使用 Helm3 在 Kubernetes Minikube 上安装 Prometheus Operator 的问题
Issue installing Prometheus Operator on Kubernetes Minikube with Helm3
我一直在尝试使用 Prometheus 来监控 http_request_rate and/or packets_per_second 的 pod 统计信息。为此，我计划使用 Prometheus Adapter，根据我的阅读，需要使用 Prometheus Operator。
我在从 helm 稳定图表安装 Prometheus Operator 时遇到了问题。当 运行 安装命令 "helm install prom stable/prometheus-operator" 我收到以下警告消息显示六次
$ manifest_sorter.go:192 info: skipping unknown hook: "crd-install".
安装继续，pods 已部署，但是，prometheus-node-exporter pod 进入状态：CrashLoopBackOff。
我看不出详细的原因，因为描述 pods 时的消息是 "Back-off restarting failed container"
我是 运行 Minikube 版本：1.7.2.
我是 运行 Helm 版本：3.1.1.
>>>更新<<<
描述有问题的 Pod 的输出
> $ kubectl describe pod prom-oper-prometheus-node-exporter-2m6vm -n default
>
> Name: prom-oper-prometheus-node-exporter-2m6vm Namespace:
> default Priority: 0 Node: max-ubuntu/10.2.40.198 Start
> Time: Wed, 04 Mar 2020 18:06:44 +0000 Labels:
> app=prometheus-node-exporter
> chart=prometheus-node-exporter-1.8.2
> controller-revision-hash=68695df4c5
> heritage=Helm
> jobLabel=node-exporter
> pod-template-generation=1
> release=prom-oper Annotations: <none> Status: Running IP: 10.2.40.198 IPs: IP: 10.2.40.198
> Controlled By: DaemonSet/prom-oper-prometheus-node-exporter
> Containers: node-exporter:
> Container ID: docker://50b2398f72a0269672c4ac73bbd1b67f49732362b4838e16cd10e3a5247fdbfe
> Image: quay.io/prometheus/node-exporter:v0.18.1
> Image ID: docker-pullable://quay.io/prometheus/node-exporter@sha256:a2f29256e53cc3e0b64d7a472512600b2e9410347d53cdc85b49f659c17e02ee
> Port: 9100/TCP
> Host Port: 9100/TCP
> Args:
> --path.procfs=/host/proc
> --path.sysfs=/host/sys
> --web.listen-address=0.0.0.0:9100
> --collector.filesystem.ignored-mount-points=^/(dev|proc|sys|var/lib/docker/.+)($|/)
> --collector.filesystem.ignored-fs-types=^(autofs|binfmt_misc|cgroup|configfs|debugfs|devpts|devtmpfs|fusectl|hugetlbfs|mqueue|overlay|proc|procfs|pstore|rpc_pipefs|securityfs|sysfs|tracefs)$
> State: Waiting
> Reason: CrashLoopBackOff
> Last State: Terminated
> Reason: Error
> Exit Code: 1
> Started: Wed, 04 Mar 2020 18:10:10 +0000
> Finished: Wed, 04 Mar 2020 18:10:10 +0000
> Ready: False
> Restart Count: 5
> Liveness: http-get http://:9100/ delay=0s timeout=1s period=10s #success=1 #failure=3
> Readiness: http-get http://:9100/ delay=0s timeout=1s period=10s #success=1 #failure=3
> Environment: <none>
> Mounts:
> /host/proc from proc (ro)
> /host/sys from sys (ro)
> /var/run/secrets/kubernetes.io/serviceaccount from prom-oper-prometheus-node-exporter-token-n9dj9 (ro) Conditions: Type
> Status Initialized True Ready False
> ContainersReady False PodScheduled True Volumes: proc:
> Type: HostPath (bare host directory volume)
> Path: /proc
> HostPathType: sys:
> Type: HostPath (bare host directory volume)
> Path: /sys
> HostPathType: prom-oper-prometheus-node-exporter-token-n9dj9:
> Type: Secret (a volume populated by a Secret)
> SecretName: prom-oper-prometheus-node-exporter-token-n9dj9
> Optional: false QoS Class: BestEffort Node-Selectors: <none> Tolerations: :NoSchedule
> node.kubernetes.io/disk-pressure:NoSchedule
> node.kubernetes.io/memory-pressure:NoSchedule
> node.kubernetes.io/network-unavailable:NoSchedule
> node.kubernetes.io/not-ready:NoExecute
> node.kubernetes.io/pid-pressure:NoSchedule
> node.kubernetes.io/unreachable:NoExecute
> node.kubernetes.io/unschedulable:NoSchedule Events: Type Reason Age From
> Message ---- ------ ---- ----
> ------- Normal Scheduled 5m26s default-scheduler Successfully assigned default/prom-oper-prometheus-node-exporter-2m6vm
> to max-ubuntu Normal Started 4m28s (x4 over 5m22s) kubelet,
> max-ubuntu Started container node-exporter Normal Pulled
> 3m35s (x5 over 5m24s) kubelet, max-ubuntu Container image
> "quay.io/prometheus/node-exporter:v0.18.1" already present on machine
> Normal Created 3m35s (x5 over 5m24s) kubelet, max-ubuntu
> Created container node-exporter Warning BackOff 13s (x30 over
> 5m18s) kubelet, max-ubuntu Back-off restarting failed container
有问题的 Pod 日志的输出
> $ kubectl logs prom-oper-prometheus-node-exporter-2m6vm -n default
> time="2020-03-04T18:18:02Z" level=info msg="Starting node_exporter
> (version=0.18.1, branch=HEAD,
> revision=3db77732e925c08f675d7404a8c46466b2ece83e)"
> source="node_exporter.go:156" time="2020-03-04T18:18:02Z" level=info
> msg="Build context (go=go1.12.5, user=root@b50852a1acba,
> date=20190604-16:41:18)" source="node_exporter.go:157"
> time="2020-03-04T18:18:02Z" level=info msg="Enabled collectors:"
