我正在尝试在 Minikube 中运行 docker 容器注册表以测试我正在编写的 CSI 驱动程序。
我在 mac 上运行 minikube 并尝试使用以下 minikube start 命令: minikube start --vm-driver=hyperkit --disk-size=40g 。我已经尝试过 kubeadm 和 localkube bootstrap 以及 virtualbox vm-driver。
这是我用于 registry pod 部署的资源定义。
---
apiVersion: v1
kind: Pod
metadata:
name: registry
labels:
app: registry
namespace: docker-registry
spec:
containers:
- name: registry
image: registry:2
imagePullPolicy: Always
ports:
- containerPort: 5000
volumeMounts:
- mountPath: /var/lib/registry
name: registry-data
volumes:
- hostPath:
path: /var/lib/kubelet/plugins/csi-registry
type: DirectoryOrCreate
name: registry-data
我尝试使用
kubectl apply -f registry-setup.yaml 创建它。在运行之前，我的 minikube 集群报告自己已准备就绪，所有正常的 minikube 容器都在运行。但是，这无法运行，并且在运行
kubectl describe pod 时，我看到以下消息: Name: registry
Namespace: docker-registry
Node: minikube/192.168.64.43
Start Time: Wed, 08 Aug 2018 12:24:27 -0700
Labels: app=registry
Annotations: kubectl.kubernetes.io/last-applied-configuration={"apiVersion":"v1","kind":"Pod","metadata":{"annotations":{},"labels":{"app":"registry"},"name":"registry","namespace":"docker-registry"},"spec":{"cont...
Status: Running
IP: 172.17.0.2
Containers:
registry:
Container ID: docker://42e5193ac563c2b2e2a2b381c91350d30f7e7c5009a30a5977d33b403a374e7f
Image: registry:2
...
TRUNCATED FOR SPACE
...
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 1m default-scheduler Successfully assigned registry to minikube
Normal SuccessfulMountVolume 1m kubelet, minikube MountVolume.SetUp succeeded for volume "registry-data"
Normal SuccessfulMountVolume 1m kubelet, minikube MountVolume.SetUp succeeded for volume "default-token-kq5mq"
Normal Pulling 1m kubelet, minikube pulling image "registry:2"
Normal Pulled 1m kubelet, minikube Successfully pulled image "registry:2"
Normal Created 1m kubelet, minikube Created container
Normal Started 1m kubelet, minikube Started container
...
TRUNCATED
...
Name: storage-provisioner
Namespace: kube-system
Node: minikube/192.168.64.43
Start Time: Wed, 08 Aug 2018 12:24:38 -0700
Labels: addonmanager.kubernetes.io/mode=Reconcile
integration-test=storage-provisioner
Annotations: kubectl.kubernetes.io/last-applied-configuration={"apiVersion":"v1","kind":"Pod","metadata":{"annotations":{},"labels":{"addonmanager.kubernetes.io/mode":"Reconcile","integration-test":"storage-provis...
Status: Pending
IP: 192.168.64.43
Containers:
storage-provisioner:
Container ID:
Image: gcr.io/k8s-minikube/storage-provisioner:v1.8.1
Image ID:
Port: <none>
Host Port: <none>
Command:
/storage-provisioner
State: Waiting
Reason: ErrImagePull
Ready: False
Restart Count: 0
Environment: <none>
Mounts:
/tmp from tmp (rw)
/var/run/secrets/kubernetes.io/serviceaccount from storage-provisioner-token-sb5hz (ro)
Conditions:
Type Status
Initialized True
Ready False
PodScheduled True
Volumes:
tmp:
Type: HostPath (bare host directory volume)
Path: /tmp
HostPathType: Directory
storage-provisioner-token-sb5hz:
Type: Secret (a volume populated by a Secret)
SecretName: storage-provisioner-token-sb5hz
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:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 1m default-scheduler Successfully assigned storage-provisioner to minikube
Normal SuccessfulMountVolume 1m kubelet, minikube MountVolume.SetUp succeeded for volume "tmp"
Normal SuccessfulMountVolume 1m kubelet, minikube MountVolume.SetUp succeeded for volume "storage-provisioner-token-sb5hz"
Normal Pulling 23s (x3 over 1m) kubelet, minikube pulling image "gcr.io/k8s-minikube/storage-provisioner:v1.8.1"
Warning Failed 21s (x3 over 1m) kubelet, minikube Failed to pull image "gcr.io/k8s-minikube/storage-provisioner:v1.8.1": rpc error: code = Unknown desc = failed to register layer: Error processing tar file(exit status 1): write /storage-provisioner: no space left on device
Warning Failed 21s (x3 over 1m) kubelet, minikube Error: ErrImagePull
Normal BackOff 7s (x3 over 1m) kubelet, minikube Back-off pulling image "gcr.io/k8s-minikube/storage-provisioner:v1.8.1"
Warning Failed 7s (x3 over 1m) kubelet, minikube Error: ImagePullBackOff
------------------------------------------------------------
...
因此，当注册表容器正确启动时，一些其他 minikube 服务(包括 dns、http 入口服务等)开始失败，原因如下:
write /storage-provisioner: no space left on device 。尽管为 minikube 分配了 40GB 的磁盘大小，但似乎 minikube 正在尝试写入只有 1GB 空间的 rootfs 或 devtempfs(取决于 vm-driver)。$ df -h
Filesystem Size Used Avail Use% Mounted on
rootfs 919M 713M 206M 78% /
devtmpfs 919M 0 919M 0% /dev
tmpfs 996M 0 996M 0% /dev/shm
tmpfs 996M 8.9M 987M 1% /run
tmpfs 996M 0 996M 0% /sys/fs/cgroup
tmpfs 996M 8.0K 996M 1% /tmp
/dev/sda1 34G 1.3G 30G 4% /mnt/sda1
有没有办法让 minikube 在拉取镜像和创建容器时实际使用分配给/mnt/sda1 而不是 rootfs 的 34GB 空间？
在此先感谢您的帮助!
最佳答案
您需要配置 Minikube 虚拟机以使用 /dev/sda1 而不是 / for Docker。要登录它，请使用 minikube ssh 命令。
比你有两个选择:
/dev/sda1 挂载到 var/lib/docker ，但不要忘记在此之前将原始 var/lib/docker 的内容复制到 /mnt/sda1 。 /mnt/sda1 而不是 var/lib/docker 来存储图像。查看此 link 以获取有关它的更多信息。 关于kubernetes - 尽管 --disk-size 标志，Minikube 空间不足并失败，我们在Stack Overflow上找到一个类似的问题：https://stackoverflow.com/questions/51755703/