olg33 <notifications@github.com> writes:

> Thanks Toke and Dave
>
> I have been testing the marking version intensively and as mentioned it's working really well. However I just found that when I put a sniffer and graph the marked TCP streams with Netdata, the Tx flow graph shows the marked data, but on the Rx side the traffic returns unmarked.
>
> In order to double-check I captured packets on both the Tx and Rx interfaces. On the Tx side I was able to see the marks:
>
>> Differentiated Services Field: 0x98 (DSCP: AF43, ECN: Not-ECT)
> Differentiated Services Field: 0x28 (DSCP: AF11, ECN: Not-ECT)
> Differentiated Services Field: 0x88 (DSCP: AF41, ECN: Not-ECT)
> Differentiated Services Field: 0x48 (DSCP: AF21, ECN: Not-ECT)
> Differentiated Services Field: 0x20 (DSCP: CS1, ECN: Not-ECT)
>
> But on Rx, all traffic has the "default" mark:
>
>> Differentiated Services Field: 0x00 (DSCP: CS0, ECN: Not-ECT)
> Differentiated Services Field: 0x00 (DSCP: CS0, ECN: Not-ECT)
> Differentiated Services Field: 0x00 (DSCP: CS0, ECN: Not-ECT)
> Differentiated Services Field: 0x00 (DSCP: CS0, ECN: Not-ECT)
> Differentiated Services Field: 0x00 (DSCP: CS0, ECN: Not-ECT)
>
> That's why I'd like to ask you, is this an expected outcome or should
> I review my setup?

Are you running these tests over the public internet? It's quite common
for DSCP fields to get zeroed out when traversing the internet...


-- 
You are receiving this because you commented.
Reply to this email directly or view it on GitHub:
https://github.com/tohojo/flent/issues/203#issuecomment-611205419