> source="node_exporter.go:97" time="2020-03-04T18:18:02Z" level=info
> msg=" - arp" source="node_exporter.go:104" time="2020-03-04T18:18:02Z"
> level=info msg=" - bcache" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - bonding"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - conntrack" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - cpu"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - cpufreq" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - diskstats"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - edac" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - entropy"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - filefd" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - filesystem"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - hwmon" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - infiniband"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - ipvs" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - loadavg"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - mdadm" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - meminfo"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - netclass" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - netdev"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - netstat" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - nfs"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - nfsd" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - pressure"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - sockstat" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - stat"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - textfile" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - time"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - timex" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - uname"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - vmstat" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - xfs"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - zfs" source="node_exporter.go:104" time="2020-03-04T18:18:02Z"
> level=info msg="Listening on 0.0.0.0:9100"
> source="node_exporter.go:170" time="2020-03-04T18:18:02Z" level=fatal
> msg="listen tcp 0.0.0.0:9100: bind: address already in use"
> source="node_exporter.go:172"
这是与 Helm 3 相关的已知问题之一。它影响了许多图表，因为 argo or ambassador. You can find in Helm docs 信息表明 crd-install 挂钩已被删除：
Note that the crd-install hook has been removed in favor of the crds/ directory in Helm 3.
我已经部署了这个图表，还得到了 Helm 跳过未知挂钩但 pods 没有问题的信息。
另一种方法是在安装图表之前创建 CRD's。可以找到执行此操作的步骤 here.
在第一步中，您有创建 CRD 的命令：
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_alertmanagers.yaml
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_podmonitors.yaml
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_prometheuses.yaml
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_prometheusrules.yaml
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_servicemonitors.yaml
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_thanosrulers.yaml
最后一步是执行 Helm install:
helm install --name my-release stable/prometheus-operator --set prometheusOperator.createCustomResource=false
但是Helm 3不会识别--name标志。
Error: unknown flag: --name
您必须删除此标志。它应该看起来像：
$ helm install prom-oper stable/prometheus-operator --set prometheusOperator.createCustomResource=false
NAME: prom-oper
LAST DEPLOYED: Wed Mar 4 14:12:35 2020
NAMESPACE: default
STATUS: deployed
REVISION: 1
NOTES:
The Prometheus Operator has been installed. Check its status by running:
kubectl --namespace default get pods -l "release=prom-oper"
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
alertmanager-prom-oper-prometheus-opera-alertmanager-0 2/2 Running 0 9m46s
...
prom-oper-prometheus-node-exporter-25b27 1/1 Running 0 9m56s
如果你有一些关于回购的问题，你只需要执行：
helm repo add stable https://kubernetes-charts.storage.googleapis.com
helm repo update
如果这种替代方法没有帮助，请添加到您的问题输出中：
kubectl describe <pod-name> -n <pod-namespace> 和 kubectl logs <pod-name> -n <pod-namespace>
这个问题原来是由 Minikube 与 --vm-driver=none 运行 造成的。为了解决这个问题，使用 --vm-driver=kvm2 和 --memory=6g 重建了 Minikube。这允许安装 stable/prometheus-operator 和所有 pods 运行 而不会崩溃。
我一直在尝试使用 Prometheus 来监控 http_request_rate and/or packets_per_second 的 pod 统计信息。为此，我计划使用 Prometheus Adapter，根据我的阅读，需要使用 Prometheus Operator。
我在从 helm 稳定图表安装 Prometheus Operator 时遇到了问题。当 运行 安装命令 "helm install prom stable/prometheus-operator" 我收到以下警告消息显示六次
$ manifest_sorter.go:192 info: skipping unknown hook: "crd-install".
