이 문서에서는 Google Distributed Cloud에서 수평형 포드 자동 확장 처리(HPA)에 대한 사용자 정의 측정항목을 구성하는 방법을 설명합니다.
이 페이지는 회사 또는 사업부의 총소유비용을 최소화하기 위해 시스템 아키텍처 및 리소스를 최적화하고 용량 및 인프라 요구사항을 계획하는 관리자, 설계자, 운영자를 대상으로 합니다. Google Cloud 콘텐츠에서 참조하는 일반적인 역할 및 예시 태스크에 대해 자세히 알아보려면 일반 GKE 사용자 역할 및 태스크를 참조하세요.
Prometheus 및 측정항목 어댑터 배포
이 섹션에서는 Prometheus를 배포하여 사용자 정의 측정항목을 스크래핑하고, prometheus-adapter를 배포하여 Prometheus를 백엔드로 사용하는 Kubernetes Custom Metrics API를 처리합니다.
다음 매니페스트를 custom-metrics-adapter.yaml이라는 파일에 저장합니다.
Prometheus 및 측정항목 어댑터용 매니페스트 파일 콘텐츠
# Copyright 2018 Google Inc
#
# 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.
apiVersion: v1
kind: ServiceAccount
metadata:
name: stackdriver-prometheus
namespace: kube-system
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
name: stackdriver-prometheus
namespace: kube-system
rules:
- apiGroups:
- ""
resources:
- nodes
- services
- endpoints
- pods
verbs:
- get
- list
- watch
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
name: stackdriver-prometheus
namespace: kube-system
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: stackdriver-prometheus
subjects:
- kind: ServiceAccount
name: stackdriver-prometheus
namespace: kube-system
---
apiVersion: v1
kind: Service
metadata:
name: stackdriver-prometheus-app
namespace: kube-system
labels:
app: stackdriver-prometheus-app
spec:
clusterIP: "None"
ports:
- name: http
port: 9090
protocol: TCP
targetPort: 9090
sessionAffinity: ClientIP
selector:
app: stackdriver-prometheus-app
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: stackdriver-prometheus-app
namespace: kube-system
labels:
app: stackdriver-prometheus-app
spec:
replicas: 1
selector:
matchLabels:
app: stackdriver-prometheus-app
template:
metadata:
labels:
app: stackdriver-prometheus-app
spec:
serviceAccount: stackdriver-prometheus
containers:
- name: prometheus-server
image: prom/prometheus:v2.45.0
args:
- "--config.file=/etc/prometheus/config/prometheus.yaml"
- "--storage.tsdb.path=/data"
- "--storage.tsdb.retention.time=2h"
ports:
- name: prometheus
containerPort: 9090
readinessProbe:
httpGet:
path: /-/ready
port: 9090
periodSeconds: 5
timeoutSeconds: 3
# Allow up to 10m on startup for data recovery
failureThreshold: 120
livenessProbe:
httpGet:
path: /-/healthy
port: 9090
periodSeconds: 5
timeoutSeconds: 3
failureThreshold: 6
resources:
requests:
cpu: 250m
memory: 500Mi
volumeMounts:
- name: config-volume
mountPath: /etc/prometheus/config
- name: stackdriver-prometheus-app-data
mountPath: /data
volumes:
- name: config-volume
configMap:
name: stackdriver-prometheus-app
- name: stackdriver-prometheus-app-data
emptyDir: {}
terminationGracePeriodSeconds: 300
nodeSelector:
kubernetes.io/os: linux
---
apiVersion: v1
data:
prometheus.yaml: |
global:
scrape_interval: 1m
rule_files:
- /etc/config/rules.yaml
- /etc/config/alerts.yaml
scrape_configs:
- job_name: prometheus-io-endpoints
kubernetes_sd_configs:
- role: endpoints
relabel_configs:
- action: keep
regex: true
source_labels:
- __meta_kubernetes_service_annotation_prometheus_io_scrape
- action: replace
regex: (.+)
source_labels:
- __meta_kubernetes_service_annotation_prometheus_io_path
target_label: __metrics_path__
- action: replace
regex: (https?)
