I hope this email finds you well. I have been successfully marking various TCP/UDP flows using rte_flow. However, I have noticed that there are frequent occurrences of checksum problems in the network. As a result, I would like to prevent flows with corrupted checksums (IP, TCP, UDP) that I marked with rte_flow from reaching the application layer. However, I still want to keep track of these flows and count the occurrences of IP, TCP, or UDP checksum issues on a per-flow basis.

 

Is there a way to achieve this? I would like to have visibility into the number of checksum problems experienced by each flow (IP, TCP, UDP) without allowing them to reach the application layer.

 

I am currently using DPDK-23.03 and my network card is Mellanox MT27710 Family [ConnectX-4]

 

Any guidance or suggestions you can provide on this matter would be greatly appreciated.

 

Thank you for your attention and support.

 

Best regards

 

- Volkan