our Linux Server kernel log report lot of "udp bad checksum "
now i am not sure if that means the udp packet was drop by the kernel
and i am wondering what knid of think would cause the checksum error in data in transfering stage .

would Router/Switch change ip/tcp/udp checksum values ?

If only udp fails some checksum then I'd suspect some device/program/connection is limiting or can't pass traffic at the requested time.

UDP from wiki. " It has no handshaking dialogues, and thus exposes the user's program to any unreliability of the underlying network: there is no guarantee of delivery, ordering, or duplicate protection" https://en.wikipedia.org/wiki/User_Datagram_Protocol

Might have to really look into this sort of issue for clues if you depend on udp. Otherwise see if you can move to a trusted end to end system or program.


Side note. Edit first post over adding. Keeps your thread on zero reply list for more members to view easily.

but, what think would cause the checksum incorrect ?

If all of your checksums are wrong then I'd suspect that the nic card is offloading the checksum process and it then reports a fake checksum to things like wireshark.

"I'd suspect some device/program/connection"

"Might have to really look into this sort of issue for clues if you depend on udp."

You would need to start some logical process to half split the issue or swap or view logs or test or replicate.

NOT ALL OF CHECKSUM, it is in a little high rate .
do you thing data would be corrupt while transmitting in the Router or Switch
any update ?