我试图使用gcr.io在GKE上建立自己的容器,并不断地获得ImagePullBackOff失败。
考虑到我做错了什么,我回到这里的教程https://cloud.google.com/kubernetes-engine/docs/tutorials/hello-app,并遵循所有的步骤,并得到相同的错误。这看起来像一个证书问题,但我遵循教程的所有步骤,但仍然没有运气。
如何调试此错误,因为日志似乎没有帮助。
本教程工作的步骤1-4。
kubectl run hello-web --image=gcr.io/${PROJECT_ID}/hello-app:v1 --port 8080使用ImagePullBackOff失败,我认为GKE和gcr.io会自动处理凭据。我做错了什么?我如何调试这个?
kubectl describe pods hello-web-6444d588b7-tqgdm
Name: hello-web-6444d588b7-tqgdm
Namespace: default
Node: gke-aia-default-pool-9ad6a2ee-j5g7/10.152.0.2
Start Time: Sat, 27 Oct 2018 06:51:38 +1000
Labels: pod-template-hash=2000814463
run=hello-web
Annotations: kubernetes.io/limit-ranger=LimitRanger plugin set: cpu request for container hello-web
Status: Pending
IP: 10.12.2.5
Controlled By: ReplicaSet/hello-web-6444d588b7
Containers:
hello-web:
Container ID:
Image: gcr.io/<project-id>/hello-app:v1
Image ID:
Port: 8080/TCP
Host Port: 0/TCP
State: Waiting
Reason: ImagePullBackOff
Ready: False
Restart Count: 0
Requests:
cpu: 100m
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from default-token-qgv8h (ro)
Conditions:
Type Status
Initialized True
Ready False
PodScheduled True
Volumes:
default-token-qgv8h:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-qgv8h
Optional: false
QoS Class: Burstable
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 45m default-scheduler Successfully assigned hello-web-6444d588b7-tqgdm to gke-aia-default-pool-9ad6a2ee-j5g7
Normal SuccessfulMountVolume 45m kubelet, gke-aia-default-pool-9ad6a2ee-j5g7 MountVolume.SetUp succeeded for volume "default-token-qgv8h"
Normal Pulling 44m (x4 over 45m) kubelet, gke-aia-default-pool-9ad6a2ee-j5g7 pulling image "gcr.io/<project-id>/hello-app:v1"
Warning Failed 44m (x4 over 45m) kubelet, gke-aia-default-pool-9ad6a2ee-j5g7 Failed to pull image "gcr.io/<project-id>/hello-app:v1": rpc error: code = Unknown desc = Error response from daemon: repository gcr.io/<project-id>/hello-app not found: does not exist or no pull access
Warning Failed 44m (x4 over 45m) kubelet, gke-aia-default-pool-9ad6a2ee-j5g7 Error: ErrImagePull
Normal BackOff 5m (x168 over 45m) kubelet, gke-aia-default-pool-9ad6a2ee-j5g7 Back-off pulling image "gcr.io/<project-id>/hello-app:v1"
Warning Failed 48s (x189 over 45m) kubelet, gke-aia-default-pool-9ad6a2ee-j5g7 Error: ImagePullBackOff群集权限:
User info Disabled
Compute Engine Read/Write
Storage Read Only
Task queue Disabled
BigQuery Disabled
Cloud SQL Disabled
Cloud Datastore Disabled
Stackdriver Logging API Write Only
Stackdriver Monitoring API Full
Cloud Platform Disabled
Bigtable Data Disabled
Bigtable Admin Disabled
Cloud Pub/Sub Disabled
Service Control Enabled
Service Management Read Only
Stackdriver Trace Write Only
Cloud Source Repositories Disabled
Cloud Debugger Disabled发布于 2018-10-26 23:15:34
在阅读了一些文档之后,我使用以下说明手动添加了access:https://cloud.google.com/container-registry/docs/access-control
现在允许部署示例代码。从gke到gcr的自动访问看起来不起作用。
发布于 2020-10-21 17:57:44
在创建GKE集群时,确保为节点设置了Storage或https://www.googleapis.com/auth/devstorage.read_only作用域。
我在通过Terraform创建GKE集群时碰到了这个问题,并拥有:
node_config {
oauth_scopes = [
"https://www.googleapis.com/auth/logging.write",
"https://www.googleapis.com/auth/monitoring",
]
...而不是
node_config {
oauth_scopes = [
"https://www.googleapis.com/auth/logging.write",
"https://www.googleapis.com/auth/monitoring",
"https://www.googleapis.com/auth/devstorage.read_only"
]
...https://stackoverflow.com/questions/53001219
复制相似问题