akshat0395 commented on PR #4012: URL: https://github.com/apache/hive/pull/4012#issuecomment-1415105579
> @akshat0395, **_one question_**: Shouldn't we need to be in-sync with Hadoop (3.3.1 used in hive) on avro version even though we are already ahead. Because difference in file format version might cause problem. Because I think the hadoop jars will be loaded first in classpath which will already contain hadoop avro version (1.7.7)? > > There is some discussion on same for this on PR 3878 ([HIVE-26829](https://issues.apache.org/jira/browse/HIVE-26829)). Thanks @Aggarwal-Raghav for sharing these details, I wasn't aware that there is a similar discussion going on. You made some valid points, I think the solution to which could be extensive testing. Right now I've run some similar tests that you have done in your upgrade, which seems to work fine to me, and after going to the discussion I feel writing some qtest and some unit test around avro APIs that are used in Hive might help confirming if the upgrade is safe or not, Till then this can be marked in progress. Regarding compatibility with Hadoop, as we are already ahead of Hadoop and the changes seems to be backward compatible I dont think upgrade should be a problem there. Also I do agree with @cnauroth regrading the upgrade to 1.11.1 rather than 1.11.0. @Aggarwal-Raghav @cnauroth Please let me know what do you think about the proposed testing solution. -- 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: gitbox-unsubscr...@hive.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: gitbox-unsubscr...@hive.apache.org For additional commands, e-mail: gitbox-h...@hive.apache.org
