Hi Saneth, In a clustered environment, the cutoff might not exactly happen on 51st request. Certain amount of requests may slip through. The amount of additional requests passing through depends on several parameters like the number of GW nodes running on the setup, replication frequency set and the amount of requests hitting the GWs.
On a standalone environment, the cutoff should happen on 51st request. On Wed, Mar 18, 2015 at 6:50 PM, Saneth Dharmakeerthi <[email protected]> wrote: > Hi APIM Team, > > I need to clarify some information about "*throttle:MaximumCount*" > > If Tier "throttle:MaximumCount" is set to 20 and UnitTime is set to 60000 > (assume we apply bellow throttle value to all API-level, Application-level > and Resource-level ) > ex: > <throttle:MaximumCount>50</throttle:MaximumCount> > <throttle:UnitTime>60000</throttle:UnitTime> > > 1. whether the api invocation cut off exactly from 51th request > within 60000 milliseconds ? > 2. If not, is there any percentage or amount that it can accept > additional requests? > > Also if there is a document that has detail information about this, please > share with me. > > Thanks and Best Regards, > > Saneth Dharmakeerthi > Senior Software Engineer > WSO2, Inc. > Mobile: +94772325511 > -- *Amila De Silva* WSO2 Inc. mobile :(+94) 775119302
_______________________________________________ Dev mailing list [email protected] http://wso2.org/cgi-bin/mailman/listinfo/dev
