You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@ozone.apache.org by GitBox <gi...@apache.org> on 2020/07/31 08:01:41 UTC

[GitHub] [hadoop-ozone] bshashikant edited a comment on pull request #1154: HDDS-3867. Extend the chunkinfo tool to display information from all nodes in the pipeline.

bshashikant edited a comment on pull request #1154:
URL: https://github.com/apache/hadoop-ozone/pull/1154#issuecomment-666245556


   > Thanks to create this patch @sadanand48
   > 
   > I have one design concern related to this change. I am not sure what is the good answer, but trying to share my view:
   > 
   > This patch modifies the `client` interface to provide more functionality for a `debug` tool. I like this more functionality, but I think we shouldn't add any debug specific feature to the main client API. Client API should be backward compatible between releases and it might have multiple implementations.
   > 
   > I would prefer to separate the `client` api from the `admin` api (used by this tool).
   > 
   > As an example: Let's say I would like to implement a new client for Erasure Coding. If the client interface contains debug specific methods I need to implement all of them, which makes introduce any new feature harder.
   > 
   > I am not sure how is it possible to keep this separation (I need your feedback). One possible solution is ot move the logic of looping over the datanodes to the `ChunkKeyHandler`. But it's not clear how can it work with RATIS pipeline. Or do you use STANDALONE client to read the data which is persisted with RATIS?
   
   I think we should remove XceiverClient interface altogether. Read and write path client functionality seem to diverge a lot .
   Adding a new functionality in XceiverClientGrpc will just be fine then.
   
   There is alredy a jira filed for the same https://issues.apache.org/jira/browse/HDDS-998.


----------------------------------------------------------------
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:
users@infra.apache.org



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