jihoonson commented on issue #6336: update insert pending segments logic to 
synchronous
URL: https://github.com/apache/incubator-druid/pull/6336#issuecomment-450693437
 
 
   I see. Since the lock contention can happen between tasks if they're writing 
into the interval of the same dataSource, 600 running tasks at the same time 
can cause this issue. If this is the case, before this PR, I think such many 
tasks could cause frequent pending segment allocation failures due to the 
allowed race in `allocatePendingSegments`. Those failed actions will be 
retried, but make segment allocation slow. Were you able to see this kind of 
problem before upgrade?
   
   Maybe we need to use separate http thread pools for API and UI. Also, maybe 
https://github.com/apache/incubator-druid/pull/6348 and 
https://github.com/apache/incubator-druid/pull/6356 would help though you said 
your pendingSegment table is small.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

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

Reply via email to