[mpich-discuss] MPICH2 Errors running MCNPX-SMPDU connectfailed?
Jayesh Krishna
jayesh at mcs.anl.gov
Tue May 8 19:37:23 CDT 2012
There are two versions of the program (interactive, icpi.c, and non-interactive, cpi.c). Don't compare the outputs to determine if your program is running correctly.
-Jayesh
----- Original Message -----
From: "Ju JiaJia" <jujj603 at gmail.com>
To: mpich-discuss at mcs.anl.gov
Sent: Tuesday, May 8, 2012 6:42:24 PM
Subject: Re: [mpich-discuss] MPICH2 Errors running MCNPX-SMPDU connectfailed?
I don't know the differences between MPICH2 packages on linux and windows. The cpi test program on linux, has one line of code, like this:
fprintf(stdout,"Process %d of %d is on %s\n",
myid, numprocs, processor_name);
On Tue, May 8, 2012 at 11:33 PM, Abrego, Celestino < Celestino.Abrego at parsons.com > wrote:
Thanks for the tip but –v does not come up as an option for mpiexec. Do you think the testcpi is not running correctly? When it executes my cpu usage goes up to around 80% which is exactly where it should be.
C:\Documents and Settings\p0038192>"c:\Program files\MPICH2\bin\mpiexec.exe" -help2
All options to mpiexec:
-n x
-np x
launch x processes
-localonly x
-n x -localonly
launch x processes on the local machine
-machinefile filename
use a file to list the names of machines to launch on
-host hostname
-hosts n host1 host2 ... hostn
-hosts n host1 m1 host2 m2 ... hostn mn
launch on the specified hosts
In the second version the number of processes = m1 + m2 + ... + mn
-binding proc_binding_scheme
Set the proc binding for each of the launched processes to a single core.
Currently "auto" and "user" are supported as the proc_binding_schemes
-map drive:\\host\share
map a drive on all the nodes
this mapping will be removed when the processes exit
-mapall
map all of the current network drives
this mapping will be removed when the processes exit
(Available currently only on windows)
-dir drive:\my\working\directory
-wdir /my/working/directory
launch processes in the specified directory
-env var val
set environment variable before launching the processes
-logon
prompt for user account and password
-pwdfile filename
read the account and password from the file specified
put the account on the first line and the password on the second
-nompi
launch processes without the mpi startup mechanism
-nopopup_debug
disable the system popup dialog if the process crashes
-exitcodes
print the process exit codes when each process exits.
-noprompt
prevent mpiexec from prompting for user credentials.
-priority class[:level]
set the process startup priority class and optionally level.
class = 0,1,2,3,4 = idle, below, normal, above, high
level = 0,1,2,3,4,5 = idle, lowest, below, normal, above, highest
the default is -priority 1:3
-localroot
launch the root process directly from mpiexec if the host is local.
(This allows the root process to create windows and be debugged.)
-port port
-p port
specify the port that smpd is listening on.
-phrase passphrase
specify the passphrase to authenticate connections to smpd with.
-smpdfile filename
specify the file where the smpd options are stored including the passphrase.
-path search_path
search path for executable, ; separated
-register [-user n]
encrypt a user name and password to the Windows registry.
optionally specify a user slot index
-remove [-user n]
delete the encrypted credentials from the Windows registry.
If no user index is specified then all entries are removed.
-validate [-user n] [-host hostname]
validate the encrypted credentials for the current or specified host.
A specific user index can be specified otherwise index 0 is the default.
-user n
use the registered user credentials from slot n to launch the job.
-timeout seconds
timeout for the job.
-plaintext
don't encrypt the data on the wire.
-delegate
use passwordless delegation to launch processes
-impersonate
use passwordless authentication to launch processes
-add_job <job_name> <domain\user>
-add_job <job_name> <domain\user> -host <hostname>
add a job key for the specified domain user on the local or specified host
requires administrator privileges
-remove_job <name>
-remove_job <name> -host <hostname>
remove a job key from the local or specified host
requires administrator privileges
-associate_job <name>
-associate_job <name> -host <hostname>
associate the current user's token with the specified job on the local or spec
ified host
-job <name>
launch the processes in the context of the specified job
-whomai
print the current user name
-l
prefix output with the process number. (This option is a lowercase L not the n
umber one)
C:\Documents and Settings\p0038192>
Celestino Abrego
From: mpich-discuss-bounces at mcs.anl.gov [mailto: mpich-discuss-bounces at mcs.anl.gov ] On Behalf Of Ju JiaJia
Sent: Monday, May 07, 2012 8:31 PM
To: mpich-discuss at mcs.anl.gov
Subject: Re: [mpich-discuss] MPICH2 Errors running MCNPX-SMPDU connectfailed?
mpiexec has an option -verbose, abbr -v. You can see it with mpiexec -help2.
