[tlug] freeze on AMD64 Dual Opteron server

[Marcus Metzler <mocm@example.com>]

>>>>> "Evan" == Evan Monroig <evan.monroig@example.com> writes:

    Evan> Dear Tlug, I have a server with two AMD64 Opteron processors
    Evan> and 8GB of ram, that I use as a simulation box here at the
    Evan> laboratory.  Communitation with it is entirely through ssh.

    Evan> I have run simulations full time for a few days, generating
    Evan> tens of gigabytes of data at high speed, all without
    Evan> problem.

    Evan> But now I would like to send the data to another computer
    Evan> using rsync, and after about 20 seconds it invariably hangs.

Looks like it may be a problem with your network card (or network
chipset on the board). Since you get the lockup as soon as you start
sending lots of data over the network.
I used to have problems like that when the realtek GBit cards when
there were still problems with the driver.

    Evan> There is no way to do anything.  The ssh connection is cut,
    Evan> the computer is doesn't appear on the network (I configured
    Evan> a static IP address).  I tried connecting a screen and
    Evan> keyboard to it while doing the rsync, but after failure the
    Evan> computer is completely hung.  Frozen screen, keyboard not
    Evan> responding (numlock, capslock..).

Since you are running an SMP system, the problem with
the network (or maybe some other) driver could be that it is not SMP save.

    Evan> I think that it might be a kernel panic, and looked at
    Evan> /var/log/syslog, /var/log/messages and others, and couldn't
    Evan> find anything related.

Usually there is no time to flush the logs before it freezes completely.


    Evan> Here is some info about the system

    Evan> dmesg: http://obakechan.net/e7Rj5pWm0N/dmesg /proc/cpuinfo:
    Evan> http://obakechan.net/e7Rj5pWm0N/cpuinfo

    Evan> uname -a Linux sim2 2.6.15-27-amd64-generic #1 SMP PREEMPT
    Evan> Sat Sep 16 01:50:50 UTC 2006 x86_64 GNU/Linux

    Evan> The kernel is a stock kernel that comes with ubuntu dapper.

    Evan> Do you have any suggestion as to how to investigate the
    Evan> problem?

You could try booting a kernel without SMP to see if that is the problem. 
If possible you can try a different NIC.

If you get any clue as to what part of the system or which driver could
be responsible you can contact the maintainer of that driver.


Hope that helps

Marcus

-- 
/--------------------------------------------------------------------\
| Dr. Marcus O.C. Metzler        |                                   |
| mocm@example.com            | http://www.metzlerbros.de/        |
\--------------------------------------------------------------------/
 |>>>             Quis custodiet ipsos custodes                 <<<|