Mailinglist Archive: opensuse (3767 mails)

< Previous Next >
Re: [SLE] maximum nproc value
  • From: Steve Graegert <graegerts@xxxxxxxxx>
  • Date: Thu, 2 Feb 2006 21:56:59 +0100
  • Message-id: <6a00c8d50602021256ia708672i42a336c44da59a85@xxxxxxxxxxxxxx>
On 2/2/06, Per Jessen <per@xxxxxxxxxxxx> wrote:
> Per Jessen wrote:
>
> > I just ran the above via a remote konsole, and my 2-way machine
> > certainly _appears_ to be locked up. I'll have to make my way to the
> > computer-room to check out the local console.
>
> Well, the local console was still responding, but after entering a
> single keystroke, I had to wait for about 5 mins before the next was
> accepted. I ended up rebooting that machine. It was not locked up as
> such, merely unusable and the fastest recovery was to reboot. It took
> about 1, maybe 2mins for it to accept the Ctrl-Alt-Del combination.
> Of course, this was still on 2.4 - it's quite possible 2.6 with proper
> pre-emption would have behaved much better.

Yes, true. I ran the code I've posted an hour ago and it didn't do
any harm. I suppose the new O(1) scheduler has proven to scale well.
Nevertheless, the fork bomb is as old as the system call itself and
should never cause a system to lock up. Solaris 2.6+ and Tru64 Unix
4.1+ have never shown any anomalies.

\Steve

--

Steve Graegert <graegerts@xxxxxxxxx>
Software Consultant {C/C++ && Java && .NET}
Office: +49 9131 7123988
Mobile: +49 1520 9289212

< Previous Next >