[ 
https://issues.apache.org/jira/browse/HBASE-18946?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16280011#comment-16280011
 ] 

Anoop Sam John commented on HBASE-18946:
----------------------------------------

Ram.. What I say is at 1st steps, there wont be 3 regions at all (consider 3 
replica count and we say abt one region)..  Till the LB is been contacted. 
There only the replicas are been generated and asked for servers.  Now this 
single region processing itself is a sub procedure.  Now we have 3 regions 
under this and so have to generate 3 sub procedures.  My Q was whether that is 
possible. A sub procedure under another (in the f/w level) and Stack says yes.. 
The issue of grouping comes as we did the replica regions create at 1st.  What 
comes always to my mind is can we NOT do that then but later.  May be am 
totally wrong as I did not do code study fully. Just putting up an 
option/Question

> Stochastic load balancer assigns replica regions to the same RS
> ---------------------------------------------------------------
>
>                 Key: HBASE-18946
>                 URL: https://issues.apache.org/jira/browse/HBASE-18946
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 2.0.0-alpha-3
>            Reporter: ramkrishna.s.vasudevan
>            Assignee: ramkrishna.s.vasudevan
>             Fix For: 2.0.0-beta-1
>
>         Attachments: HBASE-18946.patch, HBASE-18946.patch, 
> HBASE-18946_2.patch, HBASE-18946_2.patch, HBASE-18946_simple_7.patch, 
> HBASE-18946_simple_8.patch, TestRegionReplicasWithRestartScenarios.java
>
>
> Trying out region replica and its assignment I can see that some times the 
> default LB Stocahstic load balancer assigns replica regions to the same RS. 
> This happens when we have 3 RS checked in and we have a table with 3 
> replicas. When a RS goes down then the replicas being assigned to same RS is 
> acceptable but the case when we have enough RS to assign this behaviour is 
> undesirable and does not solve the purpose of replicas. 
> [~huaxiang] and [~enis]. 



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Reply via email to