Skip to content

Evaluate Spark 4.2.0 upgrade (DBR 17.3/18/19 runtime compatibility) #100

Description

@mjohns-databricks

Dependabot PR #92 proposes bumping spark.version 4.0.0 -> 4.2.0. Held — deliberate evaluation, not an auto-merge.

spark.version is pinned to 4.0.0 across all four Maven profiles so a single JAR runs on DBR 17.3 / 18 / 19 (which ship Spark 4.0.x). Building against 4.2.0 risks linking APIs the DBR-shipped Spark doesn't provide (NoSuchMethodError / ClassNotFoundException on-cluster) and jeopardises the 'one JAR on 17.3/18/19' invariant.

Acceptance criteria:

  • Confirm the Spark version each supported DBR (17.3/18/19) ships; move only to a version all provide (or drop the ones that don't).
  • Build + run the Scala/parity suites against the target Spark; validate the JAR on each DBR.
  • Bump spark.version in all four pom profiles together.

Tracked from #92.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    dependenciesPull requests that update a dependency file

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions