Hi I am using tc with htb + cfq to do some outbound traffic shaping, I have a high priority band which I want to put all my voip traffic. Currently I mark with 0x1 all traffic doing to the main sip server, but what I have noticed is that is not the destination of the rtp traffic - which is usually udp based - some where in the 10000 - 20000 range. My question is does the nf_conntrack_sip follow the initial tcp 5060 sip setup and associated the udp traffic with it and if so can I some how use the related conntrack to mark the udp packets ? i have something like iptables -I POSTROUTING -d 125.213.160.81 -j MARK --set-mark 0x1 if i changed it to iptables -I POSTROUTING -j CONNMARK --restore-mark iptables -I POSTROUTING -d 125.213.160.81 -j CONNMARK --set-mark 0x1 so is the connmark setmark the same as mark setmark - used by tc and will nf_conntrack_sip understand its a related packet even though it is to a different server. Thanks Alex Samad PS please CC me as I am not on the mailing list
Attachment:
signature.asc
Description: Digital signature
