[prev in list] [next in list] [prev in thread] [next in thread] 

List:       npaci-rocks-discussion
Subject:    [Rocks-Discuss] mpich question on 5.1 cluster
From:       János_Löbb <janos.lobb () yale ! edu>
Date:       2011-01-26 19:54:58
Message-ID: 70C7A09A-83EE-4D2F-BA81-21F3D0A6EAC6 () yale ! edu
[Download RAW message or body]

Hi,

I installed the hpc roll on my test cluster and started testing with the examples.  \
Cluster-fork runs fine.  When I tried mpirun as described here:

http://www.rocksclusters.org/roll-documentation/hpc/5.1/using-mpirun-ethernet.html

as a non-root user, I received this as the output:

<nyissz>
[janos@bml0042 mpich-test]$ /opt/mpich/gnu/bin/mpirun -nolocal -np 2 -machinefile \
                $HOME/machines $HOME/mpich-test/mpi-ring
Warning: Permanently added 'compute-0-0' (RSA) to the list of known hosts.
Warning: untrusted X11 forwarding setup failed: xauth key data not generated
Warning: No xauth data; using fake authentication data for X11 forwarding.
/usr/bin/xauth:  creating new authority file /home/janos/.Xauthority
Warning: Permanently added 'compute-0-1' (RSA) to the list of known hosts.
Warning: untrusted X11 forwarding setup failed: xauth key data not generated
Warning: No xauth data; using fake authentication data for X11 forwarding.
Process 0 on compute-0-0.local
Process 1 on compute-0-1.local
p1_12905:  p4_error: interrupt SIGSEGV: 11
p0_12869:  p4_error: interrupt SIGSEGV: 11
Killed by signal 2.
p0_12869: (8.453125) net_send: could not write to fd=4, errno = 32
<nyassz>

One post in the Internet recommends to run it with the -dbg switch.  When I do that \
the output is this: <nyissz>
[janos@bml0042 mpich-test]$ /opt/mpich/gnu/bin/mpirun -dbg ./err.out -nolocal -np 2 \
                -machinefile $HOME/machines $HOME/mpich-test/mpi-ring
Warning: Command line arguments for program should be given
after the program name.  Assuming that ./err.out is a
command line argument for the program.
Unrecognized argument -dbg ignored.
Warning: untrusted X11 forwarding setup failed: xauth key data not generated
Warning: No xauth data; using fake authentication data for X11 forwarding.
Warning: untrusted X11 forwarding setup failed: xauth key data not generated
Warning: No xauth data; using fake authentication data for X11 forwarding.
Process 0 on compute-0-0.local
Process 0 on compute-0-0.local:successfully sent (0) bytes to id (1)
Process 1 on compute-0-1.local
Process 1 on compute-0-1.local:successfully sent (0) bytes to id (0)
Process 0 on compute-0-0.local:successfully received (0) bytes from id (1)
Process 1 on compute-0-1.local:successfully received (0) bytes from id (0)
<nyassz>

So, looks like everything is all right.  There is no err.out file.  However if I \
again re-run it without the -dbg switch it reports: <nyissz>
[janos@bml0042 mpich-test]$ /opt/mpich/gnu/bin/mpirun -nolocal -np 2 -machinefile \
                $HOME/machines $HOME/mpich-test/mpi-ring
Warning: untrusted X11 forwarding setup failed: xauth key data not generated
Warning: No xauth data; using fake authentication data for X11 forwarding.
Warning: untrusted X11 forwarding setup failed: xauth key data not generated
Warning: No xauth data; using fake authentication data for X11 forwarding.
Process 0 on compute-0-0.local
Process 1 on compute-0-1.local
p0_13005:  p4_error: interrupt SIGSEGV: 11
Killed by signal 2.
<nyassz>

Any good advise ?

Thanks ahead,

János


[prev in list] [next in list] [prev in thread] [next in thread]