Improving halt package interaction...

[Thomas M Steenholdt]

Jesse Keating wrote:
> 
> Wouldn't you just make sure that the UPS shut down script is
> scheduled /after/ the database shut down, and all other critical
> services?  IE make it close to last in the shutdown schedule.
> 

It is really best to handle this as close to the actual halt as 
possible. This is the same reason why the NUT handling code has already 
been put in exactly that place in the current halt script - We just need 
to make it more generic and easier to "plug-in" package specific scripts.

Also, it seems from the feedback, that UPSes are not the only cases 
where this would prove useful. Perhaps you need to perform an operation 
after "/" has been remounted read-only, which would never be possible 
from a normal initscript.

/Thomas