What if you set the default_output_uri to false ?

@<1523701083040387072:profile|UnevenDolphin73> , basically, it scales to as many pods as you like. Very similar to the autoscaler but on top of K8s

I'm afraid that's all there is. I think security integrations are in the Scale & Enterprise versions.

The ValueError is happening because there is no queue called services it appears

I see. I don't think it's supported but I think it would be a great idea for a feature. Maybe Open a Github feature request?

What about Task.unregister_artifact ?
https://clear.ml/docs/latest/docs/references/sdk/task#unregister_artifact

VexedCat68 , what errors are you getting? What exactly is not working, the webserver or apiserver? Are you trying to access the server from the machine you set it up on or remotely?

@<1734020208089108480:profile|WickedHare16> , many different things, RBAC, users & groups, K8s dedicated support with advanced features, HyperDatasets, SSO/LDAP integration, dedicated support, dynamic GPU allocation, advanced GPU fractioning on top of K8s and much more.

You can see a more detailed list here - None

I would suggest contacting sales@clear.ml for more information 🙂

Hi @<1734020162731905024:profile|RattyBluewhale45> , from the error it looks like there is no space left on the pod. Are you able to run this code manually?

Does any exit code appear? What is the status message and status reason in the 'INFO' section?

This is doen via clearml-agent this is the second link I provided - None

TartLeopard58 , I think you need to mount apiserver.conf to the api server. This is an API configuration 🙂

This is what I just tested now for a task with a commit in the webUI:

from clearml import Task

task = Task.get_task(task_id="<TASK_ID>")

print(task.data.script.version_num)

This returned the commit ID I see in the webUI.

Are you sure there is a commit ID in the UI? Are you sure you're fetching the correct task?

I think you would need to contact the sales department for this 🙂
None

You're sure this is a generated token and not a password?

Can you please run ls -la /opt/clearml and send the output + your docker compose file

MammothParrot39 , yes it is available. This part of the Dataset module of clearml

Hi @<1856507252714770432:profile|VirtuousStork5> , what if you set the output_uri to point directly to the s3 bucket?

Hi GiganticMole91 , what version of ClearML server are you using?
Also, can you take a look inside the elastic container to see if there are any errors there?

Hmmm this is strange. Still not working for you?

You can provide it in the extra configurations sections

And are you running inside a repository or is it a stand alone script?

Yep, just make sure you show some activity in a task once every 2 hours so it won't be detected as inactive 🙂

CumbersomeParrot30 , try setting the following environment variable to true:
CLEARML_SKIP_PIP_VENV_INSTALL

The sample script you posted runs fine on server 1.6.0. I did however comment out from machine_learning.clearml_client import Task and used from clearml import Task

Can you please try with the regular import?

Hi FierceHamster54 , is this an old autoscaler instance? What is the version? You can see the version when you're on the application and click on 'More' at the top left text area

Fair point

No, resource configuration allows you higher level of control over different resources - None

I suggest you review the agent setup & deployment docs - None

Then you can view all the connected nodes through the orchestration tab - None

Yes, but this data is managed by mongodb. Also since you have full visibility at the user/passwords you can probably somehow generate a token similiar to how the UI does it when you log in to create a token

Can you try with auto_connect_streams=True ? Also, what version of clearml sdk are you using?