adoroszlai opened a new pull request, #4711:
URL: https://github.com/apache/ozone/pull/4711

   ## What changes were proposed in this pull request?
   
   
`TestDecommissionAndMaintenance#testContainerIsReplicatedWhenAllNodesGotoMaintenance`
 fails with the new replication manager (i.e. if legacy is disabled).  If all 
replicas are starting maintenance, underreplication is not fixed.  
`RatisReplicationCheckHandler` skips because there are no healthy replicas, and 
`RatisUnhealthyReplicationCheckHandler` skips because neither are there any 
unhealthy ones.  Decommissioning and maintenance replicas are counted 
separately, and we lose the information regarding their health.
   
   This change fixes the problem by counting unhealthy decom/maint replicas 
separately, and including them in total healthy/unhealthy counts 
(`getHealthyReplicaCount()` and `getUnhealthyReplicaCount()`).
   
   It also includes some refactoring as separate commits, reducing code 
duplication and duplicate calculation of some values.
   
   https://issues.apache.org/jira/browse/HDDS-8616
   
   ## How was this patch tested?
   
   New unit test is added to reproduce the problem.  
   
   Legacy replication manager in `TestDecommissionAndMaintenance` is disabled, 
since it now passes with the new one.
   
   CI:
   https://github.com/adoroszlai/hadoop-ozone/actions/runs/4972183233


-- 
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