Skip to content

Conversation

ntkathole
Copy link
Member

What this PR does / why we need it:

This PR adds codeflare-sdk to requirements for managing ray. Codeflare-sdk requires python 3.11+ and kubernetes > 27.2.0.

Which issue(s) this PR fixes:

#5633

@ntkathole ntkathole self-assigned this Sep 27, 2025
@ntkathole ntkathole requested a review from a team as a code owner September 27, 2025 14:56
@franciscojavierarceo
Copy link
Member

I thought the codeflare SDK was being dropped

@ntkathole
Copy link
Member Author

I thought the codeflare SDK was being dropped

It is due to be renamed, but it's not going away.

Copy link
Member

@franciscojavierarceo franciscojavierarceo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm

@franciscojavierarceo
Copy link
Member

unit tests failing

@HaoXuAI
Copy link
Collaborator

HaoXuAI commented Sep 27, 2025

Hey is this necessary dependency?

@franciscojavierarceo
Copy link
Member

@ntkathole can we outline the benefits of having the codeflare SDK here?

@HaoXuAI
Copy link
Collaborator

HaoXuAI commented Sep 28, 2025

Or maybe create another dep like "Ray-clodeflakre"

@ntkathole
Copy link
Member Author

Hey is this necessary dependency?

Codeflare-sdk is required for handling authentication, certs and making connection to Kuberay. I am going to add support for this in upcoming PRs by adding codeflare wrapper above current implementation. This will be completely optional as a end user, User can use simply use ray address to connect to remote cluster as well, if ray cluster do not need authentication or tls certs.

can we outline the benefits of having the codeflare SDK here?

Codeflare-sdk allows us:

  • To find existing KubeRay cluster and connect without need to perform manual steps such as setting up tls certs configuration.
from codeflare_sdk import TokenAuthentication
from codeflare_sdk import generate_cert, get_cluster
  • User don't need to know cluster details, just name and namespace of Ray cluster.
  • Eliminates need to manually deploy and configure Ray clusters in Kubernetes.

I will add this in our documentation once implementation is done.

Or maybe create another dep like "Ray-clodeflakre"

We can do this, but it could be redundant. I am thinking anyhow codeflare-sdk will pull ray as it's dependency and usually ray needs same version for both server and client. If we did it like this, we don't need to maintain the ray version and let codeflare-sdk handle the compatibility.

@franciscojavierarceo
Copy link
Member

That makes sense to me.

Please make sure to document the benefits on the usage to the users.

@HaoXuAI codeflare and the auth @ntkathole mentioned is how we recommend users securely use Ray clusters at Red Hat. See here: https://developers.redhat.com/articles/2025/07/24/submit-remote-rayjobs-ray-cluster-codeflare-sdk

@HaoXuAI
Copy link
Collaborator

HaoXuAI commented Sep 28, 2025

That makes sense to me.

Please make sure to document the benefits on the usage to the users.

@HaoXuAI codeflare and the auth @ntkathole mentioned is how we recommend users securely use Ray clusters at Red Hat. See here: https://developers.redhat.com/articles/2025/07/24/submit-remote-rayjobs-ray-cluster-codeflare-sdk

I don't have double on the benefits. Just concerned that the codeflare is not available in our team's artifacts. Though we don't use Ray at the moment but possibly in the future

@franciscojavierarceo
Copy link
Member

franciscojavierarceo commented Sep 28, 2025

@HaoXuAI it's also open source so can be installed. It was a projected started by IBM I believe.

@franciscojavierarceo franciscojavierarceo merged commit 51a0ee6 into feast-dev:master Sep 30, 2025
14 of 17 checks passed
franciscojavierarceo pushed a commit that referenced this pull request Sep 30, 2025
# [0.54.0](v0.53.0...v0.54.0) (2025-09-30)

### Bug Fixes

* Column quoting in query of `PostgreSQLOfflineStore.pull_all_from_table_or_query` ([#5621](#5621)) ([e8eae71](e8eae71))
* Correct column list polars materialization engine ([#5595](#5595)) ([39aeb0c](39aeb0c))
* Fix Go feature server entitykey serialization for version 3 ([#5622](#5622)) ([5ab18a6](5ab18a6))
* Fix hostname resolution for spark tests ([#5610](#5610)) ([8f0e22d](8f0e22d))
* Fixed filtering based on data_source for ODFVs ([#5593](#5593)) ([c3e6c56](c3e6c56))
* Fixed project_description to set in registry and UI ([#5602](#5602)) ([02c3006](02c3006))
* Fixed Registry Cache Refresh Issues ([#5604](#5604)) ([3c7a022](3c7a022))
* Fixed tls issue when running both grpc and rest servers ([#5617](#5617)) ([51c16b1](51c16b1))
* Fixed transaction handling with SQLite registry ([#5588](#5588)) ([0052754](0052754))
* Update the deprecated functions in Go feature server. ([#5632](#5632)) ([a24e06e](a24e06e))
* Updated python packages conflicting with kserve dependencies ([#5580](#5580)) ([d56baf4](d56baf4))

### Features

* Add 'featureView' in global search api result for features. ([#5626](#5626)) ([76590bf](76590bf))
* Add aggregation in OnDemandFeatureView ([#5629](#5629)) ([8715ae8](8715ae8))
* Added codeflare-sdk to requirements ([#5640](#5640)) ([51a0ee6](51a0ee6))
* Added RemoteDatasetProxy that executes Ray Data operations remotely ([7128024](7128024))
* Added support for image search ([#5577](#5577)) ([56c5910](56c5910))
* Enable ingestion without event timestamp ([#5625](#5625)) ([eb51f00](eb51f00))
* Feast dataframe phase1 ([#5611](#5611)) ([2ce4198](2ce4198))
* Feast dataframe phase2 ([#5612](#5612)) ([1d08786](1d08786))
* Feast Namespaces registry for client ConfigMaps availability ([#5599](#5599)) ([728589a](728589a))
* Support hdfs:// uris in to_remote_storage for Spark offline store ([#5635](#5635)) ([5e4b9fd](5e4b9fd))
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants