[GitHub] [ozone] sodonnel commented on pull request #3497: HDDS-6848. Ignore timeout replication tasks on datanode

[GitBox <gi...@apache.org>]

sodonnel commented on PR #3497:
URL: https://github.com/apache/ozone/pull/3497#issuecomment-1150969847

   There is some work going on to fix this sort of issue in different ways. #3482 is going to limit the number of replications that can be inflight at any given times, and hence should avoid deep queues on the DNs.
   
   As part of EC, we are working toward replacing replication manager completely, and when we do that, we will be holding back the replication work in SCM and feeding it to the datanodes when they have capacity to take it. The queues on the DNs should be very small when that happens.
   
   While the approach here looks OK, I feel we should hold off on committing it until we see how well #3482 works, and larger RM refactor. Once those are in, this change would not be needed I think.


-- 
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: issues-unsubscribe@ozone.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org