalIP]; ok {
		return addresses[0], m, nil
	}
	if addresses, ok := m[v1.NodeExternalIP]; ok {
		return addresses[0], m, nil
	}
	return "", m, fmt.Errorf("host address unknown")
}

func getNodeAddresses(nodes *v1.NodeList) ([]v1.EndpointAddress, []error) {
	addresses := make([]v1.EndpointAddress, 0)
	errs := make([]error, 0)

	for _, n := range nodes.Items {
		address, _, err := nodeAddress(n)
		if err != nil {
			errs = append(errs, errors.Wrapf(err, "failed to determine hostname for node (%s)", n.Name))
			continue
		}
		addresses = append(addresses, v1.EndpointAddress{
			IP: address,
			TargetRef: &v1.ObjectReference{
				Kind:       "Node",
				Name:       n.Name,
				UID:        n.UID,
				APIVersion: n.APIVersion,
			},
		})
	}

	return addresses, errs
}

func (c *Operator) syncNodeEndpoints() error {
	eps := &v1.Endpoints{
		ObjectMeta: metav1.ObjectMeta{
			Name: c.kubeletObjectName,
			Labels: c.config.Labels.Merge(map[string]string{
				"k8s-app": "kubelet",
			}),
		},
		Subsets: []v1.EndpointSubset{
			{
				Ports: []v1.EndpointPort{
					{
						Name: "https-metrics",
						Port: 10250,
					},
					{
						Name: "http-metrics",
						Port: 10255,
					},
					{
						Name: "cadvisor",
						Port: 4194,
					},
				},
			},
		},
	}

	nodes, err := c.kclient.CoreV1().Nodes().List(metav1.ListOptions{})
	if err != nil {
		return errors.Wrap(err, "listing nodes failed")
	}

	addresses, errs := getNodeAddresses(nodes)
	if len(errs) > 0 {
		for _, err := range errs {
			level.Warn(c.logger).Log("err", err)
		}
		c.nodeAddressLookupErrors.Add(float64(len(errs)))
	}
	eps.Subsets[0].Addresses = addresses

	svc := &v1.Service{
		ObjectMeta: metav1.ObjectMeta{
			Name: c.kubeletObjectName,
			Labels: c.config.Labels.Merge(map[string]string{
				"k8s-app": "kubelet",
			}),
		},
		Spec: v1.ServiceSpec{
			Type:      v1.ServiceTypeClusterIP,
			ClusterIP: "None",
			Ports: []v1.ServicePort{
				{
					Name: "https-metrics",
					Port: 10250,
				},
			},
		},
	}

	err = k8sutil.CreateOrUpdateService(c.kclient.CoreV1().Services(c.kubeletObjectNamespace), svc)
	if err != nil {
		return errors.Wrap(err, "synchronizing kubelet service object failed")
	}

	err = k8sutil.CreateOrUpdateEndpoints(c.kclient.CoreV1().Endpoints(c.kubeletObjectNamespace), eps)
	if err != nil {
		return errors.Wrap(err, "synchronizing kubelet endpoints object failed")
	}

	return nil
}

// TODO: Don't enque just for the namespace
func (c *Operator) handleSmonAdd(obj interface{}) {
	o, ok := c.getObject(obj)
	if ok {
		level.Debug(c.logger).Log("msg", "ServiceMonitor added")
		c.triggerByCounter.WithLabelValues(monitoringv1.ServiceMonitorsKind, "add").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

// TODO: Don't enque just for the namespace
func (c *Operator) handleSmonUpdate(old, cur interface{}) {
	if old.(*monitoringv1.ServiceMonitor).ResourceVersion == cur.(*monitoringv1.ServiceMonitor).ResourceVersion {
		return
	}

	o, ok := c.getObject(cur)
	if ok {
		level.Debug(c.logger).Log("msg", "ServiceMonitor updated")
		c.triggerByCounter.WithLabelValues(monitoringv1.ServiceMonitorsKind, "update").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

// TODO: Don't enque just for the namespace
func (c *Operator) handleSmonDelete(obj interface{}) {
	o, ok := c.getObject(obj)
	if ok {
		level.Debug(c.logger).Log("msg", "ServiceMonitor delete")
		c.triggerByCounter.WithLabelValues(monitoringv1.ServiceMonitorsKind, "delete").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

// TODO: Don't enque just for the namespace
func (c *Operator) handleRuleAdd(obj interface{}) {
	o, ok := c.getObject(obj)
	if ok {
		level.Debug(c.logger).Log("msg", "PrometheusRule added")
		c.triggerByCounter.WithLabelValues(monitoringv1.PrometheusRuleKind, "add").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

