[gentoo-dev] Strange behavior of rc scripts

[gentoo-dev] Strange behavior of rc scripts

[Norbert Kamenicky]

Hi to all !

On unixes it's quite common, that if some daemon started from
init rc scripts fails, it's possible to re/start it paintless by  hand 
again ...
( but not on Gentoo ? )

Let's try this:

#pkill -9  some_daemon

#/etc/init.d/some_daemon  start  (or restart)

You'll get stupid message "running already" and
the command fails with no action ... :o)

 so probably stop is needed before ;-)

#/etc/init.d/some_daemon  stop

You'll get  message "not running", but if you try start it
again, stupid message comes again ! :-(

It's getting me really very  nervous ... :-(

PS.
When I switched from RH to Gentoo, the first thing
I was worry about were rc init scripts, because I can
only hardly to imagine something better then system 5 ones...
let's try to convince me I am wrong :-)


Noro


--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Strange behavior of rc scripts

[leio]

On Fri, 6 Jun 2003, Norbert Kamenicky wrote:

> On unixes it's quite common, that if some daemon started from
> init rc scripts fails, it's possible to re/start it paintless by  hand 
> again ...
> ( but not on Gentoo ? )

If a service is not running but the init system thinks it is, you just 
need to zap it, just as suggested in the initscripts helpfile (I believe 
it was /etc/init.d/<a initscript> help but I could remember wrong, not at 
home and forced to use MS Windows atm :( )

/etc/init.d/<initscript> zap

> When I switched from RH to Gentoo, the first thing
> I was worry about were rc init scripts, because I can
> only hardly to imagine something better then system 5 ones...
> let's try to convince me I am wrong :-)

Gentoo initscripts are just about perfect and work is in progress to make 
it even more perfect (if I recall correct then the initscript system is 
being developed to hold initscript dependencies in a tree so that multiple 
services can be started to increase boot-up time heavily)
If you cared to read the docs from gentoo.org you wouldn't ask about it.
http://www.gentoo.org/doc/en/rc-scripts.xml which is directly linked from 
the docs link on gentoo homepage.
Also such questions do NOT go to the gentoo-dev mailing list.

Mart Raudsepp


--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Strange behavior of rc scripts

[Norbert Kamenicky]

leio@hercules.dustbite.org wrote:

>On Fri, 6 Jun 2003, Norbert Kamenicky wrote:
>
>  
>
>>On unixes it's quite common, that if some daemon started from
>>init rc scripts fails, it's possible to re/start it paintless by  hand 
>>again ...
>>( but not on Gentoo ? )
>>    
>>
>
>If a service is not running but the init system thinks it is, you just 
>need to zap it, just as suggested in the initscripts helpfile (I believe 
>it was /etc/init.d/<a initscript> help but I could remember wrong, not at 
>home and forced to use MS Windows atm :( )
>
>/etc/init.d/<initscript> zap
>
>  
>
>>When I switched from RH to Gentoo, the first thing
>>I was worry about were rc init scripts, because I can
>>only hardly to imagine something better then system 5 ones...
>>let's try to convince me I am wrong :-)
>>    
>>
>
>Gentoo initscripts are just about perfect and work is in progress to make 
>it even more perfect (if I recall correct then the initscript system is 
>being developed to hold initscript dependencies in a tree so that multiple 
>services can be started to increase boot-up time heavily)
>If you cared to read the docs from gentoo.org you wouldn't ask about it.
>http://www.gentoo.org/doc/en/rc-scripts.xml which is directly linked from 
>the docs link on gentoo homepage.
>Also such questions do NOT go to the gentoo-dev mailing list.
>
>Mart Raudsepp
>
Yeah, I know, I have to use "zap", but I can't understand
what is the advantage of  this complication ...

Perfect is something, which is simple and easy to use ... (KISS)
can you imagine, new Mercedes (automatic gearbox equiped)
with  an extra clutch ?

Noro




--
gentoo-dev@gentoo.org mailing list