siddharthteotia commented on pull request #6066:
URL: https://github.com/apache/incubator-pinot/pull/6066#issuecomment-700331216


   We don't know how many customers are receiving empty response today due to 
usage of incorrect column name.
   
   - We have quite a few tables already using SQL in production. Not sure if 
they have bad column names. But if they do, they will see errorCount go up.
   - For the ones that are currently migrating to SQL, it is expected of them 
to test out the code changes in test env first. However, they may not have 
queries with incorrect column name in their test environment and will notice 
the issue only in production. We can add to our migration guidelines and ask 
users to ensure columns names are in schema.
   
   While (2) is potentially fixable by changing the migration guidelines, I 
don't know how to handle (1). If there are any such users, they will have to go 
back, figure out where the incorrect column name is coming from, make changes 
and re-deploy. It is also possible that if there are many such cases and the 
errorRate is very high, we will end up rolling-back the release this change is 
part of.
   
   May be for now consider emitting a metric and a week or two after this code 
is deployed, we can ask users to fix their queries and then make this default? 
@jackjlli 
   


----------------------------------------------------------------
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:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to