ok, thanks - no need to specifically patch for us. Just let me know when the final fix is in so we can test on our end.
We actually ran into another error installing JupyterHub (same version as above) onto a Karpenter enabled Qovery managed EKS cluster:
https://console.qovery.com/organization/5fafa1c2-689c-4a54-8ea1-533a7230a2a5/project/0014bb51-b28a-4048-a385-262dab6ec1b5/environment/36a454d1-f2ef-4055-a397-c43f2f37402b/services/general
Hey @data_admin,
Did you set priority classes as mentioned here?
There is a known issue on Karpenter on Daemonsets and priority class needs to be set.
Hereās another topic having the same issue with a solution.
Thanks, I was able to add the priority class service from the instructions above:
However, itās not clear to me how to āModify your DaemonSet configuration to use the new priority class an redeploy itā
- I tried to search for an answer in your forum but couldnāt find it
- What piece of code do I need to add to the JupyterHub Values override as file?
https://console.qovery.com/organization/5fafa1c2-689c-4a54-8ea1-533a7230a2a5/project/0014bb51-b28a-4048-a385-262dab6ec1b5/environment/36a454d1-f2ef-4055-a397-c43f2f37402b/application/2dd114ff-b9c9-466e-898b-366de1a291a2/settings/values-override-file
Also, Iām having trouble deleting this cluster:
Hello @data_admin ,
Benjamin is looking at your Cluster deletion problem.
Regarding the priorityClass question, I checked JupyterHubās documentation and I think you should be able to delete the custom priorityClass, and use the build-in values.
fullnameOverride: "jupyterhub"
proxy:
service:
type: ClusterIP
scheduling:
podPriority:
enabled: true
globalDefault: true
defaultPriority: 50
imagePullerPriority: 1000
userPlaceholderPriority: 0
You can find the detailed documentation here: Configuration Reference ā Zero to JupyterHub with Kubernetes documentation
Please let me know if this is helpful.
Regards,
Charles-Edouard
1 Like
ok, thanks - no need to specifically patch for us. Just let me know when the final fix is in so we can test on our end.
Hello @data_admin !
The fix has just landed and is available for NON production clusters and will be available for production clusters on Monday.
In the meantime, if you need this patch on production cluster, let me know I can patch it.
You just need to trigger a cluster update to patch your cluster.
Cheers
1 Like
Thanks! I tried it and got this error:
Error: values don't meet the specifications of the schema(s) in the following chart(s):
13 28 Nov, 15:12:52.51 jupyterhub:
14 28 Nov, 15:12:52.51 - proxy: Additional property scheduling is not allowed
Hey @data_admin,
I think itās normal, you put the fields at the wrong place, it should be:
fullnameOverride: "jupyterhub"
proxy:
service:
type: ClusterIP
scheduling:
podPriority:
enabled: true
globalDefault: true
defaultPriority: 50
imagePullerPriority: 1000
userPlaceholderPriority: 0
but yours is:
scheduling should be at root level.
Cluster has been deleted.
1 Like
Thanks - didnāt have too much time to debug yesterday. Actually, the working config should be:
fullnameOverride: "jupyterhub"
proxy:
service:
type: ClusterIP
scheduling:
podPriority:
enabled: true
globalDefault: true
defaultPriority: 50
imagePullerPriority: 1000
userPlaceholderPriority: 0
I got JupyterHub to work on a Qovery managed EKS cluster with Karpenter enabled. I was able to deploy the latest JupyterHub Helm chart v 4.0.0
You may want to update Deploy JupyterHub using Helm | Qovery with the results above.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.