I'm using linux, it should be the same on windows.
On Tue, May 8, 2012 at 8:59 AM, Abrego, Celestino < Celestino.Abrego at parsons.com > wrote:
Thanks for the info. What do you mean by run with –v? It does not appear to be an option for either cpi or mpiexe.
Celestino Abrego
From: mpich-discuss-bounces at mcs.anl.gov [mailto: mpich-discuss-bounces at mcs.anl.gov ] On Behalf Of Ju JiaJia
Sent: Monday, May 07, 2012 7:00 PM
To: mpich-discuss at mcs.anl.gov
Subject: Re: [mpich-discuss] MPICH2 Errors running MCNPX-SMPDU connectfailed?
I don't think your testcpi is giving the right answer. there should be 20 processes running and process info should be printed, something like this:
Process 0 of 1 is on m1.
Please run with -v to get more informations.
On Tue, May 8, 2012 at 5:53 AM, Jayesh Krishna < jayesh at mcs.anl.gov > wrote:
Hi,
You might also want to check with the mcnpx developers if they have seen the error message (You should mention that you can run cpi.exe with the same command).
Regards,
Jayesh
----- Original Message -----
From: "Celestino Abrego" < Celestino.Abrego at parsons.com >
To: mpich-discuss at mcs.anl.gov
Sent: Monday, May 7, 2012 4:40:11 PM
Subject: Re: [mpich-discuss] MPICH2 Errors running MCNPX-SMPDU connect failed?
Jayesh
-It is a 32 bit Windows XP Pro system and I am only trying to run on the
local system.
-I read through the manual and found that MPICH2 only uses TCP protocols
so I briefly allowed all TCP ports and IP addresses. But I still had the
same errors when I executed MCNPX.
-CPI and MCNPX have the same privileges, except I could not modify
MCNPX's by click on the check boxes- they were grayed out. But the same
boxes were checked for each group or user name. At minimum all users
could read & execute applications.
-My firewall only allows me to make exceptions for particular protocols
like TCP, UDP, etc., so I could not explicitly allow MCNPX exe file
through firewall.
Just to give you a bit more information in the hopes that it may trigger
new ideas:
-Desktop computer is setup with to be on a the parsons.com Domain but is
not actively on the "Parsons" network (since I work remotely). I
mention this because the error I receive is "PMI_ConnectToHost failed:
unable to post a connect to PBRFPNM1.Parsons.com "
-Computer receives internet connection from a router with its own
firewall that I have assumed does not affect what I am doing.
-I am getting a new firewall program tomorrow!
Anything else I should try?
Celestino Abrego
-----Original Message-----
From: mpich-discuss-bounces at mcs.anl.gov
[mailto: mpich-discuss-bounces at mcs.anl.gov ] On Behalf Of Jayesh Krishna
Sent: Monday, May 07, 2012 1:58 PM
To: Abrego, Celestino
Cc: mpich-discuss at mcs.anl.gov
Subject: Re: [mpich-discuss] MPICH2 Errors running MCNPX-SMPDU connect
failed?
Hi,
# Is your machine 32-bit or 64-bit (You seem to be running a 32-bit
version of mcnp)?
# Try adding an exception for the executable
(C:\MCNP\mcnp5\bin\MCNPX27mpi_Win32.exe) in your firewall.
# Do the executables "C:\Documents and
Settings\p0038192\Desktop\Testcpi" and
"C:\MCNP\mcnp5\bin\MCNPX27mpi_Win32.exe" have different privileges?
# For more information on how MPICH2 can work with firewalls see Section
9.5 of the Windows developer's guide
( http://www.mcs.anl.gov/research/projects/mpich2/documentation/index.php
?s=docs).
