kubeflow/sdk

KEP-107: Improve Spark E2E Robustness

Open

#611 opened on Jul 19, 2026

View on GitHubContributor guide
 (7 comments) (0 reactions) (1 assignee)Python (196 forks)auto 404
area/sparkarea/testinggood first issuehelp wantedkind/feature

Repository metrics

Stars
 (124 stars)
PR merge metrics
 (PR metrics pending)

Description

What you would like to be added?

The goal of this issue is to make the Spark E2E tests more robust and align closely with upstream artifacts to enable better detection of drifts.

The following is a consolidated list of target end states:

  • Remove the vendored SparkConnect CRD and pull all the latest upstream artifacts to test Spark SDK against the latest version.
  • Remove the default service account from the RBAC.
  • Enable "cluster watcher" support for both SparkConnect and SparkApplication.

Why is this needed?

Test robustness ensures the Spark package works in realistic production settings.

Love this feature?

Give it a 👍 We prioritize the features with most 👍

Contributor guide

Research direction
Focus on pulling upstream Spark artifacts to replace the vendored CRD, then remove the default service account and enable cluster watcher support for both SparkConnect and SparkApplication.
Tech stack
pythonsparkkubernetes
Domain
backendtestinginfrastructure
Issue type
Test
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
PythonKubernetesSpark
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60