安装继续，pods 已部署，但是，prometheus-node-exporter pod 进入状态：CrashLoopBackOff。
我看不出详细的原因，因为描述 pods 时的消息是 "Back-off restarting failed container"
我是 运行 Minikube 版本：1.7.2.
我是 运行 Helm 版本：3.1.1.
>>>更新<<<
描述有问题的 Pod 的输出
> $ kubectl describe pod prom-oper-prometheus-node-exporter-2m6vm -n default
>
> Name: prom-oper-prometheus-node-exporter-2m6vm Namespace:
> default Priority: 0 Node: max-ubuntu/10.2.40.198 Start
> Time: Wed, 04 Mar 2020 18:06:44 +0000 Labels:
> app=prometheus-node-exporter
> chart=prometheus-node-exporter-1.8.2
> controller-revision-hash=68695df4c5
> heritage=Helm
> jobLabel=node-exporter
> pod-template-generation=1
> release=prom-oper Annotations: <none> Status: Running IP: 10.2.40.198 IPs: IP: 10.2.40.198
> Controlled By: DaemonSet/prom-oper-prometheus-node-exporter
> Containers: node-exporter:
> Container ID: docker://50b2398f72a0269672c4ac73bbd1b67f49732362b4838e16cd10e3a5247fdbfe
> Image: quay.io/prometheus/node-exporter:v0.18.1
> Image ID: docker-pullable://quay.io/prometheus/node-exporter@sha256:a2f29256e53cc3e0b64d7a472512600b2e9410347d53cdc85b49f659c17e02ee
> Port: 9100/TCP
> Host Port: 9100/TCP
> Args:
> --path.procfs=/host/proc
> --path.sysfs=/host/sys
> --web.listen-address=0.0.0.0:9100
> --collector.filesystem.ignored-mount-points=^/(dev|proc|sys|var/lib/docker/.+)($|/)
> --collector.filesystem.ignored-fs-types=^(autofs|binfmt_misc|cgroup|configfs|debugfs|devpts|devtmpfs|fusectl|hugetlbfs|mqueue|overlay|proc|procfs|pstore|rpc_pipefs|securityfs|sysfs|tracefs)$
> State: Waiting
> Reason: CrashLoopBackOff
> Last State: Terminated
> Reason: Error
> Exit Code: 1
> Started: Wed, 04 Mar 2020 18:10:10 +0000
> Finished: Wed, 04 Mar 2020 18:10:10 +0000
> Ready: False
> Restart Count: 5
> Liveness: http-get http://:9100/ delay=0s timeout=1s period=10s #success=1 #failure=3
> Readiness: http-get http://:9100/ delay=0s timeout=1s period=10s #success=1 #failure=3
> Environment: <none>
> Mounts:
> /host/proc from proc (ro)
> /host/sys from sys (ro)
> /var/run/secrets/kubernetes.io/serviceaccount from prom-oper-prometheus-node-exporter-token-n9dj9 (ro) Conditions: Type
> Status Initialized True Ready False
> ContainersReady False PodScheduled True Volumes: proc:
> Type: HostPath (bare host directory volume)
> Path: /proc
> HostPathType: sys:
> Type: HostPath (bare host directory volume)
> Path: /sys
> HostPathType: prom-oper-prometheus-node-exporter-token-n9dj9:
> Type: Secret (a volume populated by a Secret)
> SecretName: prom-oper-prometheus-node-exporter-token-n9dj9
> Optional: false QoS Class: BestEffort Node-Selectors: <none> Tolerations: :NoSchedule
> node.kubernetes.io/disk-pressure:NoSchedule
> node.kubernetes.io/memory-pressure:NoSchedule
> node.kubernetes.io/network-unavailable:NoSchedule
> node.kubernetes.io/not-ready:NoExecute
> node.kubernetes.io/pid-pressure:NoSchedule
> node.kubernetes.io/unreachable:NoExecute
> node.kubernetes.io/unschedulable:NoSchedule Events: Type Reason Age From
> Message ---- ------ ---- ----
> ------- Normal Scheduled 5m26s default-scheduler Successfully assigned default/prom-oper-prometheus-node-exporter-2m6vm
> to max-ubuntu Normal Started 4m28s (x4 over 5m22s) kubelet,
> max-ubuntu Started container node-exporter Normal Pulled
> 3m35s (x5 over 5m24s) kubelet, max-ubuntu Container image
> "quay.io/prometheus/node-exporter:v0.18.1" already present on machine
> Normal Created 3m35s (x5 over 5m24s) kubelet, max-ubuntu
> Created container node-exporter Warning BackOff 13s (x30 over
> 5m18s) kubelet, max-ubuntu Back-off restarting failed container
有问题的 Pod 日志的输出
> $ kubectl logs prom-oper-prometheus-node-exporter-2m6vm -n default
> time="2020-03-04T18:18:02Z" level=info msg="Starting node_exporter
> (version=0.18.1, branch=HEAD,
> revision=3db77732e925c08f675d7404a8c46466b2ece83e)"
> source="node_exporter.go:156" time="2020-03-04T18:18:02Z" level=info
> msg="Build context (go=go1.12.5, user=root@b50852a1acba,
> date=20190604-16:41:18)" source="node_exporter.go:157"
> time="2020-03-04T18:18:02Z" level=info msg="Enabled collectors:"
> source="node_exporter.go:97" time="2020-03-04T18:18:02Z" level=info
> msg=" - arp" source="node_exporter.go:104" time="2020-03-04T18:18:02Z"
> level=info msg=" - bcache" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - bonding"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - conntrack" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - cpu"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - cpufreq" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - diskstats"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - edac" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - entropy"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - filefd" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - filesystem"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - hwmon" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - infiniband"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - ipvs" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - loadavg"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - mdadm" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - meminfo"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - netclass" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - netdev"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - netstat" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - nfs"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - nfsd" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - pressure"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - sockstat" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - stat"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - textfile" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - time"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - timex" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - uname"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - vmstat" source="node_exporter.go:104"
> time="2020-03-04T18:18:02Z" level=info msg=" - xfs"
> source="node_exporter.go:104" time="2020-03-04T18:18:02Z" level=info
> msg=" - zfs" source="node_exporter.go:104" time="2020-03-04T18:18:02Z"
> level=info msg="Listening on 0.0.0.0:9100"
> source="node_exporter.go:170" time="2020-03-04T18:18:02Z" level=fatal
> msg="listen tcp 0.0.0.0:9100: bind: address already in use"
> source="node_exporter.go:172"
这是与 Helm 3 相关的已知问题之一。它影响了许多图表，因为 argo or ambassador. You can find in Helm docs 信息表明 crd-install 挂钩已被删除：
Note that the
crd-installhook has been removed in favor of thecrds/directory in Helm 3.
我已经部署了这个图表，还得到了 Helm 跳过未知挂钩但 pods 没有问题的信息。
另一种方法是在安装图表之前创建 CRD's。可以找到执行此操作的步骤 here.
在第一步中，您有创建 CRD 的命令：
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_alertmanagers.yaml
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_podmonitors.yaml
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_prometheuses.yaml
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_prometheusrules.yaml
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_servicemonitors.yaml
kubectl apply -f https://raw.githubusercontent.com/coreos/prometheus-operator/release-0.36/example/prometheus-operator-crd/monitoring.coreos.com_thanosrulers.yaml
最后一步是执行 Helm install:
helm install --name my-release stable/prometheus-operator --set prometheusOperator.createCustomResource=false
但是Helm 3不会识别--name标志。
Error: unknown flag: --name
您必须删除此标志。它应该看起来像：
$ helm install prom-oper stable/prometheus-operator --set prometheusOperator.createCustomResource=false
NAME: prom-oper
LAST DEPLOYED: Wed Mar 4 14:12:35 2020
NAMESPACE: default
STATUS: deployed
REVISION: 1
NOTES:
The Prometheus Operator has been installed. Check its status by running:
kubectl --namespace default get pods -l "release=prom-oper"
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
alertmanager-prom-oper-prometheus-opera-alertmanager-0 2/2 Running 0 9m46s
...
prom-oper-prometheus-node-exporter-25b27 1/1 Running 0 9m56s
如果你有一些关于回购的问题，你只需要执行：
helm repo add stable https://kubernetes-charts.storage.googleapis.com
helm repo update
如果这种替代方法没有帮助，请添加到您的问题输出中：
kubectl describe <pod-name> -n <pod-namespace> 和 kubectl logs <pod-name> -n <pod-namespace>
这个问题原来是由 Minikube 与 --vm-driver=none 运行 造成的。为了解决这个问题，使用 --vm-driver=kvm2 和 --memory=6g 重建了 Minikube。这允许安装 stable/prometheus-operator 和所有 pods 运行 而不会崩溃。