[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: FC6 faster boot

On Sun, 2006-04-09 at 19:32 +0200, David Nielsen wrote:
> søn, 09 04 2006 kl. 10:03 +0200, skrev dragoran:
> > Before FC4 was released there was a disussion about how to make the boot 
> > process faster.
> > But until now (FC5/FC6-rawhide) nothing has been done.
> > Early login was in development but it stopped,
> > Systemservices was never in a working state.
> > Any plans on improving the boot process in FC6?
> > I would suggest to use initng which is in review for extras now [1].
> > But to get it in the final release it would be better to have it in 
> > rawhide rather than in extras.
> > 
> > [1]: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=173459
> > 
> While I would love to see the current Init system replaced for various
> reasons, I'm concerned if InitNG is the correct technical solution.

We would have evaluate it to know more. Right? What concerns do you have
for initng?

> Do we even have a clear spec for what we require from a replacement, so
> far the document with the most effort going into this seems to be Seth
> Nickell' SystemServices document.

Have you seen this? http://fedoraproject.org/wiki/FCNewInit

Proper *runtime* dependency support. Starting a service should start its
dependencies, in-order. Changes to dependent services should be
immediately handled. Out of this work, parallelization can easily fall
out. Without this work, parallelization is pointless. :)

Full backwards compatibility. Old-style init scripts aren't going to go
away anytime soon. 

Full LSB support for LSB init scripts. Makes the spec groups happy, even
though the number of LSB scripts is basically nil. 

D-BUS support. Services should be exposed via D-BUS, and should be
available for querying to: 
                      * check what's configured 
                      * check what's actually running 
                      * start them 
                      * stop them 
                      * etc. 
The services should also post notifications via D-BUS when they've
started successfully, and (more importantly) when they haven't, or when
they've unexpectedly exited. 

Support for respawning services, if necessary. Complete with rate limits
and other fun stuff. 

How well does Initng fit into those requirements currently?

> Just saying we should use initng reminds me of that old saying - when
> all you have is a hammer every problem looks like a nail.

Unlike a hammer, Initng can potentially be modified to suit our
requirements if it doesnt fit our specifications if it is a good base to
start with and upstream is open to accepting changes if any is


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]