amahussein commented on pull request #2737:
URL: https://github.com/apache/hadoop/pull/2737#issuecomment-822024808


   The discussion on PR #2878 and HDFS-15957 sounds to be around the same 
concept.
   
   By definition, if there is a "network issue", the NameNode in general 
because all RPCs will be problematic. So what makes that  `call.sendResponse(); 
call get a special treatment compared RPC in the NameNode?
   I do not see a bug.
   
   - Network issues are not fixed by delegating the notifications to a pool of 
thread. This hides the symptoms of the problems and makes it way difficult to 
the detect the original problem.
   - This code is potentially adding a major overhead because of the **lambda 
expression** that implies triggering a sequence of object allocations inside 
each iteration. "More allocations" lead to: slow executions,  means more GC, 
more frequent Full-GC, ..etc.
   
   -1 on this patch.


-- 
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.

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



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to