Hi Youngrae, +1 on going with an allow-list, same way we do it for ICEBERG_REST. ill take a look at the PR
Thanks, *Vignesh* On Wed, 14 Oct 2026 at 22:25, youngrae kim <[email protected]> wrote: > Hi all, > > I would appreciate a review of https://github.com/apache/polaris/pull/5680 > . > > It is the follow-up deferred from #5644: connectionConfigInfo.properties is > stored and returned for HADOOP and HIVE connections but never applied when > the federated catalog is initialized. The PR forwards an explicit > allow-list of keys, following the ICEBERG_REST precedent, and documents it. > The design question (which keys, allow-list vs. merge-all) is in #5676 and > the list in the PR is a proposal I am happy to adjust. > > CI is green and the branch is rebased on main. > > Thanks, > Youngrae Kim >