// TODO: Don't enque just for the namespace
func (c *Operator) handleRuleUpdate(old, cur interface{}) {
	if old.(*monitoringv1.PrometheusRule).ResourceVersion == cur.(*monitoringv1.PrometheusRule).ResourceVersion {
		return
	}

	o, ok := c.getObject(cur)
	if ok {
		level.Debug(c.logger).Log("msg", "PrometheusRule updated")
		c.triggerByCounter.WithLabelValues(monitoringv1.PrometheusRuleKind, "update").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

// TODO: Don't enque just for the namespace
func (c *Operator) handleRuleDelete(obj interface{}) {
	o, ok := c.getObject(obj)
	if ok {
		level.Debug(c.logger).Log("msg", "PrometheusRule deleted")
		c.triggerByCounter.WithLabelValues(monitoringv1.PrometheusRuleKind, "delete").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

// TODO: Do we need to enque secrets just for the namespace or in general?
func (c *Operator) handleSecretDelete(obj interface{}) {
	o, ok := c.getObject(obj)
	if ok {
		level.Debug(c.logger).Log("msg", "Secret deleted")
		c.triggerByCounter.WithLabelValues("Secret", "delete").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

func (c *Operator) handleSecretUpdate(old, cur interface{}) {
	if old.(*v1.Secret).ResourceVersion == cur.(*v1.Secret).ResourceVersion {
		return
	}

	o, ok := c.getObject(cur)
	if ok {
		level.Debug(c.logger).Log("msg", "Secret updated")
		c.triggerByCounter.WithLabelValues("Secret", "update").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

func (c *Operator) handleSecretAdd(obj interface{}) {
	o, ok := c.getObject(obj)
	if ok {
		level.Debug(c.logger).Log("msg", "Secret added")
		c.triggerByCounter.WithLabelValues("Secret", "add").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

// TODO: Do we need to enque configmaps just for the namespace or in general?
func (c *Operator) handleConfigMapAdd(obj interface{}) {
	o, ok := c.getObject(obj)
	if ok {
		level.Debug(c.logger).Log("msg", "ConfigMap added")
		c.triggerByCounter.WithLabelValues("ConfigMap", "add").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

func (c *Operator) handleConfigMapDelete(obj interface{}) {
	o, ok := c.getObject(obj)
	if ok {
		level.Debug(c.logger).Log("msg", "ConfigMap deleted")
		c.triggerByCounter.WithLabelValues("ConfigMap", "delete").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

func (c *Operator) handleConfigMapUpdate(old, cur interface{}) {
	if old.(*v1.ConfigMap).ResourceVersion == cur.(*v1.ConfigMap).ResourceVersion {
		return
	}

	o, ok := c.getObject(cur)
	if ok {
		level.Debug(c.logger).Log("msg", "ConfigMap updated")
		c.triggerByCounter.WithLabelValues("ConfigMap", "update").Inc()

		c.enqueueForNamespace(o.GetNamespace())
	}
}

func (c *Operator) getObject(obj interface{}) (metav1.Object, bool) {
	ts, ok := obj.(cache.DeletedFinalStateUnknown)
	if ok {
		obj = ts.Obj
	}

	o, err := meta.Accessor(obj)
	if err != nil {
		level.Error(c.logger).Log("msg", "get object failed", "err", err)
		return nil, false
	}
	return o, true
}

// enqueue adds a key to the queue. If obj is a key already it gets added
// directly. Otherwise, the key is extracted via keyFunc.
func (c *Operator) enqueue(obj interface{}) {
	if obj == nil {
		return
	}

	key, ok := obj.(string)
	if !ok {
		key, ok = c.keyFunc(obj)
		if !ok {
			return
		}
	}

	c.queue.Add(key)
}

// enqueueForNamespace enqueues all Prometheus object keys that belong to the
// given namespace or select objects in the given namespace.
func (c *Operator) enqueueForNamespace(nsName string) {
	nsObject, exists, err := c.nsInf.GetStore().GetByKey(nsName)
	if err != nil {
		level.Error(c.logger).Log(
			"msg", "get namespace to enqueue Prometheus instances failed",
			"err", err,
		)
		return
	}
	if !exists {
		level.Error(c.logger).Log(
			"msg", fmt.Sprintf("get namespace to enqueue Prometheus instances failed: namespace %q does not exist", nsName),
			"err", err,
		)
		return
	}
	ns := nsObject.(*v1.Namespace)

	err = cache.ListAll(c.promInf.GetStore(), labels.Everything(), func(obj interface{}) {
		// Check for Prometheus instances in the NS.
		p := obj.(*monitoringv1.Prometheus)
		if p.Namespace == nsName {
			c.enqueue(p)
			return
		}

