The customer notices that some flows are not respecting the maximum bandwidth for the policies these flows are falling under by a large amount. This article provides a resolution for this issue.
Most likely, the problem is that these flows are being subjected to Edge Cache or Acceleration. For either of these cases, the Inbound QoS is not respected and this is the reason why the numbers configured in the policies are not followed.
Follow one of the resolutions given below:
- In case the traffic is being accelerated (yellow striping in real time), it means that there is an Exinda on the other side of the connection terminating the acceleration process. Go to that other Exinda and configure the Outbound Policies in such a way that these flows are being limited to the desired bandwidth amount. Since the outbound flow on the other side represents the inbound flow on the local Exinda, this will solve the problem,
- In case the traffic is being subject to edge cache (blue striping in real time), we are aware of this design and we are planning on changing it in a future version of v7. We can create policies for specific groups of IPs (Network Object-based) to avoid that traffic from being edge cached but not from an application level. If this cannot be a solution, the temporary workaround would be to disable edge cache until the software design is modified in the future to be able to QoS Inbound Edge Cache traffic.