Rusty,

I have two identical machines and the problem exists on both.  The machines
have Megaplex II quad  Pentium III Xeon PCI ISA system boards.  I have 550
Pentium III Xeon, 100/512 processors.  I am running test11 which I
downloaded a few days ago.  I have installed no patches.  My iptables is
1.1.2.

I compiled two version of the kernel, one is single processor and one is
SMP.  I can successfully run iptables on the single processor kernel and I
can also get it to work with two processors using the SMP kernel.  With 4
processors the system boots fine, but when I try to run iptables with 4
processors the system freezes solid.  It does not write anything to the
system log.

We currently have iptables working on a single processor machine in our
network.  It typically tracks an average of 13000 connections with peaks
reaching 21000 connections.  It works flawlessly.  We want to upgrade from
10/100 interfaces to GigE.  We hoped to use the 4 processor machines to do
this.

I am attaching my kernel config file.  If you have any ideas how to get all
four processors working please let me know.


Thanks


Roger




Rusty Russell wrote:

> In message <3A27D4D6.4DA47346@lanl.gov> you write:
> >
> > I have 2.4.0  test 10 and test 11 installed on a multiprocessor (Intel)
> > machine.  I have tried both test versions of the kernel.  I configured
> > the kernel for single
> > and multi processor.  When I boot single processor, iptables will run
> > fine.  When I boot the machine with the multiprocessor kernel and run
> > iptables, the kernel dumps several pages of hex and the final two lines
> > of output are:
> >
> > Killing interrupt handler
> > scheduling in interrupt
>
> My development box (running test10pre5) is SMP, and it works fine.  I
> haven't updated to the latest kernel version because I like my
> filesystems in one piece, and the netfilter code hasn't changed.
>
> What is your kernel configuration, and iptables version?  Have you
> patched the kernel?
>
> Thanks for the report,
> Rusty.
> --
> Hacking time.