		// Check for Prometheus instances selecting ServiceMonitors in
		// the NS.
		smNSSelector, err := metav1.LabelSelectorAsSelector(p.Spec.ServiceMonitorNamespaceSelector)
		if err != nil {
			level.Error(c.logger).Log(
				"msg", fmt.Sprintf("failed to convert ServiceMonitorNamespaceSelector of %q to selector", p.Name),
				"err", err,
			)
			return
		}

		if smNSSelector.Matches(labels.Set(ns.Labels)) {
			c.enqueue(p)
			return
		}

		// Check for Prometheus instances selecting PrometheusRules in
		// the NS.
		ruleNSSelector, err := metav1.LabelSelectorAsSelector(p.Spec.RuleNamespaceSelector)
		if err != nil {
			level.Error(c.logger).Log(
				"msg", fmt.Sprintf("failed to convert RuleNamespaceSelector of %q to selector", p.Name),
				"err", err,
			)
			return
		}

		if ruleNSSelector.Matches(labels.Set(ns.Labels)) {
			c.enqueue(p)
			return
		}
	})
	if err != nil {
		level.Error(c.logger).Log(
			"msg", "listing all Prometheus instances from cache failed",
			"err", err,
		)
	}
}

// worker runs a worker thread that just dequeues items, processes them, and
// marks them done. It enforces that the syncHandler is never invoked
// concurrently with the same key.
func (c *Operator) worker() {
	for c.processNextWorkItem() {
	}
}

func (c *Operator) processNextWorkItem() bool {
	key, quit := c.queue.Get()
	if quit {
		return false
	}
	defer c.queue.Done(key)

	err := c.sync(key.(string))
	if err == nil {
		c.queue.Forget(key)
		return true
	}

	c.reconcileErrorsCounter.With(prometheus.Labels{}).Inc()
	utilruntime.HandleError(errors.Wrap(err, fmt.Sprintf("Sync %q failed", key)))
	c.queue.AddRateLimited(key)

	return true
}

func (c *Operator) prometheusForStatefulSet(sset interface{}) *monitoringv1.Prometheus {
	key, ok := c.keyFunc(sset)
	if !ok {
		return nil
	}

	promKey := statefulSetKeyToPrometheusKey(key)
	p, exists, err := c.promInf.GetStore().GetByKey(promKey)
	if err != nil {
		level.Error(c.logger).Log("msg", "Prometheus lookup failed", "err", err)
		return nil
	}
	if !exists {
		return nil
	}
	return p.(*monitoringv1.Prometheus)
}

func prometheusNameFromStatefulSetName(name string) string {
	return strings.TrimPrefix(name, "prometheus-")
}

func statefulSetNameFromPrometheusName(name string) string {
	return "prometheus-" + name
}

func statefulSetKeyToPrometheusKey(key string) string {
	keyParts := strings.Split(key, "/")
	return keyParts[0] + "/" + strings.TrimPrefix(keyParts[1], "prometheus-")
}

func prometheusKeyToStatefulSetKey(key string) string {
	keyParts := strings.Split(key, "/")
	return keyParts[0] + "/prometheus-" + keyParts[1]
}

func (c *Operator) handleStatefulSetDelete(obj interface{}) {
	if ps := c.prometheusForStatefulSet(obj); ps != nil {
		level.Debug(c.logger).Log("msg", "StatefulSet delete")
		c.triggerByCounter.WithLabelValues("StatefulSet", "delete").Inc()

		c.enqueue(ps)
	}
}

func (c *Operator) handleStatefulSetAdd(obj interface{}) {
	if ps := c.prometheusForStatefulSet(obj); ps != nil {
		level.Debug(c.logger).Log("msg", "StatefulSet added")
		c.triggerByCounter.WithLabelValues("StatefulSet", "add").Inc()

		c.enqueue(ps)
	}
}

func (c *Operator) handleStatefulSetUpdate(oldo, curo interface{}) {
	old := oldo.(*appsv1.StatefulSet)
	cur := curo.(*appsv1.StatefulSet)

	level.Debug(c.logger).Log("msg", "update handler", "old", old.ResourceVersion, "cur", cur.ResourceVersion)

