JingsongLi commented on PR #10182:
URL: https://github.com/apache/paimon/pull/10182#issuecomment-5831333345

   Reviewed head `943499b054`. Requirement fit: SUPPORTED. Implementation: 
CLEAN.
   
   Issue #10181 identifies a reachable Hive clone failure: creating a 
HiveCatalog with an `s3://` warehouse initializes the S3 FileIO, then a source 
table at `s3a://` in the same bucket uses that FileIO and triggers Hadoop’s 
`Wrong FS` check. The change keys the per-instance cache by both scheme and 
authority, matching the existing core HadoopFileIO behavior; S3FileIO’s shared 
cache already distinguishes the two. I traced the normal and reverse scheme 
orders, same-scheme reuse, and the relevant caller path. No further code 
finding survived review.
   
   Verification on JDK 8 / Maven 3.9.9: `mvn -pl 
paimon-filesystems/paimon-s3-impl -am -Pfast-build -DwildcardSuites=none 
-Dtest=S3FileIOSchemeCacheTest -DfailIfNoTests=false test` passed, 3 tests, 0 
failures; `git diff --check HEAD^..HEAD` passed. The CI JDK 8 and JDK 11 jobs 
failed because `S3FileIOTest` could not pull the pre-existing MinIO image from 
quay.io, before exercising the S3 integration path. Please rerun or repair that 
CI dependency before merge. I did not run a live S3/Hive clone integration test.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to