-
Notifications
You must be signed in to change notification settings - Fork 40k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Kubelet.exe on Windows Server Core 1809 - Errors with Failed create pod sandbox.. rpc error: code - Unknown desc ... hnsCall Failed in Win32: The object already exits #72696
Comments
Adding sig labels: /sig Networking |
/sig Network |
Greetings, I have tested K8S on Windows with Flannel Overlay last week, it also fails. Literally, following error usually means the port is occupied by another container. You may run docker ps -a on the Windows box to verify.
|
Hi @guhuajun, What is the best way to get microsoft's help? Should I repost in https://github.com/microsoft/sdn/issues and point to this issues, or just repost and duplicate the content? There does not seem to be a conflict with the ports. Here is a docker ps command after the windows node has joined, but prior to deploying a kubernetes pod:
After running the kubectl apply -f iis-win.yml command, the results from the docker ps commands return:
If I do a |
Thanks for updating. Flannel on Windows are still under developing. Hope we will have a workaround for this issue. |
I am hitting the same |
See resolution in issue #298 referenced above. |
I am hitting the same The object already exits issue on Azure VM windows server 1809 with container. |
What happened:
Using VMWare Fusion on a MAC, I have successfully deployed a CentOS Master and CentOS Node cluster. With the examples I'll provide, it is just using a simple NAT network, but I get the same result if a create a second Network with a static IP address. The Kube Cluster works perfectly. I am also able to Successfully join a Windows 1809 Core Server to the cluster following the documentation found on "https://docs.microsoft.com/en-us/virtualization/windowscontainers/kubernetes/joining-linux-workers". I am using a Flannel network. The Windows Node shows in a ready State.
However, A failure occurs when a deployment is made to the Windows node. When I "apply" a deployment to the cluster with the nodeSelector set to beta.kubernetes.io/os: windows, the pod successfully "begins" deployment on the Windows node. (note: I'm pre-tested downloading and running the Image on the windows node with a docker command and it works, so the correct image is being downloaded).
I have deployed the example found on the previously referenced microsoft documentation, with the same result, but in the simple example shown below I used this deployment:
Here are the 3 nodes - knnode3w is the Windows Server Core 1809, the other nodes are CentOS 7
List of all the pods:
Failed POD Description
Here is the Log (kubelet.exe.knnode3w.KNNODE3W_Administrator.log.ERROR.20190108-120851.540) from the Windows Node:
I ran the following commands to join the Windows Node to the Cluster:
What you expected to happen:
That a simple sample pod would deploy to the Windows Node.
How to reproduce it (as minimally and precisely as possible):
Anything else we need to know?:
Environment:
kubectl version
):On Linux:
On Windows
Cloud provider or hardware configuration: On Premises... running on Mac with vmware fusion Version Professional Version 10.1.5
OS (e.g. from /etc/os-release):
Master and Node1
Windows Server
uname -a
): Linux knmaster 3.10.0-957.el7.x86_64 Unit test coverage in Kubelet is lousy. (~30%) #1 SMP Thu Nov 8 23:39:32 UTC 2018 x86_64 x86_64 x86_64 GNU/LinuxThe text was updated successfully, but these errors were encountered: