于 2011年05月22日 23:22, Henry Gebhardt 写道:
> On Sun, May 22, 2011 at 02:50:00PM +0800, microcai wrote:
>> Heya,
>>
>> Here is the 2nd version.
>>
>> Split the DHCP and TFTP socket and use Servie=dnsmasq.servie to point to
>> dnsmasq.
>>
>> Add einfo to the user tell then how to do with the socket unit
2011/5/20 Lennart Poettering :
> On Fri, 20.05.11 17:40, Chen Jie ([email protected]) wrote:
>
>> 2011/5/19 Chen Jie :
>> > 2011/5/18 Kay Sievers :
>> >>
>> >> Completely different! That's an Intel Core Duo 1.4 GHz laptop:
>> >> systemd-analyze blame | grep udev
>> >> 87ms udev-trigger.service
>>
On Sun, May 22, 2011 at 07:55:09PM +0200, Michał Górny
wrote:
> Well, that's not entirely guaranteed here. The point is to use
> systemd_dounit() and similar functions because they can decide whether
> to install the service files or not.
Hm, but why don't you patch dnsmasq as described in daemo
On Sun, 22 May 2011 17:22:30 +0200
Henry Gebhardt wrote:
> On Sun, May 22, 2011 at 02:50:00PM +0800, microcai wrote:
> > Heya,
> >
> > Here is the 2nd version.
> >
> > Split the DHCP and TFTP socket and use Servie=dnsmasq.servie to
> > point to dnsmasq.
> >
> > Add einfo to the user tell then
On Sun, May 22, 2011 at 02:50:00PM +0800, microcai wrote:
> Heya,
>
> Here is the 2nd version.
>
> Split the DHCP and TFTP socket and use Servie=dnsmasq.servie to point to
> dnsmasq.
>
> Add einfo to the user tell then how to do with the socket unit files.
>
> both DNS and tftp have been test.
