gianm commented on issue #5589:
URL: https://github.com/apache/druid/issues/5589#issuecomment-1029467029


   >  I'm seeing quite a lot of errors in the Historicals like the following 
and similar ones in the Middlemanagers (OoM on Direct buffer). The instances 
are identical, same heap, same everything than the ones with Java 8, but only 
Java 11 ones are throwing this error
   
   @tanisdlj Hmm, one idea about what's going on here. If you have *not* set a 
value for `druid.processing.buffer.sizeBytes`, then you'll see a log message on 
server startup like "Auto sizing buffers to [%,d] bytes each for [%,d] 
processing and [%,d] merge buffers". The calculation is different on Java 8 and 
11, because on Java 11 we can't get the max direct memory size directly, so we 
have to guess. Due to this missing info, the Java 11 version of the calculation 
is less accurate. So if you're hitting this case, you can fix it by setting 
`druid.processing.buffer.sizeBytes` explicitly.
   
   Another thing to check: is it possible that your two deployments have 
different values of `-XX:MaxDirectMemorySize`?


-- 
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