Firefly and PC GAMESS-related discussion club


 
Learn how to ask questions correctly  
 
 
We are NATO-free zone
 



Re^7: Error 0x00000040 (Unable to create/chdir to temporary working directory) Firefly 7.1.G/Linux/MPICH2

Alex Granovsky
gran@classic.chem.msu.su


Hi,

You need to apply the patch.

See

http://classic.chem.msu.su/cgi-bin/ceilidh.exe/gran/gamess/forum/?C35e9ea936bHW-7660-1225-00.htm

and multiple other posts to this forum concerning signal 7 issue.

Kind regards,
Alex Granovsky


On Fri Oct 19 '12 8:37pm, Denis Zavelev wrote
---------------------------------------------
>Hi,

>Thanks! This works...
>But Firefly exits with error.

>Here's output, attached goes log file:

>computing BENCH13.inp

>TID 4079 caught signal 7, exiting.

>Dump of registers follows

>eax :: 0x000105a4, edx :: 0xff99f470
>ecx :: 0xff982244, ebx :: 0xff99f250
>esi :: 0xf5c12498, edi :: 0xff99f2dc
>ebp :: 0xff99f298, esp :: 0xff99f23c
>eip :: 0xf5bf4177, eflags :: 0x00210202

>cs  :: 0x0023
>ds  :: 0x002b
>es  :: 0x002b
>ss  :: 0x002b
>fs  :: 0x00d7
>gs  :: 0x0063

>Stack backtrace

>esp :: 0xff99f298, ebp :: 0x00000002, eip :: 0xf5bfd5fc

>TID 4078 caught signal 7, exiting.

>Dump of registers follows

>eax :: 0x0000f5a4, edx :: 0xff9ec670
>ecx :: 0xff9ce444, ebx :: 0xff9ec450
>esi :: 0xf5c22498, edi :: 0xff9ec4dc
>ebp :: 0xff9ec498, esp :: 0xff9ec43c
>eip :: 0xf5c04177, eflags :: 0x00210202

>cs  :: 0x0023
>ds  :: 0x002b
>es  :: 0x002b
>ss  :: 0x002b
>fs  :: 0x00d7
>gs  :: 0x0063

>Stack backtrace

>esp :: 0xff9ec498, ebp :: 0x00000002, eip :: 0xf5c0d5fc

>===================================================================================
>=   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
>=   EXIT CODE: 9
>=   CLEANING UP REMAINING PROCESSES
>=   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
>===================================================================================
>YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Killed (signal 9)
>This typically refers to a problem with your application.
>Please see the FAQ page for debugging suggestions
>
>
>
>On Fri Oct 19 '12 8:08pm, Alex Granovsky wrote
>----------------------------------------------
>>Hi,

>>then try the following:

>>

>>mpiexec -n 2 ./firefly -r -f -p -i $inputfile -o $outfile.out -t /home/metalian/Quant/firefly_71g_linux_mpich2_p2/scratch  -b /home/metalian/Quant/firefly_71g_linux_mpich2_p2/metalian.lib
>>

>>Kind regards,
>>Alex
>>
>>
>>On Fri Oct 19 '12 7:16pm, Denis Zavelev wrote
>>---------------------------------------------
>>>Hello again, Alex!

>>>Here it goes:

>>>     linux-gate.so.1 =>  (0xf7715000)
>>>     libm.so.6 => /lib/i386-linux-gnu/libm.so.6 (0xf76c7000)
>>>     libmpich.so.1.1 => /home/metalian/mpich2-install/lib/libmpich.so.1.1 (0xf7467000)
>>>     libpthread.so.0 => /lib/i386-linux-gnu/libpthread.so.0 (0xf744b000)
>>>     librt.so.1 => /lib/i386-linux-gnu/librt.so.1 (0xf7442000)
>>>     libc.so.6 => /lib/i386-linux-gnu/libc.so.6 (0xf7298000)
>>>     /lib/ld-linux.so.2 (0xf7716000)
>>>     libmpl.so.1 => /home/metalian/mpich2-install/lib/libmpl.so.1 (0xf7292000)
>>>     libgcc_s.so.1 => /lib/i386-linux-gnu/libgcc_s.so.1 (0xf7274000)
>>>
>>>
>>>
>>>On Fri Oct 19 '12 7:08pm, Alex Granovsky wrote
>>>----------------------------------------------
>>>>Hello,

>>>>could you please post here the result of:

>>>>
>>>>ldd ./firefly
>>>>

>>>>Kind regards,
>>>>Alex
>>>>
>>>>
>>>>On Fri Oct 19 '12 6:49pm, Denis Zavelev wrote
>>>>---------------------------------------------
>>>>>Hello, Alex!

>>>>>Thanks for the answer!

>>>>>>my two points are:
>>>>>>1. It seems you are using mpich2-linked Firefly version but is
>>>>>>trying to run it as if it were mpich1 binaries.
>>>>>What's the difference? Should I execute mpirun with firefly as argument instead of just firefly?

>>>>>>2. You cannot not use relative paths with Firefly v. 7.1.G
>>>>>>(you can with Firefly v. 8.0.0 onward).
>>>>>I've changed the command:
>>>>>./firefly -f -p -o $inpfile.out -t /home/metalian/Quant/firefly_71g_linux_mpich2_p2/scratch -p4pg /home/metalian/Quant/firefly_71g_linux_mpich2_p2/procgrp -b /home/metalian/Quant/firefly_71g_linux_mpich2_p2/metalian.lib

>>>>>But it hasn't changed anything.
>>>>>
>>>>>
>>>>>
>>>>>>On Fri Oct 19 '12 3:52pm, Denis Zavelev wrote
>>>>>>---------------------------------------------
>>>>>>>Hello, all!

>>>>>>>After reinstalling Ubuntu 12.04 on one of my PCs (with Athlon 64 x2 CPU), I've decided to use it again for computations. I've successfully (after several efforts) built MPICH2 v. 1.5 from sources (I had to do it as my Ubuntu is 64-bit and there were some tricks).

>>>>>>>But when I tried to launch firefly, I've got the following error instead of output file:

>>>>>>> FATAL ERROR INITIALIZING FIREFLY! CHECK YOUR COMMAND LINE OPTIONS!

>>>>>>> ON MASTER NODE, ERROR CODE IS : 0x00000040
>>>>>>>    Unable to create/chdir to  temporary working directory

>>>>>>>Here's a fragment of my firefly executing script:

>>>>>>>     echo "computing $inpfile.inp"
>>>>>>>      cp -f $inpfile.inp ./input
>>>>>>>      cp -f $inpfile.inp scratch.0/input
>>>>>>>      ./firefly -f -p -o $inpfile.out -ex . -t scratch -p4pg ./procgrp -b ./metalian.lib

>>>>>>>I've randomly taken (with edition of extensions!) BENCH13.inp and also EXAM31.inp from samples.zip.
>>>>>>>Yes, I understand that -f, -p and -b keys for such computation are not needed, but this script is being successfully used on my PCs, so there can't be the reason of error.

>>>>>>>The script is executed from Firefly working directory under user.
>>>>>>>procgrp file contains the only string "local 1"
>>>>>>>Working directory contains scratch.0 and scratch.1 those are symlinks to temporary directories on /mnt/sata2 and /mnt/sata3. Both temporary directories contain all the files needed (Actually working and temporary directories remained unaffected since I stopped to use this PC for computations and I haven't formatted these partitions when installing Ubuntu but of course I had to make "chown -R" to get all the rights after installing).
>>>>>>>All the directories and files there are owned by the same user account and (I've checked this manually) allow copying files to them. All of the directories have enough free space (about 13Gb in working directory and about 450G in each of temporary directories).

>>>>>>>What have I done wrong? Any ideas?


[ Previous ] [ Next ] [ Index ]           Fri Oct 19 '12 9:04pm
[ Reply ] [ Edit ] [ Delete ]           This message read 1128 times