Re: [patch 1/2] fork_connector: add a fork connector

[Guillaume Thouvenin <guillaume.thouvenin@bull.net>]

On Tue, 2005-03-29 at 22:06 -0800, dean gaudet wrote:
> On Tue, 29 Mar 2005, Jay Lan wrote:
> 
> > The fork_connector is not designed to solve accounting data collection
> > problem.
> > 
> > The accounting data collection must be done via a hook from do_exit().
> 
> by the time do_exit() occurs the parent may have disappeared... you do 
> need to record something at fork() time so that you can account to the 
> correct ancestor.

You're right. At fork(), the "job daemon", provided by ELSA, records
information about parent PID, child PID and also about the group of
processes they belong to. At exit(), accounting data are recorded by CSA
or BSD-like accounting. 

> an example of where this ancestry is useful would be the summation of all 
> cpu time spent by children of apache, spamd, clamd, ...

You're right. One usage can be: apache, spamd and clamd can be put in a
job (a group of processes) by using the "job daemon" and automatically,
all children will belong to the same jobs respectively. So the gaol here
is really to perform per-group of processes accounting using ELSA and
CSA accounting data.

Best Regards,
Guillaume

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/