source_labels:
- __meta_kubernetes_service_annotation_prometheus_io_scheme
target_label: __scheme__
- action: replace
regex: ([^:]+)(?::\d+)?;(\d+)
replacement: $1:$2
source_labels:
- __address__
- __meta_kubernetes_service_annotation_prometheus_io_port
target_label: __address__
- action: replace
source_labels:
- __meta_kubernetes_namespace
target_label: namespace
- action: replace
source_labels:
- __meta_kubernetes_pod_name
target_label: pod
- action: keep
regex: (.+)
source_labels:
- __meta_kubernetes_endpoint_port_name
- job_name: prometheus-io-services
kubernetes_sd_configs:
- role: service
metrics_path: /probe
params:
module:
- http_2xx
relabel_configs:
- action: replace
source_labels:
- __address__
target_label: __param_target
- action: replace
replacement: blackbox
target_label: __address__
- action: keep
regex: true
source_labels:
- __meta_kubernetes_service_annotation_prometheus_io_probe
- action: replace
source_labels:
- __meta_kubernetes_namespace
target_label: namespace
- action: replace
source_labels:
- __meta_kubernetes_pod_name
target_label: pod
- job_name: prometheus-io-pods
kubernetes_sd_configs:
- role: pod
relabel_configs:
- action: keep
regex: true
source_labels:
- __meta_kubernetes_pod_annotation_prometheus_io_scrape
- action: replace
regex: (.+)
source_labels:
- __meta_kubernetes_pod_annotation_prometheus_io_path
target_label: __metrics_path__
- action: replace
regex: ([^:]+)(?::\d+)?;(\d+)
replacement: $1:$2
source_labels:
- __address__
- __meta_kubernetes_pod_annotation_prometheus_io_port
target_label: __address__
- action: replace
source_labels:
- __meta_kubernetes_namespace
target_label: namespace
- action: replace
source_labels:
- __meta_kubernetes_pod_name
target_label: pod
kind: ConfigMap
metadata:
name: stackdriver-prometheus-app
namespace: kube-system
---
# The main section of custom metrics adapter.
kind: ServiceAccount
apiVersion: v1
metadata:
name: custom-metrics-apiserver
namespace: kube-system
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
name: custom-metrics:system:auth-delegator
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: system:auth-delegator
subjects:
- kind: ServiceAccount
name: custom-metrics-apiserver
namespace: kube-system
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
name: custom-metrics-server-resources
rules:
- apiGroups:
- custom.metrics.k8s.io
resources: ["*"]
verbs: ["*"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
name: custom-metrics-resource-reader
rules:
- apiGroups:
- ""
resources:
- nodes
- namespaces
- pods
- services
verbs:
- get
- watch
- list
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
name: custom-metrics-resource-reader
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: custom-metrics-resource-reader
subjects:
- kind: ServiceAccount
name: custom-metrics-apiserver
namespace: kube-system
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
name: custom-metrics-auth-reader
namespace: kube-system
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: Role
name: extension-apiserver-authentication-reader
subjects:
- kind: ServiceAccount
name: custom-metrics-apiserver
namespace: kube-system
---
apiVersion: v1
kind: ConfigMap
metadata:
name: adapter-config
namespace: kube-system
data:
config.yaml: |
rules:
default: false
# fliter all metrics
- seriesQuery: '{pod=~".+"}'
seriesFilters: []
resources:
# resource name is mapped as it is. ex. namespace -> namespace
template: <<.Resource>>
name:
matches: ^(.*)$
as: ""
# Aggregate metric on resource level
metricsQuery: sum(<<.Series>>{<<.LabelMatchers>>}) by (<<.GroupBy>>)
---
apiVersion: apps/v1
kind: Deployment
metadata:
labels:
app: custom-metrics-apiserver
name: custom-metrics-apiserver
namespace: kube-system
spec:
replicas: 1
selector:
matchLabels:
app: custom-metrics-apiserver
template:
metadata:
labels:
app: custom-metrics-apiserver
name: custom-metrics-apiserver
spec:
serviceAccountName: custom-metrics-apiserver
containers:
- name: custom-metrics-apiserver
resources:
requests:
cpu: 15m
memory: 20Mi
limits:
cpu: 100m
memory: 150Mi
image: registry.k8s.io/prometheus-adapter/prometheus-adapter:v0.11.0
args:
- /adapter
- --cert-dir=/var/run/serving-cert
- --secure-port=6443
- --prometheus-url=http://stackdriver-prometheus-app.kube-system.svc:9090/
- --metrics-relist-interval=1m
- --config=/etc/adapter/config.yaml
ports:
- containerPort: 6443
volumeMounts:
- name: serving-cert
mountPath: /var/run/serving-cert
- mountPath: /etc/adapter/
name: config
readOnly: true
nodeSelector:
kubernetes.io/os: linux
volumes:
- name: serving-cert
emptyDir:
medium: Memory
- name: config
configMap:
name: adapter-config
---
apiVersion: v1
kind: Service
metadata:
name: custom-metrics-apiserver
namespace: kube-system
spec:
ports:
- port: 443
targetPort: 6443
selector:
app: custom-metrics-apiserver
---
apiVersion: apiregistration.k8s.io/v1
kind: APIService
metadata:
name: v1beta1.custom.metrics.k8s.io
spec:
service:
name: custom-metrics-apiserver
namespace: kube-system
group: custom.metrics.k8s.io
version: v1beta1
insecureSkipTLSVerify: true
groupPriorityMinimum: 100
versionPriority: 100
---
apiVersion: apiregistration.k8s.io/v1
kind: APIService
metadata:
name: v1beta2.custom.metrics.k8s.io
spec:
service:
name: custom-metrics-apiserver
namespace: kube-system
group: custom.metrics.k8s.io
version: v1beta2
insecureSkipTLSVerify: true
groupPriorityMinimum: 100
versionPriority: 100
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
name: hpa-controller-custom-metrics
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: custom-metrics-server-resources
subjects:
- kind: ServiceAccount
name: horizontal-pod-autoscaler
namespace: kube-system
배포와 서비스를 만듭니다.
kubectl --kubeconfig USER_CLUSTER_KUBECONFIG apply -f custom-metrics-adapter.yaml
다음 단계는 측정항목 수집을 위해 사용자 애플리케이션에 주석을 추가하는 것입니다.
