stoty commented on PR #6584:
URL: https://github.com/apache/hbase/pull/6584#issuecomment-2662231879

   > IIRC we only depend on curator in hbase-examples and now we do not ship 
hbase-examples with our binary tarball, so we do not need to manage the 
dependency for curator.
   
   Yes, I agree.
   I thought that I've removed the dependency management, but apparently I  
haven't.
   
   > For binaries without hadoop where we need to depend on the hadoop 
installation on the machine, I think curator 4 and 5 can both work with zk 
3.8.x? 
   
   That is my understanding, too.
   
   > And for developers, we use maven flatten plugin so there is no 
dependencyManagement section in our pom. so it will not mess up the end users' 
dependency management.
   
   I meant the case when we run HBase from the source repo, not the projects 
that depend on HBase.
   
   > 
   > This is my understanding. Correct me if I'm wrong.
   > 
   > Thanks.
   
   Since we don't use curator, this only really comes into play when some 
Hadoop code which uses Curator is being run from the HBase classpath, where 
have overridden ZK but not curator.
   
   TBH I don't know if that's a realistic scenario.
   
   I'm gonna open a ticket to remove the dependency management, and we'll see 
if the test suite shows any problems when using ZK 3.8 with the ancient Curator 
coming from Hadoop.
   


-- 
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