Christian Smith wrote:

> Paul (NCC/CS). uttered:
>
> >
> > hi ,
> >
> > i have same as you.    96 mb ram and 110Mhz ss5.
> >
> > and cgsix.
> >
> > so i'm surprised you boot past it.!
> >
> > looking like something wrong with my machine.
> >
> > but what.
>
> Quite possibly memory. Try pulling SIMMs and see if the problem goes away.
>
> Christian



Hi, thanks,

well it's looking more and more like this is the source of my problems.

In trying to simply get the simplest fastest way to a working machine,
i've got a successful install of 2.0 running. But building things in package
source has been a problem. Basically i've had heaps of builds die. And one
of them reported ---> signal 11.   (the dreaded.) Plus clisp failed in "minitests"
which i remember from years ago was due to faulty ram.

It's likely that 3.0 is failing due to this.

Another reason i had weird and massive build problems was that I found that
some builds would die after spawning a shell. BUT rather than exiting the shell
and coming back up the stack to the calling build it would sit there in that
shell!  Half the process's still running!  Only saw this by , well , looking, ps -aux.
Half a dozen makes and other stuff still running made for unpredictable
results when attempting a new build. I had to type "exit" at the prompt to
get the whole damn thing to unravel and kill all the process's.

At the moment i'm getting things built slowly through persistence.
Like for example "t1lib" died during building xpdf. But later building
t1lib alone was a success. So slowly i'm getting what i want. It'd probably
dependent upon how much is running at one time and where it ends up
in the SIMM at the time. Is my guess.

When this is finished i will try to find out what is wrong with my hardware
and fix it. Then i should be able to build a machine base on 3.0. On another
disk mind you.

Do you know off hand the full specification of the SIMM's my SS5 uses?
I might need to seek some out. And do you know where to get them nowadays?

thanks for your help.

Paul.







>
>
> --
>      /"\
>      \ /    ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL
>       X                           - AGAINST MS ATTACHMENTS
>      / \

--

____________________________________________

Paul
Australian Bureau Of Meteorology
____________________________________________