	// Periodic resync may resend the StatefulSet without changes
	// in-between. Also breaks loops created by updating the resource
	// ourselves.
	if old.ResourceVersion == cur.ResourceVersion {
		return
	}

	if ps := c.prometheusForStatefulSet(cur); ps != nil {
		level.Debug(c.logger).Log("msg", "StatefulSet updated")
		c.triggerByCounter.WithLabelValues("StatefulSet", "update").Inc()

		c.enqueue(ps)
	}
}

func (c *Operator) sync(key string) error {
	obj, exists, err := c.promInf.GetIndexer().GetByKey(key)
	if err != nil {
		return err
	}
	if !exists {
		// Dependent resources are cleaned up by K8s via OwnerReferences
		return nil
	}

	p := obj.(*monitoringv1.Prometheus)
	p = p.DeepCopy()
	p.APIVersion = monitoringv1.SchemeGroupVersion.String()
	p.Kind = monitoringv1.PrometheusesKind

	if p.Spec.Paused {
		return nil
	}

	level.Info(c.logger).Log("msg", "sync prometheus", "key", key)

	ruleConfigMapNames, err := c.createOrUpdateRuleConfigMaps(p)
	if err != nil {
		return err
	}

	// If no service monitor selectors are configured, the user wants to
	// manage configuration themselves.
	if p.Spec.ServiceMonitorSelector != nil {
		// We just always regenerate the configuration to be safe.
		if err := c.createOrUpdateConfigurationSecret(p, ruleConfigMapNames); err != nil {
			return errors.Wrap(err, "creating config failed")
		}
	}

	// Create empty Secret if it doesn't exist. See comment above.
	s, err := makeEmptyConfigurationSecret(p, c.config)
	if err != nil {
		return errors.Wrap(err, "generating empty config secret failed")
	}
	sClient := c.kclient.CoreV1().Secrets(p.Namespace)
	_, err = sClient.Get(s.Name, metav1.GetOptions{})
	if apierrors.IsNotFound(err) {
		if _, err := c.kclient.Core().Secrets(p.Namespace).Create(s); err != nil && !apierrors.IsAlreadyExists(err) {
			return errors.Wrap(err, "creating empty config file failed")
		}
	}
	if !apierrors.IsNotFound(err) && err != nil {
		return err
	}

	// Create governing service if it doesn't exist.
	svcClient := c.kclient.Core().Services(p.Namespace)
	if err := k8sutil.CreateOrUpdateService(svcClient, makeStatefulSetService(p, c.config)); err != nil {
		return errors.Wrap(err, "synchronizing governing service failed")
	}

	ssetClient := c.kclient.AppsV1beta2().StatefulSets(p.Namespace)
	// Ensure we have a StatefulSet running Prometheus deployed.
	obj, exists, err = c.ssetInf.GetIndexer().GetByKey(prometheusKeyToStatefulSetKey(key))
	if err != nil {
		return errors.Wrap(err, "retrieving statefulset failed")
	}

	newSSetInputHash, err := createSSetInputHash(*p, c.config, ruleConfigMapNames)
	if err != nil {
		return err
	}

	sset, err := makeStatefulSet(*p, &c.config, ruleConfigMapNames, newSSetInputHash)
	if err != nil {
		return errors.Wrap(err, "making statefulset failed")
	}

	if !exists {
		level.Debug(c.logger).Log("msg", "no current Prometheus statefulset found")
		level.Debug(c.logger).Log("msg", "creating Prometheus statefulset")
		if _, err := ssetClient.Create(sset); err != nil {
			return errors.Wrap(err, "creating statefulset failed")
		}
		return nil
	}

	oldSSetInputHash := obj.(*appsv1.StatefulSet).ObjectMeta.Annotations[sSetInputHashName]
	if newSSetInputHash == oldSSetInputHash {
		level.Debug(c.logger).Log("msg", "new statefulset generation inputs match current, skipping any actions")
		return nil
	}

	level.Debug(c.logger).Log("msg", "updating current Prometheus statefulset")

