Hi Spark devs,

One follow-up while I am waiting for Jira permissions.

I have just subscribed to this list and saw the current discussion around
contributor workflow and PR review. Since I am new to the Spark community,
I would like to make sure I approach this in a useful way instead of just
opening a larger PR out of nowhere.

The reason I am interested in SPARK-47017 specifically is that it is very
close to the area I work on: query diagnostics, execution profiles, and
making runtime metrics easier to understand. Showing the physical plan and
metrics behind RDDScanExec's internal RDD in the History Server looks like
a practical observability improvement, not just a UI change.

My initial plan would be:

1. Reproduce the issue using the existing attachment / event log.
2. Trace where the metric definitions and metric values stop being
connected in the SQL UI / History Server path.
3. Start with a small focused PR, ideally with a regression test or minimal
internal groundwork, before attempting a broader UI change.

Eric Yang, I also noticed that there had been earlier activity around this
issue. If you are still planning to work on a PR for it, I would be happy
to coordinate and avoid duplicating your work.

Does this sound like a reasonable way to approach SPARK-47017? If there is
a better first scope, or someone familiar with this area who would prefer a
different direction, I would appreciate the guidance.

Thanks,
Aleksandr

сб, 30 мая 2026 г. в 11:44, Aleksandr Efimov <[email protected]>:

> Hi Spark devs,
>
> I would like to start contributing to Apache Spark and work on SPARK-47017.
>
> Could someone please grant me the Jira permissions needed to assign the
> issue to myself and update its status?
>
> My Apache Jira username is: alexandrefimov
>
> I plan to start with a focused change for SPARK-47017 and submit a PR for
> review once I have a clean implementation.
>
> Thanks,
> Aleksandr
>

Reply via email to