lkml.org 
[lkml]   [1996]   [Sep]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: Bad news
On Sep 20, 12:45am, Boris Tobotras wrote:

} Subject: Bad news

Good morning to Boris and everyone else on the list. Hopefully this
note finds your end of the week going well.

> Hi,
>
> yes, there are bad news for me :(
>
> I tried 1.2.13, kernel `scsi' from slackware 3.0. Runs eight
> parallel crashmes without any fault. Rebooted then into 2.0.20 with
> this configuration:

... [ Configuration file deleted. ] ...

> (all other is 'N'). crashme OOPSes and reboots immediately, so
> I can't even write down OOPS sympthoms. And it is with slowest chipset
> settings, and 1.2.13 was with fastest. I have absolutely no idea what
> to do further. I'm just sure it _is_ a software bug somewhere in
> kernel...

I can perhaps add a comment which may or may not be helpful to this
discussion. The pharmacy in the Cancer Center has been supported for
almost 5 years on an old workhorse. This machine is an early
Gateway-2K 80386dx with 16Mbyte of memory.

The performance of this machine under 1.2.13 was phenomenal. In fact
it was this machine that was pushing 1 year of uptime with a total
process count of well over 3 million. I had posted a note to this
list when the machine rolled over the 1 million process mark. This
machine ran X about 18 hours a day, pushed three printers and handled
about 3-4 logins running our clinic support software. Never heard as
much as a peep out of the machine.

I kicked up 2.0.10 on the machine shortly after it came out. The
performance of this kernel was of course markedly superior to when the
machine was running 1.2.13. The problem we began to notice was that
certain of our resource hungry processes began to fault. As long as
the machine is under moderate load there is no problem. As memory
utilization and load increases we began to see occassional erratic
behavior. Totally reproducible. Performance with 1.2.13 is
unchanged.

I attributed this to the fact that the 2.0.x series is simply pushing
the hardware closer to the limits of its performance. The machine is
scheduled to come off-line and be retired to more sedentary uses so I
have not been too concerned about this.

The above comments are of course far from scientific but hopefully a
useful datapoint. I have long been under the suspicion that we may
flush out really marginal hardware as we push the performance envelope
of the architecture.

I hope that Boris doesn't interpret this as an attack on his
hardware. I just wanted to offer a datapoint that there is evidence
for a machine that functions flawlessly with 1.2.13 but not with the
2.0.x series.

Have a pleasant weekend everyone.

Greg

}-- End of excerpt from Boris Tobotras

As always,
Dr. G.W. Wettstein Velocity, LLC - Specialists in leveraging
4206 N. 19th Ave. inter-networking technology for enterprise groups.
Fargo, ND 58102 INTERNET: greg@wind.rmcc.com
Phone: 701-281-1686 INTERNET: greg@wind.enjellic.com
------------------------------------------------------------------------------
"According to the philosopher Ly Tin Wheedle, chaos is found in greatest
abundance wherever order is being sought. It always defeats order,
because it is better organized."
-- Terry Pratchett
Interesting Times

\
 
 \ /
  Last update: 2005-03-22 13:38    [W:0.020 / U:1.904 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site