Kubernetes 水平吊舱自动缩放无法读取指标
我使用的是卡夫卡头盔图表。 我也在尝试水平吊舱自动缩放 我在templates文件夹中添加了一个hpa.yaml文件，如下所示Kubernetes 水平吊舱自动缩放无法读取指标,kubernetes,apache-kafka,kubernetes-helm,Kubernetes,Apache Kafka,Kubernetes Helm,我使用的是卡夫卡头盔图表。 我也在尝试水平吊舱自动缩放 我在templates文件夹中添加了一个hpa.yaml文件，如下所示 apiVersion: autoscaling/v2beta1 kind: HorizontalPodAutoscaler metadata: name: kafka-hpa spec: scaleTargetRef: apiVersion: extensions/v1beta1 kind: Deployment name: {{ inc
apiVersion: autoscaling/v2beta1
kind: HorizontalPodAutoscaler
metadata:
name: kafka-hpa
spec:
scaleTargetRef:
apiVersion: extensions/v1beta1
kind: Deployment
name: {{ include "kafka.fullname" . }}
minReplicas: {{ .Values.replicas }}
maxReplicas: 5
metrics:
- type: Resource
resource:
name: cpu
targetAverageUtilization: 50
- type: Resource
resource:
name: memory
targetAverageValue: 8000Mi
NAME REFERENCE TARGETS MINPODS MAXPODS REPLICAS AGE
kafka-hpa Deployment/whopping-walrus-kafka <unknown>/8000Mi, <unknown>/50% 3 5 0 1h .
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
metrics-server-55cbf87bbb-vm2v5 1/1 Running 0 15m
whopping-walrus-kafka-0 1/1 Running 1 1h
whopping-walrus-kafka-1 1/1 Running 0 1h
whopping-walrus-kafka-2 1/1 Running 0 1h
whopping-walrus-kafka-exporter-5c66b5b4f9-mv5kv 1/1 Running 1 1h
whopping-walrus-zookeeper-0 1/1 Running 0 1h
$ kubectl get statefulset
NAME DESIRED CURRENT AGE
original-bobcat-kafka 3 2 2m
original-bobcat-zookeeper 1 1 2m
$ kubectl describe hpa
Name: kafka-hpa
Namespace: default
Labels: <none>
Annotations: <none>
CreationTimestamp: Fri, 18 Jan 2019 12:13:59 +0530
Reference: StatefulSet/original-bobcat-kafka
Metrics: ( current / target )
resource memory on pods: <unknown> / 8000Mi
resource cpu on pods (as a percentage of request): <unknown> / 5%
Min replicas: 3
Max replicas: 5
Conditions:
Type Status Reason Message
---- ------ ------ -------
AbleToScale False FailedGetScale the HPA controller was unable to get the target's current scale: no matches for kind "StatefulSet" in group "extensions"
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Warning FailedGetScale 15s (x17 over 8m) horizontal-pod-autoscaler no matches for kind "StatefulSet" in group "extensions"
$ kubectl describe hpa
Name: kafka-hpa
Namespace: default
Labels: <none>
Annotations: <none>
CreationTimestamp: Fri, 18 Jan 2019 12:30:07 +0530
Reference: Deployment/good-elephant-kafka
Metrics: ( current / target )
resource memory on pods: <unknown> / 8000Mi
resource cpu on pods (as a percentage of request): <unknown> / 5%
Min replicas: 3
Max replicas: 5
Conditions:
Type Status Reason Message
---- ------ ------ -------
AbleToScale False FailedGetScale the HPA controller was unable to get the target's current scale: could not fetch the scale for deployments.extensions good-elephant-kafka: deployments/scale.extensions "good-elephant-kafka" not found
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Warning FailedGetScale 9s horizontal-pod-autoscaler could not fetch the scale for deployments.extensions good-elephant-kafka: deployments/scale.extensions "good-elephant-kafka" not found
$kubectl描述hpa
姓名：卡夫卡hpa
名称空间：默认值
标签：
注释：
CreationTimestamp:2019年1月18日星期五12:13:59+0530
参考：StatefulSet/原始山猫卡夫卡
指标：（当前/目标）
吊舱上的资源内存：/8000Mi
POD上的资源cpu（占请求的百分比）：/5%
最小副本数：3
最大副本数：5
条件：
键入状态原因消息
---- ------ ------ -------
AbleToScale False FailedGetScale HPA控制器无法获取目标的当前比例：组“extensions”中的种类“StatefulSet”没有匹配项
活动：
从消息中键入原因年龄
---- ------ ---- ---- -------
警告FailedGetScale 15s（x17大于8m）水平吊舱自动缩放器“扩展”组中的“StatefulSet”类型不匹配
$ kubectl describe hpa
Name: kafka-hpa
Namespace: default
Labels: <none>
Annotations: <none>
CreationTimestamp: Fri, 18 Jan 2019 12:13:59 +0530
Reference: StatefulSet/original-bobcat-kafka
Metrics: ( current / target )
resource memory on pods: <unknown> / 8000Mi
resource cpu on pods (as a percentage of request): <unknown> / 5%
Min replicas: 3
Max replicas: 5
Conditions:
Type Status Reason Message
---- ------ ------ -------
AbleToScale False FailedGetScale the HPA controller was unable to get the target's current scale: no matches for kind "StatefulSet" in group "extensions"
