2.4 Success story

2.4 Success story

[Colonel]

A brief note of _thanks_ to all that create the linux kernel.

Mid-morning, I checked on a database fetch that has occaisonally
gotten out of hand and overrun the system -- depending on the kernel
version.  It was overrun but this time, it was probably due to a
change I made in the fetch code yesterday....  but the _real_ issue is
that this is the first time I have ever had a very responsive system
under a load of 13+ and a couple hundred processes.  There have been
countless times in the past that getting control when the system load
has gone past 6-7 was nearly impossible.

I was completely amazed and spent about 30 seconds to tame the rogue
processes, with system access feeling just like normal load
conditions.

This particular gem of a kernel is:

2.4.9-ac10 #1 SMP Tue Sep 11 21:47:15 PDT 2001 i686

THANKS!


I'd suggest that those running large applications experiencing
problems in transition from 2.2 -> 2.4 give this kernel a try.

Re: 2.4 Success story

[José Luis Domingo López]

On Tuesday, 18 September 2001, at 12:20:03 -0700,
Colonel wrote:

> A brief note of _thanks_ to all that create the linux kernel.
> [...]
> This particular gem of a kernel is:
> 
> 2.4.9-ac10 #1 SMP Tue Sep 11 21:47:15 PDT 2001 i686
> 
>From my particular experience with a couple of very low-end computers
moderately loaded (mainly used as workstations) I can say that in _my_ own
setup and hardware, memory management in 2.4.9-ac10 feels much better than
2.4.9's. In 2.4.9-ac10 I've been using:
echo "1" > /proc/sys/vm/page_aging_tactic

and the result is a much better behavior of the system. Now, with the same
applications loaded, daily cron doesn't force all of my appliactions into
swap (something 2.4.9 did). Now swap usage is much lower, the system seems
to have a faster response to user interaction than before and "swapoff" is
_much_ (maybe x10) faster than in 2.4.9.

It seems that we are getting closer to a production-quality kernel :)

-- 
José Luis Domingo López
Linux Registered User #189436     Debian GNU/Linux Potato (P166 64 MB RAM)
 
jdomingo EN internautas PUNTO org  => ¿ Spam ? Atente a las consecuencias
jdomingo AT internautas DOT   org  => Spam at your own risk

Re: 2.4 Success story

[Jeffrey Ingber]

[-- Attachment #1: Type: text/plain, Size: 2023 bytes --]

I agree.  This is a fantastic kernel - so good infact, it'll probably be
awhile before I try out a new one unless some other issue(s) comes up. 
I had a nagging problem with XF864 dying with Sig'11 on SMP and this has
completely cleared it up.  

There's probably several issues with this kernel, but as the saying goes
'It works for me.'.  Excellent work.

Jeffrey H. Ingber (jhingber _at_ ix.netcom.com)


On Wed, 2001-09-19 at 15:33, José Luis Domingo López wrote:
> On Tuesday, 18 September 2001, at 12:20:03 -0700,
> Colonel wrote:
> 
> > A brief note of _thanks_ to all that create the linux kernel.
> > [...]
> > This particular gem of a kernel is:
> > 
> > 2.4.9-ac10 #1 SMP Tue Sep 11 21:47:15 PDT 2001 i686
> > 
> >From my particular experience with a couple of very low-end computers
> moderately loaded (mainly used as workstations) I can say that in _my_ own
> setup and hardware, memory management in 2.4.9-ac10 feels much better than
> 2.4.9's. In 2.4.9-ac10 I've been using:
> echo "1" > /proc/sys/vm/page_aging_tactic
> 
> and the result is a much better behavior of the system. Now, with the same
> applications loaded, daily cron doesn't force all of my appliactions into
> swap (something 2.4.9 did). Now swap usage is much lower, the system seems
> to have a faster response to user interaction than before and "swapoff" is
> _much_ (maybe x10) faster than in 2.4.9.
> 
> It seems that we are getting closer to a production-quality kernel :)
> 
> -- 
> José Luis Domingo López
> Linux Registered User #189436     Debian GNU/Linux Potato (P166 64 MB RAM)
>  
> jdomingo EN internautas PUNTO org  => ¿ Spam ? Atente a las consecuencias
> jdomingo AT internautas DOT   org  => Spam at your own risk
> 
> -
> 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/


[-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --]

Re: 2.4 Success story

[Alan Cox]

> I agree.  This is a fantastic kernel - so good infact, it'll probably be
> awhile before I try out a new one unless some other issue(s) comes up.=20
> I had a nagging problem with XF864 dying with Sig'11 on SMP and this has
> completely cleared it up. =20

As far as I can tell the XFree problem was almost certainly the LDT thing, 
and if so then it was fixed when I merged Linus changes, so should also be
fine in Linux tree now.

If not then some hard looking is needed to make sure whatever was the fix
is definitely in Linus tree