running the example

Hello Mohsen.

mpipython, not bingle.py, should be the executable. I was about to suggest that you check if you have mpipython, but then i noticed that i am now getting precisely your error on my machine, both from the ESyS-Particle package and from the installation that i built from source. I have been running a recent revision of ESyS-Particle on our supercomputer, but i have not used it recently on my desktop computer. Maybe there is a third-party issue here. My desktop runs Ubuntu 12.04. What is on your virtual box?

Regards,

Vince

Hi Mohsen and Vince,

Perhaps you could try running the test script using the following command:
mpirun -np 2 esysparticle bingle.py

or include the full path to esysparticle executable:
mpirun -np 2 /usr/local/bin/esysparticle bingle.py

Cheers,

Dion

Hi Dion,

It gives same error message again. I checked the directory, In /usr/local/bin there is just one executable file, "povray".
Sorry i am new at Linux so am just wondering how could i be sure i installed the esysparticle correctly?

I am using Oracle VM Virtual box V 4.2 at the moment.

Thanks,

Hi,

I tried to install the esys-particle again. when i run make, after lots of lines at the end i noticed the following lines.

g++: internal compiler error: Killed (program cc1plus)
Please submit a full bug report,
with preprocessed source if appropriate.
See <file:///usr/share/doc/gcc-4.6/README.Bugs> for instructions.
make[2]: *** [SubLatticeControler.lo] Error 1
make[2]: Leaving directory `/home/mohsen/esys-particle/esys-particle/Parallel'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/mohsen/esys-particle/esys-particle'
make: *** [all] Error 2

I just wondering what is the problem.

Thanks, Dion, for the suggestion. I tried both of these but with the same message. And that is when i realized that i had inadvertently omitted the number of processes. I had used:

  mpirun -np `which esysparticle` bingle.py

instead of:

  mpirun -np 2 `which esysparticle` bingle.py

With the correct command, bingle.py runs correctly.

Mohsen, can you advise whether or not this resolves your problem?

Vince

Hi Mohsen,

Take a look at the last few lines when you type "dmesg" after you hit that compile error.
You may need to assign more memory and swap space to your virtual machine.

Regards,
Will

Hi Mohsen.

If your build has failed during "make", you won't have an installation of the code, so there will be no "esysparticle" executable. At this stage i have no idea what caused the compiler error. Some of the lines before "g++: internal compiler error: Killed (program cc1plus)" may shed some light. Otherwise, this may be a problem that the gcc compiler developers are better equipped to handle.

Vince

Hi Will

These are last few lines when typing "dmesg".

[ 107.329680] Console: switching to colour frame buffer device 80x30
[ 107.329696] fb0: VESA VGA frame buffer device
[ 112.263508] Bluetooth: Core ver 2.16
[ 112.263533] NET: Registered protocol family 31
[ 112.263536] Bluetooth: HCI device and connection manager initialized
[ 112.263538] Bluetooth: HCI socket layer initialized
[ 112.263540] Bluetooth: L2CAP socket layer initialized
[ 112.264416] Bluetooth: SCO socket layer initialized
[ 112.917946] Bluetooth: RFCOMM TTY layer initialized
[ 112.917952] Bluetooth: RFCOMM socket layer initialized
[ 112.917954] Bluetooth: RFCOMM ver 1.11
[ 113.406906] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
[ 113.406910] Bluetooth: BNEP filters: protocol multicast
[ 113.502548] type=1400 audit(1349836428.034:8): apparmor="STATUS" operation="profile_load" name="/usr/lib/cups/backend/cups-pdf" pid=799 comm="apparmor_parser"
[ 113.503336] type=1400 audit(1349836428.034:9): apparmor="STATUS" operation="profile_load" name="/usr/sbin/cupsd" pid=799 comm="apparmor_parser"
[ 113.947732] type=1400 audit(1349836428.478:10): apparmor="STATUS" operation="profile_load" name="/usr/lib/lightdm/lightdm/lightdm-guest-session-wrapper" pid=837 comm="apparmor_parser"
[ 113.950969] type=1400 audit(1349836428.482:11): apparmor="STATUS" operation="profile_replace" name="/sbin/dhclient" pid=844 comm="apparmor_parser"
[ 113.951524] type=1400 audit(1349836428.482:12): apparmor="STATUS" operation="profile_replace" name="/usr/lib/NetworkManager/nm-dhcp-client.action" pid=844 comm="apparmor_parser"
[ 113.951841] type=1400 audit(1349836428.482:13): apparmor="STATUS" operation="profile_replace" name="/usr/lib/connman/scripts/dhclient-script" pid=844 comm="apparmor_parser"
[ 115.412048] ADDRCONF(NETDEV_UP): eth0: link is not ready
[ 115.412739] e1000: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
[ 115.413975] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[ 115.790586] type=1400 audit(1349836430.326:14): apparmor="STATUS" operation="profile_load" name="/usr/bin/evince" pid=845 comm="apparmor_parser"
[ 115.797667] type=1400 audit(1349836430.334:15): apparmor="STATUS" operation="profile_load" name="/usr/bin/evince//launchpad_integration" pid=845 comm="apparmor_parser"
[ 115.799695] type=1400 audit(1349836430.334:16): apparmor="STATUS" operation="profile_load" name="/usr/bin/evince//sanitized_helper" pid=845 comm="apparmor_parser"
[ 115.800528] type=1400 audit(1349836430.334:17): apparmor="STATUS" operation="profile_load" name="/usr/bin/evince-previewer" pid=845 comm="apparmor_parser"
[ 126.144050] eth0: no IPv6 routers present
[ 127.446262] [drm] Initialized drm 1.1.0 20060810
[ 127.528697] pci 0000:00:02.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
[ 127.528739] pci 0000:00:02.0: setting latency timer to 64
[ 127.529676] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
[ 127.529680] [drm] No driver support for vblank timestamp query.
[ 127.529684] [drm] Initialized vboxvideo 1.0.0 20090303 for 0000:00:02.0 on minor 0

I didn't get where is the error. I increased the ram of the virtual machine as well.

Thanks,
Mohsen

Thanks Will,

I guess the problem as you mentioned was the memory. I increased the memeory more and i was able to install the package.

Vince,

As Will adviced, i added my virtual machine memeory and i was able to install the package. Then i used your suggested code to run the example. It works correctly.

Thanks for your help.

Hi Mohsen,

The error i was expecting dmesg to write out was something along these lines of:

[ 1377.575785] Out of memory: Kill process 12305 (cc1plus) score 905 or sacrifice child
[ 1377.575800] Killed process 12305 (cc1plus) total-vm:579928kB, anon-rss:546144kB, file-rss:0kB

Which should have been written to the log at the point of compiler failure, signalling that the machine had run out of memory. Anyway, good to know you were finally able to compile the code.

Regards,
Will