Fix queue handling in K8sIntegration and k8s_glue_example.py #183
Conversation
examples/k8s_glue_example.py
Outdated
| @@ -91,7 +95,10 @@ def k8s_user_props_cb(pod_number=0): | |||
| ssh_port_number=args.ssh_server_port) if args.ssh_server_port else None, | |||
| namespace=args.namespace, max_pods_limit=args.max_pods or None, | |||
| ) | |||
| k8s.k8s_daemon(args.queue, use_owner_token=args.use_owner_token) | |||
| if "," in args.queue: | |||
There was a problem hiding this comment.
This seems too elaborate, you can simply do:
args.queue = [q.strip() for q in args.queue.split(",") if q.strip()]Which would also handle input such as "queue1,"
| COPY ./clearml.conf /root/clearml.conf | ||
|
|
||
| #TODO: should be adjusted | ||
| RUN python3 -m pip uninstall clearml-agent -y && python3 -m pip install git+https://github.com/FeU-aKlos/clearml-agent.git@feature/support_multi_queeus_with_k8s |
There was a problem hiding this comment.
there is still the dockerimage linked, which i created my self, I was not sure how to handle the thing, that the clearml-agent python package should be installed, for usage in the helm chart, without the actual (official) repo having the changes. Any Ideas on that? Should i adjust the setup.sh script and reference my fork there for pip install or the main branch, if i expect, that this PR will be accepted...
There was a problem hiding this comment.
or in the entrypoint.sh
There was a problem hiding this comment.
Don't leave your repo here - once we merge the PR we'll build a new image and publish it
There was a problem hiding this comment.
Okay, great, thanks for clarification
Facilitate, that in the clearml-agent helm chart, multiple queues can be assigned with following format: 'queue1,queue2'.