You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It'd be nice for users to have a way to replicate their user pod in the job pod so they could access all their data similarly to how they would if they were running in their Jupyterhub user pod. I'd propose the addition of a --replicateUserPod flag which would cause the initContainers, volumes, volumeMounts, and image of the user pod to be added to the job pod. This might be implemented in the following way:
user submits job kbatch submit job --replicateUserPod ...
then kbatch client looks for socket.gethostname() to find what pod we're in, and passes hostname to kbatch-proxy
kbatch-proxy calls kubernetes.client.CoreV1Api().read_namespaced_pod(hostname, namespace) to get pod spec
pulls out initContainers, volumes, volumeMounts, image and add these to spec of job pod to be submitted
kbatch-proxy submits job otherwise the same as it does currently
The text was updated successfully, but these errors were encountered:
It'd be nice for users to have a way to replicate their user pod in the job pod so they could access all their data similarly to how they would if they were running in their Jupyterhub user pod. I'd propose the addition of a
--replicateUserPod
flag which would cause the initContainers, volumes, volumeMounts, and image of the user pod to be added to the job pod. This might be implemented in the following way:kbatch submit job --replicateUserPod ...
socket.gethostname()
to find what pod we're in, and passes hostname to kbatch-proxykubernetes.client.CoreV1Api().read_namespaced_pod(hostname, namespace)
to get pod specinitContainers
,volumes
,volumeMounts
,image
and add these to spec of job pod to be submittedThe text was updated successfully, but these errors were encountered: