[GitHub] [spark] HyukjinKwon edited a comment on pull request #30167: [SPARK-33240][SQL][3.0] Fail fast when fails to instantiate configured v2 session catalog

HyukjinKwon edited a comment on pull request #30167: URL: https://github.com/apache/spark/pull/30167#issuecomment-718336795 _In a way_, a user _could_ think it's all internal as long as the tables are created somewhere and they can be read regardless of where the tables are created. And,

[GitHub] [spark] HyukjinKwon edited a comment on pull request #30167: [SPARK-33240][SQL][3.0] Fail fast when fails to instantiate configured v2 session catalog

HyukjinKwon edited a comment on pull request #30167: URL: https://github.com/apache/spark/pull/30167#issuecomment-718295109 @HeartSaVioR, I fully agree that the newer behaviour makes much sense now. I also get that we can have many benefits by doing this fix. We have many fixes that

[GitHub] [spark] HyukjinKwon edited a comment on pull request #30167: [SPARK-33240][SQL][3.0] Fail fast when fails to instantiate configured v2 session catalog

HyukjinKwon edited a comment on pull request #30167: URL: https://github.com/apache/spark/pull/30167#issuecomment-717706506 Hm, is it something we should port back? I think it's hard to call it a bug. Maintenance release shouldn't have such behaviour changes in general according to semver