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Warning FailedGetScale 15s (x17 over 8m) horizontal-pod-autoscaler no matches for kind "StatefulSet" in group "extensions"
$ kubectl describe hpa
Name: kafka-hpa
Namespace: default
Labels: <none>
Annotations: <none>
CreationTimestamp: Fri, 18 Jan 2019 12:30:07 +0530
Reference: Deployment/good-elephant-kafka
Metrics: ( current / target )
resource memory on pods: <unknown> / 8000Mi
resource cpu on pods (as a percentage of request): <unknown> / 5%
Min replicas: 3
Max replicas: 5
Conditions:
Type Status Reason Message
---- ------ ------ -------
AbleToScale False FailedGetScale the HPA controller was unable to get the target's current scale: could not fetch the scale for deployments.extensions good-elephant-kafka: deployments/scale.extensions "good-elephant-kafka" not found
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Warning FailedGetScale 9s horizontal-pod-autoscaler could not fetch the scale for deployments.extensions good-elephant-kafka: deployments/scale.extensions "good-elephant-kafka" not found
$kubectl描述hpa
姓名：卡夫卡hpa
名称空间：默认值
标签：
注释：
CreationTimestamp:2019年1月18日星期五12:30:07+0530
参考：部署/大象卡夫卡
指标：（当前/目标）
吊舱上的资源内存：/8000Mi
POD上的资源cpu（占请求的百分比）：/5%
最小副本数：3
最大副本数：5
条件：
键入状态原因消息
---- ------ ------ -------
AbleToScale False FailedGetScale HPA控制器无法获取目标的当前比例：无法获取部署的比例。extensions good elephant kafka:deployments/scale.extensions找不到“good elephant kafka”
活动：
从消息中键入原因年龄
---- ------ ---- ---- -------
警告FailedGetScale 9s水平吊舱自动缩放器无法获取用于部署的比例。extensions good elephant kafka:deployments/scale.extensions找不到“good elephant kafka”
$ kubectl describe pods metrics-server-55cbf87bbb-vm2v5
Name: metrics-server-55cbf87bbb-vm2v5
Namespace: default
Node: docker-for-desktop/192.168.65.3
Start Time: Fri, 18 Jan 2019 11:26:33 +0530
Labels: app=metrics-server
pod-template-hash=1176943666
release=metrics-server
Annotations: <none>
Status: Running
IP: 10.1.0.119
Controlled By: ReplicaSet/metrics-server-55cbf87bbb
Containers:
metrics-server:
Container ID: docker://ee4b3d9ed1b15c2c8783345b0ffbbc565ad25f1493dec0148f245c9581443631
Image: gcr.io/google_containers/metrics-server-amd64:v0.3.1
Image ID: docker-pullable://gcr.io/google_containers/metrics-server-amd64@sha256:78938f933822856f443e6827fe5b37d6cc2f74ae888ac8b33d06fdbe5f8c658b
Port: <none>
Host Port: <none>
Command:
/metrics-server
--kubelet-insecure-tls
--kubelet-preferred-address-types=InternalIP
--logtostderr
State: Running
Started: Fri, 18 Jan 2019 11:26:35 +0530
Ready: True
Restart Count: 0
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from metrics-server-token-d2g7b (ro)
Conditions:
Type Status
Initialized True
Ready True
PodScheduled True
Volumes:
metrics-server-token-d2g7b:
Type: Secret (a volume populated by a Secret)
SecretName: metrics-server-token-d2g7b
Optional: false
QoS Class: BestEffort
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events: <none>
$kubectl描述pods metrics-server-55cbf87bbb-vm2v5
名称：metrics-server-55CBF87BB-vm2v5
名称空间：默认值
节点：docker for desktop/192.168.65.3
开始时间：2019年1月18日星期五11:26:33+0530
标签：app=metrics服务器
pod模板哈希=1176943666
release=metrics服务器
注释：
状态：正在运行
IP:10.1.0.119
控制人：ReplicaSet/metrics-server-55CBF87BB
容器：
度量服务器：
容器ID:docker://ee4b3d9ed1b15c2c8783345b0ffbbc565ad25f1493dec0148f245c9581443631
图片：gcr.io/google_containers/metrics-server-amd64:v0.3.1
图像ID:docker-pullable://gcr.io/google_containers/metrics-server-amd64@sha256:78938f933822856f443e6827fe5b37d6cc2f74ae888ac8b33d06fdbe5f8c658b
端口：
主机端口：
命令：
/度量服务器
--kubelet不安全tls
--kubelet首选地址类型=内部IP
--logtostderr
状态：正在运行
开始时间：2019年1月18日星期五11:26:35+0530
准备好了吗
重新启动计数：0
环境：
挂载：
/var/run/secrets/kubernetes.io/serviceCount来自metrics-server-token-d2g7b（ro）
条件：
类型状态
初始化为True
准备好了吗
播客预定为真
卷数：
metrics-server-token-d2g7b：
类型：Secret（由Secret填充的卷）
SecretName:metrics-server-token-d2g7b
可选：false
QoS等级：最佳努力
节点选择器：
容差：node.kubernetes.io/未就绪：不执行300秒
node.kubernetes.io/不可访问：不执行300秒
活动：
非常感谢您的帮助。您需要在
metrics服务器部署文件中添加以下命令：
containers:
- command:
- /metrics-server
- --metric-resolution=30s
- --kubelet-insecure-tls
- --kubelet-preferred-address-types=InternalIP
name: metrics-server
我相信metrics服务器找不到具有InternalIP
的kubelet，因此出现了问题。有关更多信息，请查看我的以下答案，了解设置HPA的分步说明
您需要在度量服务中添加以下命令