[k8s] Prototype - SkyServe on K8s #3109
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…o serve_k8s_playground # Conflicts: # sky/clouds/kubernetes.py # sky/serve/core.py
This was referenced Feb 20, 2024
…o serve_k8s_playground # Conflicts: # sky/cli.py
…o serve_k8s_playground
…t-org/skypilot into serve_k8s_playground # Conflicts: # sky/serve/core.py # sky/serve/replica_managers.py
Michaelvll
reviewed
Mar 12, 2024
There was a problem hiding this comment.
Thanks for adding the support sky serve on kubernetes @romilbhardwaj! Just did a quick pass and left several comments. : )
Comment on lines
+1963
to
+1976
| if endpoint: | ||
| cluster_endpoint = backend_utils.get_endpoints( | ||
| cluster_record['name'], endpoint) | ||
| click.echo(cluster_endpoint) | ||
| else: | ||
| cluster_endpoints = backend_utils.get_endpoints( | ||
| cluster_record['name']) | ||
| assert isinstance(cluster_endpoints, dict) | ||
| for port, port_endpoint in cluster_endpoints.items(): | ||
| click.echo( | ||
| f'{colorama.Fore.BLUE}{colorama.Style.BRIGHT}{port}' | ||
| f'{colorama.Style.RESET_ALL}: ' | ||
| f'{colorama.Fore.CYAN}{colorama.Style.BRIGHT}' | ||
| f'{port_endpoint}{colorama.Style.RESET_ALL}') |
There was a problem hiding this comment.
nit:
Suggested change
| if endpoint: | |
| cluster_endpoint = backend_utils.get_endpoints( | |
| cluster_record['name'], endpoint) | |
| click.echo(cluster_endpoint) | |
| else: | |
| cluster_endpoints = backend_utils.get_endpoints( | |
| cluster_record['name']) | |
| assert isinstance(cluster_endpoints, dict) | |
| for port, port_endpoint in cluster_endpoints.items(): | |
| click.echo( | |
| f'{colorama.Fore.BLUE}{colorama.Style.BRIGHT}{port}' | |
| f'{colorama.Style.RESET_ALL}: ' | |
| f'{colorama.Fore.CYAN}{colorama.Style.BRIGHT}' | |
| f'{port_endpoint}{colorama.Style.RESET_ALL}') | |
| cluster_endpoint = backend_utils.get_endpoints(cluster_record['name'], endpoint) | |
| if endpoint: | |
| click.echo(cluster_endpoint) | |
| return | |
| for port, port_endpoint in cluster_endpoints.items(): | |
| click.echo( | |
| f'{colorama.Fore.BLUE}{colorama.Style.BRIGHT}{port}' | |
| f'{colorama.Style.RESET_ALL}: ' | |
| f'{colorama.Fore.CYAN}{colorama.Style.BRIGHT}' | |
| f'{port_endpoint}{colorama.Style.RESET_ALL}') |
There was a problem hiding this comment.
backend_utils.get_endpoints() is invoked differently in the two calls - one is with a specific endpoint, which returns a str, and the other is endpoint=None, which returns a dict.
| @@ -23,6 +23,7 @@ | |||
| from sky import exceptions | |||
| from sky import global_user_state | |||
| from sky import status_lib | |||
| from sky.backends import backend_utils | |||
There was a problem hiding this comment.
Importing backend_utils here may caues circular import. We should be careful about this.
romilbhardwaj
changed the title
8 tasks
|
Since #3377 was merged, closing this PR for now. : ) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Working prototype of SkyServe on k8s. Both the controller and the replicas are run on k8s cluster.
TODOs
Automatically invoke- no longer required with SA supportgenerate_static_kubeconfig.shfor exec based kubeconfigssky-keyand instead use the solution from [k8s] Share SSH jump pod #2826 to allow multiple keys to auth to ssh jump podquery_endpointsimplementation.Tested (run the relevant ones):
bash format.shpytest tests/test_smoke.py