Jens Persson escribió:
> Ismael Luceno wrote:
>>> Yes I have the latest. But if you say there shouldn't be warnings at all 
>>> I should be looking somewhere else to fix it. This is on my own distro 
>>> Enlisy and is the only thing left to fix, otherwise I have a 100% clean 
>>> Initng boot. But it's late here in Sweden now but I'm open for 
>>> suggestions what could be the problem so I could fix it later. These 
>>> warnings comes early so it might be a service that starts "too early" 
>>> that causes it. Anyway, thanks for the help so far.
>>>     
>> I assume you have udev and a tmpfs /dev.
>>
>> Then the /dev/initng/* stuff should be regenerated when
>> system/udev/mountdev service mounts a tmpfs on /dev.
>>
>> That is done by sending a SIGHUP to initng; the problem here is that
>> udev.i uses pidof to get the pid of initng.
>>
>> If you don't have something "strange" in your boot process, then you can
>> replace the `pidof initng` to 1 in udev.i .
>>
>> I hope that fix your problem. :)
>>
>> I will add pidof to initng, when I can commit again...
>>
>>
>> Jimmy, if you are there, what is going on with the server? :(
>>   
> 
> Pidof would be nice to have. Right now I'm using a my own hack for it.

The only thing needed is a symlink to killalli5 :), but was never
added..., maybe because everybody has sysvinit.

> 
> /bin/kill -SIGHUP `/bin/ps -C initng -o pid=`
> 
> I guess that's the line you were talking about.

Yes, but replacing the `...` thing with a 1 is more efficient,
even more than making the symlink. ;)

-- 
_______________________________________________
Initng mailing list
[email protected]
http://jw.dyndns.org/mailman/listinfo/initng

Reply via email to