bharatviswa504 edited a comment on issue #536: HDDS-2988. Use 
getPropertiesByPrefix instead of regex in matching ratis client and server 
properties.
URL: https://github.com/apache/hadoop-ozone/pull/536#issuecomment-584742986
 
 
   > > The reason for server, client and grpc prefix is to group the 
configurations by the group
   > 
   > Thanks the answer. I understand that grouping is useful. I don't 
understand why do we need double grouping.
   > 
   > > Do we really need the server and client part here? Why don't we use 
`datanode.ratis.raft.client.*` and `datanode.ratis.raft.server.*` instead of 
`datanode.ratis.server.raft.server`? Is there any use case where we need to 
configure the client configs on the server side? 
(`datanode.ratis.server.raft.client.*` for example?)
   > 
   > If we already have a grouping on the ratis side why do we introduce an 
other one on the ozone side?
   
   The reasons for a prefix "datanode" is in future if OM also uses this 
similar approach, we need a way to distinguish which config is for which 
component. So, that is the reason for prefix. So, created 3 config classes, one 
for ratis server with datanode.ratis.server, similarly for others. I understand 
the ratis.server is duplicated again with this approach. If any other thoughts 
how to handle this?
   
    Main reasons for doing this way.
   1. Distinguish these properties for each component. (As OM also uses OM 
ratis Server, sooner or later, we might do similar work for OM also)
   
   One way I think we can remove this is.
   
    So, thinking more again, we can use  "datanode.raft.server" for RatisServer 
config and for client "datanode.raft.client" and for grpc "datanode.raft.grpc". 
In this way we can use directly ratis grouping. Is this is what you are 
suggesting here?
   

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


With regards,
Apache Git Services

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

Reply via email to