-
Notifications
You must be signed in to change notification settings - Fork 73
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
expected mirror to respond with 200 OK but received: 500 Internal Server Error #634
Comments
This error seems to be similar to #620. I do not believe that the error comes from the TTL running out because the image pulls are right after each other. I am going to see if I can reproduce this outside of EKS. Are you able to share your Containerd configuration? |
I am deploying spegel on eks. Is there any special configuration that needs to be done when working with artifactory jfrog? |
There are no specific requirements for the upstream registry. Spegel does not communicate with the registry, if the image does not exist within the cluster the request will fail and Containerd will be responsible to pull from the original registry. There is however specific configuration required for EKS. As Spegel is running I assume that you have already followed the compatibility guide. |
I missed this required configuration for the EKS, I will check that as soon as possible. |
@phillebaba I'm going to try to do some testing here as well; I spotted this thread and know for certain it's something we see quite often here on our deployment. In the past 7 days I see ~100k 500's (~15% of logs). I looked into these a bit a month or two ago and at the time figured it was exactly what you mentioned, that the node had cleaned up the manifest/blob and was still advertising it; but I'm noticing now when I look closer that there are actually multiple spegel mirrors 500ing on the same request chain for the same blob. |
I have same problem when run spegel on talos
The logs error:
|
I have been able to reproduce this with Talos, so I assume it is related to Containerd v2. I am going to add an e2e test with Talos to debug the issue and validate a possible solution. |
@phillebaba
|
Spegel version
v0.0.27
Kubernetes distribution
EKS
Kubernetes version
1.29
CNI
Amazon VPC CNI
Describe the bug
configure spegel to work with jfrog Artifactory.
I deploy the pod on a specific node that takes 102 sec, and then deploy him on another node at the same time.
It looks like the mirroring is not working.
this is the error logs I see:
there is any specific configuration that needs to be added when using jfrog?
The text was updated successfully, but these errors were encountered: