I would expect there to be major issues when mixing 32b OF with 64b NOX,
because some of the fields in openflow.h (at least as of v0.8.1 - haven't
checked if this got updated) change size depending on 32/64b.  Some of the
substructures in ofp messages, like flow_expiration, are not 64b multiples,
and so padding gets inserted to ensure that all substructs start on 64b
boundaries.   This will have to be addressed with v0.9.

I don't know that this is your issue for sure, but I thought you might want
the warning.

-Brandon

On Wed, Jul 16, 2008 at 9:45 PM, kk yap <[EMAIL PROTECTED]> wrote:

> Hi Keith,
>
> No... They are 32bit qemu, that's why I have to run 32bit switches.
> But NOX is running the host, so it has to be 64-bit.  Let me know if
> you want to try any setup.  I can always do it and send you the
> result.
>
> Regards
> KK
>
> 2008/7/16 Keith Amidon <[EMAIL PROTECTED]>:
> > I'm afraid I don't have any idea what the problem is from the information
> > supplied.  I do have another question about the setup however.  It sounds
> > like you are running a 64-bit NOX with 32-bit OpenFlow Linux switches.
>  Are
> > those switches running in 32-bit KVMs on the 64-bit machine?
> >
> > While I don't know of any reason why it shouldn't work, to my knowledge
> we
> > have not done any mixed 64-bit controller 32-bit switch testing.  On my
> > setup where I run a 64-bit NOX I also use 64-bit switches in KVMs.  I'm
> > pretty busy with other activities but when I can get an hour or two free
> > I'll try running a mixed setup to see if I can recreate the problem.
> >
> >            --- Keith
> >
> >
> > On Wed, 2008-07-16 at 11:26 -0700, kk yap wrote:
> >
> > Hi,
> >
> > Attached is the verbose output.  The error occurs when  the OpenFlow
> > switches are being brought up.  The core dump shows the attached.  Any
> > advice?
> >
> > Regards
> > KK
> >
> > ==============================
> > GNU gdb 6.4.90-debian
> > Copyright (C) 2006 Free Software Foundation, Inc.
> > GDB is free software, covered by the GNU General Public License, and you
> are
> > welcome to change it and/or distribute copies of it under certain
> > conditions.
> > Type "show copying" to see the conditions.
> > There is absolutely no warranty for GDB.  Type "show warranty" for
> details.
> > This GDB was configured as "x86_64-linux-gnu"...BFD:
> > /home/ykk/nox/noxcore/src/.libs/lt-nox_core: don't know how to handle
> > OS specific section `.gnu.hash' [0x6ffffff6]
> > "/home/ykk/nox/noxcore/src/.libs/lt-nox_core": not in executable
> > format: File format not recognized
> >
> > Core was generated by `/home/ykk/nox/noxcore/src/.libs/lt-nox_core -v
> > -i ptcp:2525 pyrouting pyauthent'.
> > Program terminated with signal 11, Segmentation fault.
> > #0  0x00002ad2287c3f67 in ?? ()
> >
> >
> > 2008/7/11 Martin Casado <[EMAIL PROTECTED]>:
> >> Can you paste a trace using -v ?   If you gdb the core file, what line
> is
> >> it
> >> faulting on?  You might want to try deleting testing.sqlite and
> re-running
> >> ...
> >>
> >> btw, to run gdb on a libtool binary
> >>
> >> libtool --mode=execute gdb nox_core core
> >>
> >> .martin
> >>
> >>> Hi,
> >>>
> >>> I was trying to run NOX in a Dell AMD64 machine.  The whole
> >>> compilation process was okay.  I am using openflow/linux kernel
> >>> binaries compiled on an i386 machine, while compiling the NOX on the
> >>> AMD.  The nox_core binary run fine, until after a while, the following
> >>> error appears.  I can run the setup properly on the i386 machine and
> >>> so I wonder why it fails on this machine.
> >>>
> >>> $ ./nox_core -i ptcp:2525 pyauthenticator pyrouting
> >>> NOX 0.2.1 (nox_core), compiled Jul 10 2008 16:37:53
> >>> Compiled with OpenFlow 0x84 (exp)
> >>> 00001|sqlite3-impl|WARN:Transactional storage file not given in the
> >>> command line; defaulting to 'testing.sqlite'
> >>> Caught signal 11.
> >>>  0x2b8cd5719132 6668976 (vigil::fault_handler(int)+0x22)
> >>>  0x00d12190
> >>>  0x505400000003
> >>> Segmentation fault (core dumped)
> >>>
> >>> Anyone in the know?
> >>>
> >>> Thanks.
> >>>
> >>> Regards
> >>> KK
> >>>
> >>> _______________________________________________
> >>> nox-dev mailing list
> >>> [email protected]
> >>> http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org
> >>>
> >>
> > _______________________________________________
> > nox-dev mailing list
> > [email protected]
> > http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org
> >
>
> _______________________________________________
> nox-dev mailing list
> [email protected]
> http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org
>
_______________________________________________
nox-dev mailing list
[email protected]
http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org

Reply via email to