[Dnsmasq-discuss] Is there any use case for "--dhcp-range=0.0.0.0, static, 0.0.0.0" to support all subnet

Simon Kelley simon at thekelleys.org.uk
Thu Mar 15 17:59:52 GMT 2018


There's something very similar to this already in place for DHCPv6,
where it's allowed to specify

--dhcp-range=::,static

To allow any defined static allocation. For consistency, it may make
sense to simply use 0.0.0.0

so --dhcp-range=0.0.0.0,static

represents the special case. Netmask handling in that case would be have
to be automatic and hidden.

Would that make sense?


Cheers,

Simon.






On 12/03/18 08:00, Yuki Nisiwaki wrote:
> Hi dnsmasq guys.
> 
> I'm Yuki Nishiwaki.
> Recently I started to work on dnsmasq stuff. I could ask stupid thing
> though I want to ask if anyone think the same thing as me.
> 
> I'm using dnsmasq under cloud controller and dnsmasq have responsibility
> to lease IP for
> our all virtual machines connected to Host via tap on the same host
> running dnsmasq.
> 
> In order to do that, I had to specify same number of --dhcp-range as the
> number of subnet VM belong to but
> we can not add dynamically "--dhcp-range" and I don't wanna restart
> everytime new subnet added.
> 
> So I was going to specify "--dhcp-range=0.0.0.0,static,0.0.0.0"  or
> "--dhcp-range=0.0.0.0,static,128.0.0.0,
> --dhcp-range=128.0.0.0,static,128.0.0.0" to
> support all subnets possibly will be added in the future and lease IP
> for only known client.
> It kind of worked but the netmask of leased IP was the same as
> dhcp-range specified(/1). 
> I expected the netmask will be got from the NIC receiving DHCP request
> and dhcp-range's netmask is only used for judging if dnsmasq should
> reply or not.
> 
> After read the code around
> https://github.com/imp/dnsmasq/blob/master/src/dhcp.c#L501-L561, I
> understood why I can not.
> But If there is any chance I can add new option or other behaviour to
> cover my usecase
> so that dnsmasq can use the netmask of NIC's IP receiving DHCP request
> for the IP to be leased while the netmask dhcp-range specified is used
> to judge whether
> dnsmasq should reply or not, I'm willing to write the patch but
> beforehand I want to ask what do you think about such additional change.
> Is this good approach, feature or bad approach, feature nobody want but me?
> 
> 
> Regards
> 
> Yuki Nishiwaki
> 
> 
> _______________________________________________
> Dnsmasq-discuss mailing list
> Dnsmasq-discuss at lists.thekelleys.org.uk
> http://lists.thekelleys.org.uk/mailman/listinfo/dnsmasq-discuss
> 




More information about the Dnsmasq-discuss mailing list