[Kea-users] Identify on VENDOR_CLASS_

Mathias Aichinger mathias.aichinger at easysol.at
Thu Jul 20 12:55:44 UTC 2023


Hi Ronald,

I get your point, but I think the chance that they change it in their product lifecyle is low. When changing to a new product it's a different story as I have already showed with the example of the P6426.

The Geneos guide says the following:
firmware name, hardware platform, CPE name and model, dslforum.org

One alternative for the live (configured) devices would be to change it per config and put them into another vendor class:

geneos(config-if-vlan)# dhcp client send-option 60 “genexis”

But this client related, maybe somebody has another "KEA" approach 😉

Best Regards,

Mathias

> On 20.07.2023, at 14:42, DDFR | Ronald Blaas <ronald.blaas at ddfr.nl> wrote:
> 
> Hi Mathias,
> 
> I was just about to sent the exact same thing. 🙂
> 
> The problem I have with the substring is that I do not know if the substring will change (with a newer software update or whatever..) 
> 
> as for you expression starting on 0 and a length of 6 will give me (in the currect string) 'geneos'  but what gaurantees are there option 60 string will not change (to polar,geneos...xxxx., whatever) 
> 
> until now I can make the expression by using split with the same limitations as mentioned above. 
> "test": "split(option[60].text, 0x2C, 2) == 'polar'",
> 
> 
> Thanks for the insights. 
> 
> 
> Van: Mathias Aichinger <mathias.aichinger at easysol.at>
> Verzonden: donderdag 20 juli 2023 14:37
> Aan: DDFR | Ronald Blaas <ronald.blaas at ddfr.nl>
> CC: kea-users at lists.isc.org <kea-users at lists.isc.org>
> Onderwerp: Re: [Kea-users] Identify on VENDOR_CLASS_
>  
> Hi,
> 
> I have just checked with my device. Genexis is sending it already with colons therefore the automatically created class would be " VENDOR_CLASS_geneos-polar-3.13.1-R,polar,fibertwist-p2410,dslforum.org <http://dslforum.org/>" which does not help you. 
> 
> I personally would use the substring function to extract geneos-polar without the version stuff as this is changing.
> 
> Unrelated example: "test": "substring(option[60].hex,0,6) == 'docsis'"
> 
> A newer fibertwist with the new software platform is sending this:
> FiberTwist-P6426,GenXOS-Panda-GNX-11.8.0-R,dslforum.org <http://dslforum.org/>
> So here I would extract "FiberTwist-P6426".
> 
> Best Regards,
> 
> Mathias
> 
>> On 20.07.2023, at 14:25, DDFR | Ronald Blaas <ronald.blaas at ddfr.nl> wrote:
>> 
>> Hi Mathias,
>> 
>> thanks for the quick reply. 
>> Unfortunately this does not work
>> When I try your solution kea-dhcp does not start:
>>  expression: [member('polar')] error: <string>:1.8-14: Not defined client class 'polar'
>> 
>> Regards,
>> 
>> Van: Mathias Aichinger <mathias.aichinger at easysol.at>
>> Verzonden: donderdag 20 juli 2023 14:15
>> Aan: DDFR | Ronald Blaas <ronald.blaas at ddfr.nl>
>> CC: kea-users at lists.isc.org <kea-users at lists.isc.org>
>> Onderwerp: Re: [Kea-users] Identify on VENDOR_CLASS_
>>  
>> Hi Ronald,
>> 
>> I think you don't need the "VENDOR_CLASS_".
>> 
>> try with "test": "member('polar')"
>> 
>> Best Regards,
>> 
>> Mathias
>> 
>>> On 20.07.2023, at 14:06, DDFR | Ronald Blaas <ronald.blaas at ddfr.nl> wrote:
>>> 
>>> Hi all,
>>> 
>>> I am trying to identify on VENDOR_CLASS_ bu t seems not to be working.. 
>>> 
>>> This is what I tried so far:
>>> 
>>> kea-conf
>>> (snippet)
>>>  "client-classes": [
>>>         {
>>>             "name": "Genexis",
>>>             "test": "member('VENDOR_CLASS_polar')",
>>> 
>>> 
>>> 
>>> Debug Log
>>> 
>>> DEBUG [kea-dhcp4.dhcp4/115727.139716649023040] DHCP4_CLASS_ASSIGNED [hwtype=1 58:00:32:31:34:f0], cid=[ff:00:01:00:01:00:03:00:01:58:00:32:31:34:f0], tid=0x7dba8a8d: client packet has been assigned to the following class(es): ALL, HA_dhcp2, VENDOR_CLASS_geneos-polar-3.13.1-R,polar,fibertwist-p2410,dslforum.org <http://dslforum.org/>, UNKNOWN
>>> 
>>> DEBUG [kea-dhcp4.eval/115727.139716649023040] EVAL_DEBUG_MEMBER Checking membership of 'VENDOR_CLASS_polar', pushing result 'false'
>>> 
>>> 
>>> 
>>> I was succesfull using this test:
>>>           "test": "split(option[60].text, 0x2C, 2) == 'polar'",
>>> 
>>> But I have no clue if 'polar' is always on the second value.... 
>>> 
>>> Any insights is appreciated.
>>> 
>>> Regards
>>> -- 
>>> ISC funds the development of this software with paid support subscriptions. Contact us at https://www.isc.org/contact/ for more information.
>>> 
>>> To unsubscribe visit https://lists.isc.org/mailman/listinfo/kea-users.
>>> 
>>> Kea-users mailing list
>>> Kea-users at lists.isc.org <mailto:Kea-users at lists.isc.org>
>>> https://lists.isc.org/mailman/listinfo/kea-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/kea-users/attachments/20230720/5c326921/attachment-0001.htm>


More information about the Kea-users mailing list