측정항목 수집을 위해 사용자 애플리케이션에 주석 추가
스크레이핑할 사용자 애플리케이션과 Cloud Monitoring으로 전송된 로그에 주석을 추가하려면 해당 annotations를 서비스, pod, 엔드포인트의 메타데이터에 추가해야 합니다.
metadata:
name: "example-monitoring"
namespace: "default"
annotations:
prometheus.io/scrape: "true"
prometheus.io/path: "" - Overriding metrics path (default "/metrics")
예시 사용자 애플리케이션 배포
이 섹션에서는 로그 및 Prometheus 호환 측정항목과 함께 샘플 애플리케이션을 배포합니다.
다음 서비스 및 배포 매니페스트를
my-app.yaml파일에 저장합니다. 서비스에는prometheus.io/scrape: "true"주석이 있습니다.kind: Service apiVersion: v1 metadata: name: "example-monitoring" namespace: "default" annotations: prometheus.io/scrape: "true" spec: selector: app: "example-monitoring" ports: - name: http port: 9090 --- apiVersion: apps/v1 kind: Deployment metadata: name: "example-monitoring" namespace: "default" labels: app: "example-monitoring" spec: replicas: 1 selector: matchLabels: app: "example-monitoring" template: metadata: labels: app: "example-monitoring" spec: containers: - image: gcr.io/google-samples/prometheus-dummy-exporter:v0.2.0 name: prometheus-example-exporter command: - ./prometheus-dummy-exporter args: - --metric-name=example_monitoring_up - --metric-value=1 - --port=9090 resources: requests: cpu: 100m배포와 서비스를 만듭니다.
kubectl --kubeconfig USER_CLUSTER_KUBECONFIG apply -f my-app.yaml
HPA에서 커스텀 측정항목 사용
HPA 객체를 배포하여 이전 단계에서 노출된 측정항목을 사용합니다. 다양한 유형의 커스텀 측정항목에 대한 자세한 내용은 여러 측정항목 및 커스텀 측정항목에 대한 자동 확장을 참조하세요.
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
name: example-monitoring-hpa
namespace: default
spec:
scaleTargetRef:
apiVersion: apps/v1
kind: Deployment
name: example-monitoring
minReplicas: 1
maxReplicas: 5
metrics:
- type: Pods
pods:
metric:
name: example_monitoring_up
target:
type: AverageValue
averageValue: 20
포드 유형 측정항목에는 대상 포드의 라벨에 대한 기본 측정항목 선택자가 있으며, 이에 따라 kube-controller-manager가 작동합니다. 이 예시에서는 대상 pod에서 사용할 수 있는 {matchLabels: {app: example-monitoring}} 선택기로 example_monitoring_up 측정항목을 쿼리할 수 있습니다. 지정된 다른 선택기가 목록에 추가됩니다. 기본 선택기를 사용하지 않으려면 대상 pod에서 라벨을 삭제하거나 객체 유형 측정항목을 사용하면 됩니다.
사용자 정의 애플리케이션 측정항목이 HPA에서 사용되는지 확인
사용자 정의 애플리케이션 측정항목이 HPA에서 사용되는지 확인합니다.
kubectl --kubeconfig=USER_CLUSTER_KUBECONFIG describe hpa example-monitoring-hpa
출력은 다음과 같습니다.
Name: example-monitoring-hpa Namespace: default Labels:Annotations: autoscaling.alpha.kubernetes.io/conditions: [{"type":"AbleToScale","status":"True","lastTransitionTime":"2023-08-23T22:07:24Z","reason":"ReadyForNewScale","message":"recommended size... autoscaling.alpha.kubernetes.io/current-metrics: [{"type":"Pods","pods":{"metricName":"example_monitoring_up","currentAverageValue":"1"}}] autoscaling.alpha.kubernetes.io/metrics: [{"type":"Pods","pods":{"metricName":"example_monitoring_up","targetAverageValue":"20"}}] CreationTimestamp: Wed, 23 Aug 2023 22:07:09 +0000 Reference: Deployment/example-monitoring Min replicas: 1 Max replicas: 5 Deployment pods: 1 current / 1 desired
비용
HPA의 커스텀 측정항목을 사용해도 추가 Cloud Monitoring 비용이 발생하지 않습니다. 커스텀 측정항목을 사용 설정하는 포드는 스크래핑하는 측정항목의 양에 따라 추가 CPU 및 메모리를 사용합니다.