Re: huge load average follow up

Andi Gutmans (andi@vipe.technion.ac.il)
Mon, 28 Apr 1997 23:37:14 +0300 (IDT)


Hi,

Well this seems to be happening to people but noone knows what it is.
Is this ever going to be fixed Linus? I'm keeping my system up in case
someone wants debugging information. Load average is now 34. The system is
very usable with 20 people logged in etc...

Andi

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Andi Gutmans - Computer Science, Technion

PGP public key: finger andi@vipe.technion.ac.il

On Mon, 28 Apr 1997, Tim Ricketts wrote:

> On Sat, 26 Apr 1997, Andi Gutmans wrote:
>
> > Hi,
> >
> > OK this really looks like a kernel bug to me. kernel 2.0.30.
> > PID TTY STAT TIME COMMAND
> > 26070 ? D 0:00 in.ntalkd
> > 1:10 vipe /home/staff/andi# strace -p 26070
> > attach: ptrace(PTRACE_ATTACH, ...): Operation not permitted
> > I can't kill this and another 32 in.ntalkd's.
> > load average is 32+ with system running fast. no slowdown noticeable
> >
> This looks exactly like what happened to me yesterday.
> I was experimenting with linux-nfs-0.4.21 and I did
> mount -t nfs localhost:/home/ftp/pub ftp
> as root, and it hung. The mount process was in the same state as your
> in.ntalkd's and I couldn't kill it. Unfortunately init was in the same
> state so I had to press reset.
>
> -- Tim
> Quidquid latine dictum sit, altum viditur.
>
>