New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
HDDS-2120. Remove hadoop classes from ozonefs-current jar #1434
Conversation
💔 -1 overall
This message was automatically generated. |
Thanks for the quick fix @elek. Should we also exclude Ratis.. since HBase potentially has a dependency on Ratis and we should use the in-build version instead. What about third party dependencies? Downstream components could run into conflicts.. I believe that will require shading. |
We need Ratis on the classpath. I am not sure what is the best approach for support HBase. If we exclude it, we need to put two jar files to the classpath of hive/spark: ozonefs and ratis. Do you know which HBase version uses Ratis?
Yes, they can be shaded in HDDS-2121 (at least half of them. Some of them should be kept in original form such as protobuf or the logging). |
/retest |
LGTM +1 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
+1
/retest |
It looks like this failed compilation in Jenkins... does the Jenkins job need to be updated to use the separated pom compilation command? |
We have two kind of ozone file system jars: current and legacy. current is designed to work only with exactly the same hadoop version which is used for compilation (3.2 as of now).
But as of now the hadoop classes are included in the current jar which is not necessary as the jar is expected to be used in an environment where the hadoop classes (exactly the same hadoop classes) are already there. They can be excluded.
See: https://issues.apache.org/jira/browse/HDDS-2120