DHCPD issues.

[Sten Carlsen]

On 12/01/12 0:15, ~Stack~ wrote:
> On 01/11/2012 08:22 AM, Glenn Satchell wrote:
>> You also seem to be missing a definition for the default gateway on each
>> subnet. This is needed so that the client can communicate with the dhcp
>> server. This also means that the range should *not* include this address.
>>
>> So assume that there is a router, or the dhcp server has a secondary IP
>> address for each subnet, then you need
>>
>> option routers 172.31.201.1;
>> range 172.31.201.2 172.31.201.254;
>>
>> and so on in each subnet.
>>
>> Classes operate at the pool level, so you need a pool for each range,
>> like this:
>>
>> shared subnet ...
>>   subnet ...
>>     pool {
>>       range ...
>>       allow members of "class_a"
>>     }
>>   }
>>   subnet ...
>>
>> regards,
>> -glenn
> Thanks for replying.
>
> I can not seem to get the pools to work. The only time I was able to get
> a non-error config file where the hosts all were pulling their proper
> range, they couldn't ping each other because dhcpd insists that the
> netmask be 255.255.255.0 otherwise it throws an error on startup. If I
> manually change the netmask on the systems to 255.255.0.0 it works but
> that isn't a solution.
>
> So I gave up on that. I have been digging through man pages and
> documentation and I can't seem to find an example of what I am trying to
> do. I am two days down on this and tired of finding lots of partial
> examples that don't work.  :-)
>
> Instead, I scrapped it all for the configuration posted below. This
> works well enough and knowing that the host declaration is global helps.
> Putting the PXE boot information into the group suites my needs for the
> moment. However, I now have the problem that for some reason it is _not_
> talking to the bind DNS system (on the same host). Nodes that PXE boot
> get the right IP yet they can not seem to find their name; they also
> can't ping anyone by DNS except for 172.31.200.2. This is very
> frustrating. Can anyone spot why dhcpd isn't talking to bind?
>
> Thanks again for the replys!
>
> ~Stack~
>
> ______________________________________________
>
> option domain-name "project.local";
> option domain-name-servers 172.31.200.2;
> authoritive;
> ddns-updates on;
You may want this as well:
/ddns-update-style xxx;/
> ddns-domainname "project.local";
> ddns-rev-domainname "in-addr.arpa.";
>
> shared-network project.local{
> 	option routers 172.31.0.1;
> 	default-lease-time 86400;
> 	max-lease-time 86400;
> 	deny unknown-clients;
> 	subnet 172.31.0.0 netmask 255.255.0.0 {}
>
> 	host dev001.project.local { hardware ethernet 08:00:27:11:4B:E8;
> fixed-address 172.31.201.1; }
>
> 	group {
> 		filename "pxelinux.0";
> 		next-server 172.31.200.2;
> 	        host aa001.project.local { hardware ethernet 08:00:27:15:BF:F5;
> fixed-address 172.31.202.1; }
> 	        host hp001.project.local { hardware ethernet 08:00:27:15:A3:1D;
> fixed-address 172.31.203.1; }
> 	}
> }
I wonder why you have a shared-network as there is only one subnet here?

I also want to mention that by placing the host statements inside your
shared-network, you will risk some very hard to locate issues with the
hosts inheriting from the apparently wrong places. The intention is to
place the host declarations in the global scope, i.e. among the options.
>
>
>
> _______________________________________________
> dhcp-users mailing list
> dhcp-users at lists.isc.org
> https://lists.isc.org/mailman/listinfo/dhcp-users

-- 
Best regards

Sten Carlsen

No improvements come from shouting:

       "MALE BOVINE MANURE!!!" 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/dhcp-users/attachments/20120112/06d6b4fa/attachment.html>