using option 82 to assign a static ip-address

[Sten Carlsen]

>From original mail:

binary-to-ascii(16, 8, ":", substring(option agent.remote-id, 2, 7)) =
"84:78:ac:84:e5:80"

Reformat to:

binary-to-ascii(
16,                                          ---------> result is in hex
 8,                                           ---------> it works in
8bit wide increments
 ":",                                          ---------> separator is a ":"
 substring(
option agent.remote-id,          ---------> assumed to be a MAC address
-> string of 6 bytes: 0x112233445566
 2,
 7
)                                              ---------> 0x1122
->33445566<- This is the result of the substring operation.
)
                                               ---------> result of
binary-to-ascii should then be: "33:44:55:66" as a text string, 11 chars
long.

 = "84:78:ac:84:e5:80"

Something here is mysterious, not quite sure what. Can't see why this
has ever been a match?


On 17/12/12 1:03, Glenn Satchell wrote:
> Hi Nico
>
> Please see comments below.
>
> On Mon, December 17, 2012 12:17 am, Nico De Ranter wrote:
>> Hi,
>>
>>
>>
>> I'm trying to use DHCP option 82 to assign a different address to a host
>> depending on which network port it is connected to.
>> I'vve created the following section in my config:
>>
>> ############
>> stash-agent-options true;
>> ...
>> subnet 10.103.0.0 netmask 255.255.255.0
>> {
>>         option routers 10.103.0.1;
>>         class "OK1Endo" {
>>                 match if  binary-to-ascii(16, 8, ":", substring(option
>> agent.remote-id, 2, 7)) = "84:78:ac:84:e5:80"
> This string "8 4 : 7 8 : ... : 8 0" is 17 characters long. So a 7 char
> substring will never be equal to it. I can't see how it ever matches.
Precedence: first substring, then binary-to-ascii, I still don't see the
match.
>
>>                         and     binary-to-ascii (16, 8, ":", hardware) =
>> "1:0:14:2d:40:f:15"
>>                         and     binary-to-ascii (16, 8, ":",
>> substring(option agent.circuit-id, 5, 5)) = "e";
> This may or may not match. If the agent.circuit-id is exactly 5 characters
> long it will return that last char only and if that is an 'e' then it will
> match.
>
> These can all be re-written without the binary-to-ascii and be much clearer:
>
> match if substring(option agent.remote-id, 2, 7)) = 84:78:ac:84:e5:80
>      and hardware = 1:0:14:2d:40:f:15
>      and substring(option agent.circuit-id, 5, 1)) = e;
>
> Finally, putting the class definition inside the subnet can cause some
> unexpected inheritance. Probably ok if you only have a single subnet in
> your definition, but better to put the class outside the subnet
> definition.
>
>>         }
>>         pool {
>>                 filename "pxelinux.0";
>>                 next-server 10.103.0.202;
>>                 option mobile-ip-home-agent 10.103.0.202;
>>                 option log-servers 10.103.0.202;
>>                 option tz "Europe/Brussels";
>>                 use-host-decl-names on;
>>                 allow members of "OK1Endo";
>>                 range 10.103.0.220 10.103.0.220;
>>         }
>>   }
>> ###########
>>
>> I'm sure the class definition is correct and the pool matches as I also
>> added a log statement which only seems to trigger if the client matches
>> the
>> pool (if I change the class definition or remove the 'allow members' I
>> don't see the logs appearing in the log files anymore).
>>
>> However the dhcpd server doesn't seem to send any reply to the host: in
>> tcpdump I see a bootp/dhcp request from the clients MAC but no replies
>> from
>> the server.  If I replace the 'pool' section with a simple 'host'
>> definition the server responds immediately.
>>
>> Any ideas what I'm doing wrong?
>>
>> Note: I'm using a very old version of dhcpd (3.1.1) which unfortunately I
>> cannot upgrade at the moment.
>>
>> Thanks in advance,
>>
>> Nico
>>
>> --
>> Nico De Ranter
>>
>> Operations Engineer
>>
>>
>> eSATURNUS
>>
>> T. +32 16 40 12 82
>>
>> M. +32 497 91 53 78
>>
>> www.esaturnus.com
>>
>>
>> <http://www.esaturnus.com/>
>>
>>
>>  <http://vcard.esaturnus.com/>
>> _______________________________________________
>> dhcp-users mailing list
>> dhcp-users at lists.isc.org
>> https://lists.isc.org/mailman/listinfo/dhcp-users
>
> _______________________________________________
> 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/20121217/bea4be22/attachment.html>