	_, err = ssetClient.Update(sset)
	sErr, ok := err.(*apierrors.StatusError)

	if ok && sErr.ErrStatus.Code == 422 && sErr.ErrStatus.Reason == metav1.StatusReasonInvalid {
		level.Debug(c.logger).Log("msg", "resolving illegal update of Prometheus StatefulSet")
		propagationPolicy := metav1.DeletePropagationForeground
		if err := ssetClient.Delete(sset.GetName(), &metav1.DeleteOptions{PropagationPolicy: &propagationPolicy}); err != nil {
			return errors.Wrap(err, "failed to delete StatefulSet to avoid forbidden action")
		}
		return nil
	}

	if err != nil {
		return errors.Wrap(err, "updating StatefulSet failed")
	}

	return nil
}

func createSSetInputHash(p monitoringv1.Prometheus, c Config, ruleConfigMapNames []string) (string, error) {
	hash, err := hashstructure.Hash(struct {
		P monitoringv1.Prometheus
		C Config
		R []string `hash:"set"`
	}{p, c, ruleConfigMapNames},
		nil,
	)
	if err != nil {
		return "", errors.Wrap(
			err,
			"failed to calculate combined hash of Prometheus CRD, config and"+
				" rule ConfigMap names",
		)
	}

	return fmt.Sprintf("%d", hash), nil
}

func ListOptions(name string) metav1.ListOptions {
	return metav1.ListOptions{
		LabelSelector: fields.SelectorFromSet(fields.Set(map[string]string{
			"app":        "prometheus",
			"prometheus": name,
		})).String(),
	}
}

// PrometheusStatus evaluates the current status of a Prometheus deployment with
// respect to its specified resource object. It return the status and a list of
// pods that are not updated.
func PrometheusStatus(kclient kubernetes.Interface, p *monitoringv1.Prometheus) (*monitoringv1.PrometheusStatus, []v1.Pod, error) {
	res := &monitoringv1.PrometheusStatus{Paused: p.Spec.Paused}

	pods, err := kclient.Core().Pods(p.Namespace).List(ListOptions(p.Name))
	if err != nil {
		return nil, nil, errors.Wrap(err, "retrieving pods of failed")
	}
	sset, err := kclient.AppsV1beta2().StatefulSets(p.Namespace).Get(statefulSetNameFromPrometheusName(p.Name), metav1.GetOptions{})
	if err != nil {
		return nil, nil, errors.Wrap(err, "retrieving stateful set failed")
	}

	res.Replicas = int32(len(pods.Items))

	var oldPods []v1.Pod
	for _, pod := range pods.Items {
		ready, err := k8sutil.PodRunningAndReady(pod)
		if err != nil {
			return nil, nil, errors.Wrap(err, "cannot determine pod ready state")
		}
		if ready {
			res.AvailableReplicas++
			// TODO(fabxc): detect other fields of the pod template
			// that are mutable.
			if needsUpdate(&pod, sset.Spec.Template) {
				oldPods = append(oldPods, pod)
			} else {
				res.UpdatedReplicas++
			}
			continue
		}
		res.UnavailableReplicas++
	}

	return res, oldPods, nil
}

// needsUpdate checks whether the given pod conforms with the pod template spec
// for various attributes that are influenced by the Prometheus CRD settings.
func needsUpdate(pod *v1.Pod, tmpl v1.PodTemplateSpec) bool {
	c1 := pod.Spec.Containers[0]
	c2 := tmpl.Spec.Containers[0]

	if c1.Image != c2.Image {
		return true
	}
	if !reflect.DeepEqual(c1.Args, c2.Args) {
		return true
	}

	return false
}

func loadAdditionalScrapeConfigsSecret(additionalScrapeConfigs *v1.SecretKeySelector, s *v1.SecretList) ([]byte, error) {
	if additionalScrapeConfigs != nil {
		for _, secret := range s.Items {
			if secret.Name == additionalScrapeConfigs.Name {
				if c, ok := secret.Data[additionalScrapeConfigs.Key]; ok {
					return c, nil
				}

				return nil, fmt.Errorf("key %v could not be found in Secret %v", additionalScrapeConfigs.Key, additionalScrapeConfigs.Name)
			}
		}
		return nil, fmt.Errorf("secret %v could not be found", additionalScrapeConfigs.Name)
	}
	return nil, nil
}

func extractCredKey(secret *v1.Secret, sel v1.SecretKeySelector, cred string) (string, error) {
	if s, ok := secret.Data[sel.Key]; ok {
		return string(s), nil
	}
	return "", fmt.Errorf("secret %s key %q in secret %q not found", cred, sel.Key, sel.Name)
}

func getCredFromSecret(c corev1client.SecretInterface, sel v1.SecretKeySelector, cred string, cacheKey string, cache map[string]*v1.Secret) (_ string, err error) {
	var s *v1.Secret
	var ok bool

