Dear sir,

You mean below configuration can make me to request dhcp request from 192.168.1.1 first, once 192.168.1.1 server is down, it will drop to the second one

172.16.1.1 ?

My configuration is

int vlan 100

ip address 100.100.100.1 255.255.255.0

ip helper-address 192.168.1.1

ip helper-address 172.16.1.1

Then if I change to

int vlan100

ip address 100.100.100.1 255.255.255.0

ip helper-address 172.16.1.1

ip helper-address 192.168.1.1

that means first query dhcp server 172.16.1.1 first, once 172.16.1.1 is down, it will query to 192.168.1.1 ?

Any document has publish  it ?

Thanks!

Unfortunately Dipesh is incorrect when he says that you can specify a DHCP server to querry first. If you configure 2 ip helper address commands then IOS will forward requests to both of them. It is likely that both will respond and in that case the client usually accepts the first one that responds. There is not a way in IOS to achieve what you want, that the request will go to a primary and only if that one is down will go to a secondary.

HTH

Rick

Dear sir,

You mean there is no way to control what is first and what is second, Client receive IP address is base on which dhcp respond is faster. Am I right ?

Yes you are right. With ip helper-address there is not a way to establish priority.

HTH

Rick

Dear,

I am not saying the priority of helper address.

But the reachability to DHCP server.

Suppose One DHCP server is in other segment and one DHCP server in the sane local segment than the dhcp server in the local segment will reponse first.

for 172.X.X.1 is other side of router so it will be reached second. 1st will use the dhcp server on local segment and if it will down than go for second one.

Regards,

Reachability could be checked by SLA. May be it is possible to do anything this way?

Clearly what the original poster wanted to achieve is to have 2 servers configured in a way that the router would send requests to only one server. If 192.168.1.1 is available the request is sent to this and if that server is no longer available then the request is sent to 172.16.1.1. How would you suggest that routing metrics could achieve this?

This question is very old and technology has changed. If I were answering that question today I might suggest that you might be able to combine tracking availability of 192.168.1.1 and an EEM script that could detect loss of connectivity to 192.168.1.1 and would then change the configuration to use 172.16.1.1. (and probably another EEM script to detect when 192.168.1.1 becomes available and revert to the original server)