Re: [gentoo-user] Re: openrc fails

[Dale <rdalek1967@gmail.com>]

covici@ccs.covici.com wrote:
> Dale<rdalek1967@gmail.com>  wrote:
>
>    
>> walt wrote:
>>      
>>> On 10/10/2010 03:17 AM, Stéphane Guedon wrote:
>>>        
>>>> I have A problem on my gentoo server (but not my gentoo laptop, which is
>>>> bizzare as they together have the same release of system basic
>>>> packages !).
>>>>
>>>> At launching default runlevel, several services appear to be not
>>>> launched
>>>> whereas they are ! This concerns proftpd, mysql and console-kit (and all
>>>> services dependant of course).
>>>>
>>>> For exemple, when launching proftpd, openrc says :
>>>>
>>>> * Starting ProFTPD ...
>>>>    * start-stop-daemon: /usr/sbin/proftpd is already running
>>>>    [ !! ]
>>>>    * ERROR: proftpd failed to start
>>>>
>>>> mysql :
>>>>    * "/var/run/mysqld/mysqld.pid" is still present and the process
>>>> is running.
>>>>    * Please stop it "kill 5478" maybe ?
>>>>    [ !! ]
>>>>    * ERROR: mysql failed to start
>>>>
>>>> I have set :
>>>> rc_parallel="NO"
>>>> rc_depend_strict="YES"
>>>>
>>>> Anyone have an idea where to start to solve the problem ?
>>>>          
>>> When you stop the machine do you see any messages about services
>>> that fail
>>> to stop?  Any reason to suspect that those *.pid files don't get deleted?
>>>
>>> You could try booting into single-user mode to see if those *.pid files
>>> really are still there from the previous session.
>>>
>>> Do the two machines run the same version of baselayout?
>>>
>>>        
>> Does the zap option exist in openrc?  That may get rid of the pid file too.
>>
>> Just a thought.
>>      
> I wonder if he has something in two different runlevels because
> start-stop says something is already running, not just that the pid file
> is there.
>
>    

Based on a post made later, it would appear that he has more pressing 
issues.  If it thinks the service is running after a reboot, there is 
problems.  On my rig, those directories are cleared during boot because 
there can't be a service running right after the kernel loads.  There 
shouldn't be any started yet since it hasn't even got to a runlevel.

Your thoughts would be a logical one tho.  If he has a service added to 
both boot and default runlevels, then after it gets through the boot 
level it will try to start the service a second time.  You could be on 
to at least one part of the problem.   I'm not sure on the part after a 
reboot tho.

Dale

:-)  :-)