simbadzina commented on PR #6404:
URL: https://github.com/apache/hadoop/pull/6404#issuecomment-1876005260

   @LiuGuH can you expand more on the following
   > And [HDFS-16890](https://issues.apache.org/jira/browse/HDFS-16890) maybe 
lead to many read requests into active NN at the same time.
   
   I see how multiple concurrent calls can read the same `false` value from 
`isNamespaceStateIdFresh(nsId)` before the accumulator is updated, but the 
windows for these reads should be very small. I'm wondering if we can instead 
fix the existing mechanism such that only a single read is sent to the active, 
vs. adding a new mechanism.
   
   Additionally, the periodic redirection of calls to the active only happens 
in the case when there are no calls going to the active already so having some 
reads be sent to the active should not overload it.


-- 
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]


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

Reply via email to