	if s, ok = cache[cacheKey]; !ok {
		if s, err = c.Get(sel.Name, metav1.GetOptions{}); err != nil {
			return "", fmt.Errorf("unable to fetch %s secret %q: %s", cred, sel.Name, err)
		}
		cache[cacheKey] = s
	}
	return extractCredKey(s, sel, cred)
}

func loadBasicAuthSecretFromAPI(basicAuth *monitoringv1.BasicAuth, c corev1client.CoreV1Interface, ns string, cache map[string]*v1.Secret) (BasicAuthCredentials, error) {
	var username string
	var password string
	var err error

	sClient := c.Secrets(ns)

	if username, err = getCredFromSecret(sClient, basicAuth.Username, "username", ns+"/"+basicAuth.Username.Name, cache); err != nil {
		return BasicAuthCredentials{}, err
	}

	if password, err = getCredFromSecret(sClient, basicAuth.Password, "password", ns+"/"+basicAuth.Password.Name, cache); err != nil {
		return BasicAuthCredentials{}, err
	}

	return BasicAuthCredentials{username: username, password: password}, nil
}

func loadBasicAuthSecret(basicAuth *monitoringv1.BasicAuth, s *v1.SecretList) (BasicAuthCredentials, error) {
	var username string
	var password string
	var err error

	for _, secret := range s.Items {

		if secret.Name == basicAuth.Username.Name {
			if username, err = extractCredKey(&secret, basicAuth.Username, "username"); err != nil {
				return BasicAuthCredentials{}, err
			}
		}

		if secret.Name == basicAuth.Password.Name {
			if password, err = extractCredKey(&secret, basicAuth.Password, "password"); err != nil {
				return BasicAuthCredentials{}, err
			}

		}
		if username != "" && password != "" {
			break
		}
	}

	if username == "" && password == "" {
		return BasicAuthCredentials{}, fmt.Errorf("basic auth username and password secret not found")
	}

	return BasicAuthCredentials{username: username, password: password}, nil

}

func gzipConfig(buf *bytes.Buffer, conf []byte) error {
	w := gzip.NewWriter(buf)
	defer w.Close()
	if _, err := w.Write(conf); err != nil {
		return err
	}
	return nil
}

func (c *Operator) loadBasicAuthSecrets(
	mons map[string]*monitoringv1.ServiceMonitor,
	remoteReads []monitoringv1.RemoteReadSpec,
	remoteWrites []monitoringv1.RemoteWriteSpec,
	apiserverConfig *monitoringv1.APIServerConfig,
	SecretsInPromNS *v1.SecretList,
) (map[string]BasicAuthCredentials, error) {

	secrets := map[string]BasicAuthCredentials{}
	nsSecretCache := make(map[string]*v1.Secret)
	for _, mon := range mons {
		for i, ep := range mon.Spec.Endpoints {
			if ep.BasicAuth != nil {
				credentials, err := loadBasicAuthSecretFromAPI(ep.BasicAuth, c.kclient.CoreV1(), mon.Namespace, nsSecretCache)
				if err != nil {
					return nil, fmt.Errorf("could not generate basicAuth for servicemonitor %s. %s", mon.Name, err)
				}
				secrets[fmt.Sprintf("serviceMonitor/%s/%s/%d", mon.Namespace, mon.Name, i)] = credentials
			}
		}
	}

	for i, remote := range remoteReads {
		if remote.BasicAuth != nil {
			credentials, err := loadBasicAuthSecret(remote.BasicAuth, SecretsInPromNS)
			if err != nil {
				return nil, fmt.Errorf("could not generate basicAuth for remote_read config %d. %s", i, err)
			}
			secrets[fmt.Sprintf("remoteRead/%d", i)] = credentials
		}
	}

	for i, remote := range remoteWrites {
		if remote.BasicAuth != nil {
			credentials, err := loadBasicAuthSecret(remote.BasicAuth, SecretsInPromNS)
			if err != nil {
				return nil, fmt.Errorf("could not generate basicAuth for remote_write config %d. %s", i, err)
			}
			secrets[fmt.Sprintf("remoteWrite/%d", i)] = credentials
		}
	}

