On Fre, 2008-02-15 at 15:08 +0530, Pravin wrote:
> On Thu, Feb 14, 2008 at 7:14 PM, Bernd Petrovitsch <bernd@xxxxxxxxx> wrote:
> > On Don, 2008-02-14 at 14:35 +0100, Thomas De Schampheleire wrote:
> > > On Thu, Feb 14, 2008 at 2:22 PM, Pravin <shindepravin@xxxxxxxxx> wrote:
> > > I doubt that init will have a lot to do while your application is
> > > running. According to me, init will just sleep and performance of
> > > your application will not be degraded. Maybe someone else can confirm
> > > this.
> >
> > ACK. "init" will just wait for signals (be it a SIGCHLD from a
> > terminated process or an `kill -1` to reread /etc/initttab) or the
> > effects of a `telinit`.
> I do agree that once main initialization is over then only work for init is
Of course after initialization which happens at start/boot time and if
you tell init to reread /etc/inittab.
> to acknowledge the processes without parent, and it should b signal based.
> > Just boot Linux somewhere and do `strace -p 1` and see yourself.
> I tried it, but got an error saying "Operation not permitted" even
> though I was root...!!!
> {{{
> # strace -p 1
> attach: ptrace(PTRACE_ATTACH, ...): Operation not permitted
> }}}
Ooops, ACK. Obvious I never tried it myself;-)
Well, "init" the first normal process. However it has a few specialities
and it seems that this is one more.
> To "Thomas De Schampheleire"
> My application is not hard real time system, but its performance is important.
> Its an Intrusion Prevention System(IPS), and any delay introduced by
> these systems
> are directly noted by users in there internet experience.
Personally I don't think that a (sane) "init" can cause any measurable
performance degradation on an IPS.
Bernd
--
Firmix Software GmbH http://www.firmix.at/
mobil: +43 664 4416156 fax: +43 1 7890849-55
Embedded Linux Development and Services
--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
Please read the FAQ at http://kernelnewbies.org/FAQ
