Hi TrickySheep9
Long story short, clearml-session fully supports k8s (using k8s glue)
The --remote-gateway along side ports mode will basically allow you to setup a k8s service so that every session will register with a specific port so k8s does ingest foe you and route the SSH connection to the pod itslef, everything else is tunneled over the original SSH connection.
Make sense ?

The way I read that is if you have exposed your clearml-server via k8s ingress then you can, from outside the k8s, say to clearml-session this is the k8s ingress/ip

How is clearml-session intended to be used?

Thanks AgitatedDove14 - i get overall what you are saying. Have to get glue setup, which I couldn’t understand fully, so that’s a different topic 🙂

maybe I am entirely wrong .. I read clearml-session as running a remote 'headless container with jupyter'.. but not on k8s.. or rather... the pre-requisites say that 'ssh access', which you won't have in a k8s cluster (I would hope). I mean.. exposing the ssh port on a k8s cluster is jst a security nightmare.

AlertBlackbird30 - i don’t understand why it can’t be a focus though. Probably missing some context.

I am guessing it could be but.. I don't feel that k8s is clearml-session's main focus/push

Ok so that’s nothing more than what I would configured in the clearml config then

I believe that's correct, yes

AlertBlackbird30 :

--remote-gateway [REMOTE_GATEWAY] Advanced: Specify gateway ip/address to be passed to interactive session (for use with k8s ingestion / ELB)I see this in clearml-session - what’s the intent here?

But ok the summary is I guess it doesn’t work in a k8s env

“notebook as a service” right

Hey Manoj, I am not sure how clearml-session would know how to setup kube-proxy, if that's your intent.

Personally, I would run the clearml-server and agents on the k8s cluster, and then expose the server endpoints via kube proxy or some other nicer ingress. Then you can run jupyter locally and you should be good. Jupyter session remotely running on k8s would be a logistical nightmare 🙂

I would like to create a notebook instance and start using it without having to do anything on a dev box

There is kubectl exec 🙂

Ah ok. Kind of getting it, will have to try the glue mode

Have to get glue setup, which I couldn’t understand fully, so that’s a different topic

I suggest using the apply template setup (basically you provide a Job/Service template, and it uses that to setup k8s jobs based on the Tasks coming in from the specific queue)

👍