apurtell commented on PR #4924: URL: https://github.com/apache/hbase/pull/4924#issuecomment-1371249713
> I do not think adding WAL attributes to mutation directly is the correct way to deal with this. For what it's worth, I think it would be fine, but I see your point. Consider what we are designing end to end. On the client side, the client creates a mutation with an attribute containing some metadata. On the server side, a coprocessor extracts the metadata and decorates the associated WAL entry via an attribute. Then the WAL entry is replicated. On the sink side, the WAL entry is processed and the metadata should be recovered from the attribute. The question is how to do that last part. I think introducing a default conversion from WAL entry attribute to mutation attribute would sufficient, but to @Apache9 's point, a symmetric flow here would have a pair of coprocessors involved: the first transforming attribute to metadata at the source client, the latter transforming metadata to attribute at the sink client. What I would not like to see here is a new config. -- 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]
