nanamikon edited a comment on issue #4382: URL: https://github.com/apache/apisix/issues/4382#issuecomment-855622043
In my company, many users set 200 r/m on remote_address or even lower, If it divided by 60, 1r/s or 2r/s will be reached easily, because modern browser support http2, so one user visit one nignx over 1r/s is very common. Now, we transfer this conf to apisix, some of them reach limit and delay request online。 I think we use nodelay in nginx conf while apisix only support delay cause this happen， because both of them reach the rate limit, but not over burst. I will test it ``` The rate is specified in requests per second (r/s). If a rate of less than one request per second is desired, it is specified in request per minute (r/m). For example, half-request per second is 30r/m. ``` ``` ctx->rate = rate * 1000 / scale; ``` Maybe multiply by 1000 is the key to solve the decimal point ？ -- 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]