	// load apiserver basic auth secret
	if apiserverConfig != nil && apiserverConfig.BasicAuth != nil {
		credentials, err := loadBasicAuthSecret(apiserverConfig.BasicAuth, SecretsInPromNS)
		if err != nil {
			return nil, fmt.Errorf("could not generate basicAuth for apiserver config. %s", err)
		}
		secrets["apiserver"] = credentials
	}

	return secrets, nil

}

func (c *Operator) createOrUpdateConfigurationSecret(p *monitoringv1.Prometheus, ruleConfigMapNames []string) error {
	smons, err := c.selectServiceMonitors(p)
	if err != nil {
		return errors.Wrap(err, "selecting ServiceMonitors failed")
	}

	sClient := c.kclient.CoreV1().Secrets(p.Namespace)
	SecretsInPromNS, err := sClient.List(metav1.ListOptions{})
	if err != nil {
		return err
	}

	basicAuthSecrets, err := c.loadBasicAuthSecrets(smons, p.Spec.RemoteRead, p.Spec.RemoteWrite, p.Spec.APIServerConfig, SecretsInPromNS)
	if err != nil {
		return err
	}

	additionalScrapeConfigs, err := loadAdditionalScrapeConfigsSecret(p.Spec.AdditionalScrapeConfigs, SecretsInPromNS)
	if err != nil {
		return errors.Wrap(err, "loading additional scrape configs from Secret failed")
	}
	additionalAlertRelabelConfigs, err := loadAdditionalScrapeConfigsSecret(p.Spec.AdditionalAlertRelabelConfigs, SecretsInPromNS)
	if err != nil {
		return errors.Wrap(err, "loading additional alert relabel configs from Secret failed")
	}
	additionalAlertManagerConfigs, err := loadAdditionalScrapeConfigsSecret(p.Spec.AdditionalAlertManagerConfigs, SecretsInPromNS)
	if err != nil {
		return errors.Wrap(err, "loading additional alert manager configs from Secret failed")
	}

	// Update secret based on the most recent configuration.
	conf, err := c.configGenerator.generateConfig(
		p,
		smons,
		basicAuthSecrets,
		additionalScrapeConfigs,
		additionalAlertRelabelConfigs,
		additionalAlertManagerConfigs,
		ruleConfigMapNames,
	)
	if err != nil {
		return errors.Wrap(err, "generating config failed")
	}

	s := makeConfigSecret(p, c.config)
	s.ObjectMeta.Annotations = map[string]string{
		"generated": "true",
	}

	// Compress config to avoid 1mb secret limit for a while
	var buf bytes.Buffer
	if err = gzipConfig(&buf, conf); err != nil {
		return errors.Wrap(err, "couldnt gzip config")
	}
	s.Data[configFilename] = buf.Bytes()

	curSecret, err := sClient.Get(s.Name, metav1.GetOptions{})
	if apierrors.IsNotFound(err) {
		level.Debug(c.logger).Log("msg", "creating configuration")
		_, err = sClient.Create(s)
		return err
	}

	var (
		generatedConf             = s.Data[configFilename]
		curConfig, curConfigFound = curSecret.Data[configFilename]
	)
	if curConfigFound {
		if bytes.Equal(curConfig, generatedConf) {
			level.Debug(c.logger).Log("msg", "updating Prometheus configuration secret skipped, no configuration change")
			return nil
		}
		level.Debug(c.logger).Log("msg", "current Prometheus configuration has changed")
	} else {
		level.Debug(c.logger).Log("msg", "no current Prometheus configuration secret found", "currentConfigFound", curConfigFound)
	}

	level.Debug(c.logger).Log("msg", "updating Prometheus configuration secret")
	_, err = sClient.Update(s)
	return err
}

func (c *Operator) selectServiceMonitors(p *monitoringv1.Prometheus) (map[string]*monitoringv1.ServiceMonitor, error) {
	namespaces := []string{}
	// Selectors might overlap. Deduplicate them along the keyFunc.
	res := make(map[string]*monitoringv1.ServiceMonitor)

	servMonSelector, err := metav1.LabelSelectorAsSelector(p.Spec.ServiceMonitorSelector)
	if err != nil {
		return nil, err
	}

	// If 'ServiceMonitorNamespaceSelector' is nil only check own namespace.
	if p.Spec.ServiceMonitorNamespaceSelector == nil {
		namespaces = append(namespaces, p.Namespace)
	} else {
		servMonNSSelector, err := metav1.LabelSelectorAsSelector(p.Spec.ServiceMonitorNamespaceSelector)
		if err != nil {
			return nil, err
		}

		namespaces, err = c.listMatchingNamespaces(servMonNSSelector)
		if err != nil {
			return nil, err
		}
	}

	level.Debug(c.logger).Log("msg", "filtering namespaces to select ServiceMonitors from", "namespaces", strings.Join(namespaces, ","), "namespace", p.Namespace, "prometheus", p.Name)

