albertobastos commented on PR #15264: URL: https://github.com/apache/pinot/pull/15264#issuecomment-2745198616
> We need to apply the same throttling as the `QueryLogger` does. This change can flood the log for super high QPS use cases (imagine over 10k query lines per second) I get the point, but two questions about it: - Would it be acceptable that this query logging become unreliable under high QPS scenarios? Once there's a chance for a query to not be logged and cannot be detected during throubleshooting, we miss the whole point of the PR. - Should we share the same throttling `QueryLogger` uses or add a separate one? I was even thinking on using `QueryLogger` itself, splitting the current `log` method into two: `logRequest` and `logResponse`. While the latter would remain the same, the first will only log the `requestId` and query string as intended. -- 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: commits-unsubscr...@pinot.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org For additional commands, e-mail: commits-h...@pinot.apache.org
