1. Two DHCP servers:
First: Description: Debian GNU/Linux 6.0.10 (squeeze) Release: 6.0.10 Codename: squeeze
Second NAME="Debian GNU/Linux" VERSION_ID="8" VERSION="8 (jessie)" ID=debian
2. One network, let's say 192.168.0.0/24
3. Two interfaces 192.168.0.1 and 192.168.0.2, one per each DHCP server. Both are working and offering addresses from one subnet, but pool for first dhcp is .10-.20 gateway ip.1, pool for second ip.21-.31 with gateway ip.2
4. Task - devices in that network should receive addresses from DHCP #2 rather than DHCP #1. Why so? Main task is to migrate from one system to another, so i can't simply disconnect all devices - those of them, that are working now, must stay in working state. It's TV migration in Hotel.
5. What i've tryed? My thought was to manually increase delay in response of DCHP #1 on DHCP request from client.
6. Commands used on DHCP #1: tc qdisc add dev eth1 root handle 1: prio tc qdisc add dev eth1 parent 1:3 handle 30: netem delay 500ms tc filter add dev eth1 protocol ip parent 1:0 prio 3 u32 \ match ip sport 67 0xffff flowid 1:3 tc filter add dev eth1 protocol ip parent 1:0 prio 3 u32 \ match ip dport 68 0xffff flowid 1:3
7. Result: pinging DHCP #1 from device, i see delays in 500+ ms, but when i perform rebute of device and issueing "tcpdump -n -i eth1 portrange 67-68" command on both DHCP servers, or command "tail -f /var/log/syslog" for more details, i see, that DHCP #1 replies faster, than DHCP #2.
I will be very gratefull for ideas and clues, what to do else, or another solution. Thank you!