Skip to content

Conversation

@yangyuxia
Copy link
Contributor

@yangyuxia yangyuxia commented Nov 13, 2025

What changes were proposed in this pull request?

Overriding the invalidateTable() method in BaseCatalog

Why are the changes needed?

After a Spark session started using spark-connecto plugin, and then a table is modified via a Hive client, Spark's refresh table statement fails to clear the table's fileStatusCache,making it impossible to query the latest data from the table.

Fix: #9110

Does this PR introduce any user-facing change?

N/A.

How was this patch tested?

ITs

@jerqi
Copy link
Contributor

jerqi commented Nov 13, 2025

Do the other catalogs have similar issues? Could you add some UT for this fix?

@yangyuxia
Copy link
Contributor Author

So far, I've only used Hive catalog, and this pr depends on the fileStatusCache lifecycle management of kyuubi-hive-connector. apache/kyuubi#7191

@FANNG1
Copy link
Contributor

FANNG1 commented Dec 1, 2025

@yangyuxia , would you like to move the fix to BaseCatalog? since it's a general problem for all catalogs.

@FANNG1 FANNG1 changed the title clean cache when execute refresh table sql [#9110] fix(spark): clean cache when execute refresh table sql Dec 2, 2025
@FANNG1 FANNG1 merged commit 80c9bd7 into apache:main Dec 2, 2025
24 checks passed
@FANNG1
Copy link
Contributor

FANNG1 commented Dec 2, 2025

merged to main, @yangyuxia thanks for the contribution!

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[Bug report] The fileStatusCache cannot be cleared when executing a refresh table SQL statement using spark-connector

3 participants