Reply to remote unicast DHCP requests

Paúl Ortiz Imedio portizimedio at yahoo.es
Tue May 3 15:23:24 UTC 2011


Simon Hobson wrote:
>In that case, your setup is operating correctly - it is your 
>expectations that are wrong !
>You could have saved a lot of effort by asking earlier, and 
>mentioning the authoritative/DHCPNAK element to the story.
 
>If setting the server as authoritative results in it DHCPNAKing
>the request, then the requested address is not valid according 
>to the DHCP server setup. Setting the server to 
>non-authoritative results in it simply ignoring the invalid 
>request - by design.
 
>So you need to determine why the server believes the requested 
>address to be invalid.
 
I know that is what I need to solve, but I cannot guess what is
wrong because my scenario is quite simple:
   
CLIENT -    net1    - RELAY -    net2    - SERVER
 
The server configuration only defines the remote subnet (net1). 
It has no subnets defined for DHCP at local interfaces (net2).
I looked for an explanation at the code and then I found what I
reported in my first post: the request is being discarded 
because the server receives the packet through a local 
interface (connected to net2) with no shared network associated.
 
The content of the DHCPREQUEST is:
- ciaddr = IP of the valid lease
- yiaddr = siaddr = giaddr = 0
- client HW address
- options: 53 and 55
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/dhcp-users/attachments/20110503/778ddce8/attachment.html>


More information about the dhcp-users mailing list