	for _, ns := range namespaces {
		cache.ListAllByNamespace(c.smonInf.GetIndexer(), ns, servMonSelector, func(obj interface{}) {
			k, ok := c.keyFunc(obj)
			if ok {
				res[k] = obj.(*monitoringv1.ServiceMonitor)
			}
		})
	}

	serviceMonitors := []string{}
	for k := range res {
		serviceMonitors = append(serviceMonitors, k)
	}
	level.Debug(c.logger).Log("msg", "selected ServiceMonitors", "servicemonitors", strings.Join(serviceMonitors, ","), "namespace", p.Namespace, "prometheus", p.Name)

	return res, nil
}

// listMatchingNamespaces lists all the namespaces that match the provided
// selector.
func (c *Operator) listMatchingNamespaces(selector labels.Selector) ([]string, error) {
	var ns []string
	err := cache.ListAll(c.nsInf.GetStore(), selector, func(obj interface{}) {
		ns = append(ns, obj.(*v1.Namespace).Name)
	})
	if err != nil {
		return nil, errors.Wrap(err, "failed to list namespaces")
	}
	return ns, nil
}

func (c *Operator) createCRDs() error {
	crds := []*extensionsobj.CustomResourceDefinition{
		k8sutil.NewCustomResourceDefinition(c.config.CrdKinds.Prometheus, c.config.CrdGroup, c.config.Labels.LabelsMap, c.config.EnableValidation),
		k8sutil.NewCustomResourceDefinition(c.config.CrdKinds.ServiceMonitor, c.config.CrdGroup, c.config.Labels.LabelsMap, c.config.EnableValidation),
		k8sutil.NewCustomResourceDefinition(c.config.CrdKinds.PrometheusRule, c.config.CrdGroup, c.config.Labels.LabelsMap, c.config.EnableValidation),
	}

	crdClient := c.crdclient.ApiextensionsV1beta1().CustomResourceDefinitions()

	for _, crd := range crds {
		oldCRD, err := crdClient.Get(crd.Name, metav1.GetOptions{})
		if err != nil && !apierrors.IsNotFound(err) {
			return errors.Wrapf(err, "getting CRD: %s", crd.Spec.Names.Kind)
		}
		if apierrors.IsNotFound(err) {
			if _, err := crdClient.Create(crd); err != nil {
				return errors.Wrapf(err, "creating CRD: %s", crd.Spec.Names.Kind)
			}
			level.Info(c.logger).Log("msg", "CRD created", "crd", crd.Spec.Names.Kind)
		}
		if err == nil {
			crd.ResourceVersion = oldCRD.ResourceVersion
			if _, err := crdClient.Update(crd); err != nil {
				return errors.Wrapf(err, "creating CRD: %s", crd.Spec.Names.Kind)
			}
			level.Info(c.logger).Log("msg", "CRD updated", "crd", crd.Spec.Names.Kind)
		}
	}

	crdListFuncs := []struct {
		name     string
		listFunc func(opts metav1.ListOptions) (runtime.Object, error)
	}{
		{
			monitoringv1.PrometheusesKind,
			listwatch.MultiNamespaceListerWatcher(c.config.Namespaces, func(namespace string) cache.ListerWatcher {
				return &cache.ListWatch{
					ListFunc: func(options metav1.ListOptions) (runtime.Object, error) {
						return c.mclient.MonitoringV1().Prometheuses(namespace).List(options)
					},
				}
			}).List,
		},
		{
			monitoringv1.ServiceMonitorsKind,
			listwatch.MultiNamespaceListerWatcher(c.config.Namespaces, func(namespace string) cache.ListerWatcher {
				return &cache.ListWatch{
					ListFunc: func(options metav1.ListOptions) (runtime.Object, error) {
						return c.mclient.MonitoringV1().ServiceMonitors(namespace).List(options)
					},
				}
			}).List,
		},
		{
			monitoringv1.PrometheusRuleKind,
			listwatch.MultiNamespaceListerWatcher(c.config.Namespaces, func(namespace string) cache.ListerWatcher {
				return &cache.ListWatch{
					ListFunc: func(options metav1.ListOptions) (runtime.Object, error) {
						return c.mclient.MonitoringV1().PrometheusRules(namespace).List(options)
					},
				}
			}).List,
		},
	}

	for _, crdListFunc := range crdListFuncs {
		err := k8sutil.WaitForCRDReady(crdListFunc.listFunc)
		if err != nil {
			return errors.Wrapf(err, "waiting for %v crd failed", crdListFunc.name)
		}
	}

	return nil
}