Regards,
Jayesh
----- Original Message -----
From: "Celestino Abrego" < Celestino.Abrego at parsons.com >
To: "Jayesh Krishna" < jayesh at mcs.anl.gov >, mpich-discuss at mcs.anl.gov
Sent: Monday, May 7, 2012 12:05:04 PM
Subject: RE: [mpich-discuss] MPICH2 Errors running MCNPX-SMPDU connect
failed?
I have not been able to disable firewalls yet but here are the results
of your other requests. Commands give similar results. What exactly
needs to go through the firewalls? I can add exceptions and modify
both of the firewalls. I just have not figured out how to temporarily
disable them for testings. I have request with our IT department for
that. Any other ideas? Thanks for the fast response!!!
c:\mcnp\mcnp5\bin>TestMCNPXrunMPI_27
c:\mcnp\mcnp5\bin>rem ++++++Validates MPICH2 USERNAME AND PW
registration+++++++
+
c:\mcnp\mcnp5\bin>"C:\Program Files\MPICH2\bin\mpiexec.exe" -validate
SUCCESS
c:\mcnp\mcnp5\bin>rem +++++Renamed MPICH2 example Cpi and moved to
Desktop for T
esting++++++
c:\mcnp\mcnp5\bin>"C:\Program Files\MPICH2\bin\mpiexec.exe" -hosts 1
192.168.1.3
20 C:\Documents and Settings\p0038192\Desktop\Testcpi Enter the number
of intervals: (0 quits) 10000000000000000000000000000000 pi is
approximately -0.0000000000000000, Error is 3.1415926535897931 wall
clock time = 0.000069 Enter the number of intervals: (0 quits) 0
c:\mcnp\mcnp5\bin>rem ++++++++Test MCNP Run with HOST name using MCNPX
Win32_MPI
++++++++
c:\mcnp\mcnp5\bin>"C:\Program Files\MPICH2\bin\mpiexec.exe" -hosts 1
192.168.1.3
2 C:\MCNP\mcnp5\bin\MCNPX27mpi_Win32.exe i=beam0 [01:396]..ERROR:Error
creating a socket, A system call that should never fail ha s failed.
(10107) [01:396]..ERROR:Error creating a socket SMPDU_Sock_post_connect
failed.
[1] PMI_ConnectToHost failed: unable to post a connect to
PBRFPNM1.Parsons.com 1
92.168.1.3 192.168.79.1 192.168.102.1 :4345, error: Unknown error class,
error s
tack:
(unknown)(): Invalid buffer pointer
uPMI_ConnectToHost returning PMI_FAIL
[1] PMI_Init failed.
Fatal error in MPI_Init: Other MPI error, error stack:
MPIR_Init_thread(388):
MPID_Init(107).......: channel initialization failed
MPID_Init(367).......: PMI_Init returned -1 [01:7480]..ERROR:Error
creating a socket, A system call that should never fail h as failed.
(10107) [01:7480]..ERROR:Error creating a socket SMPDU_Sock_post_connect
failed.
[0] PMI_ConnectToHost failed: unable to post a connect to
PBRFPNM1.Parsons.com 1
92.168.1.3 192.168.79.1 192.168.102.1 :4352, error: Unknown error class,
error s
tack:
(unknown)(): Invalid buffer pointer
uPMI_ConnectToHost returning PMI_FAIL
[0] PMI_Init failed.
Fatal error in MPI_Init: Other MPI error, error stack:
MPIR_Init_thread(388):
MPID_Init(107).......: channel initialization failed
MPID_Init(367).......: PMI_Init returned -1
c:\mcnp\mcnp5\bin>pause
Press any key to continue . . .
c:\mcnp\mcnp5\bin>^A
Celestino Abrego
-----Original Message-----
From: Jayesh Krishna [mailto: jayesh at mcs.anl.gov ]
Sent: Monday, May 07, 2012 11:50 AM
To: mpich-discuss at mcs.anl.gov
Cc: Abrego, Celestino
Subject: Re: [mpich-discuss] MPICH2 Errors running MCNPX-SMPDU connect
failed?
Hi,
Some suggestions,
# Make sure that any firewall running on the machine is turned off # In
the step where Testcpi is run (c:\mcnp\mcnp5\bin>"C:\Program
Files\MPICH2\bin\mpiexec.exe" -hosts 1 PBRFPNM1 20 C:\Documents and
Settings\p0038192\Desktop\Testcpi ), try specifying a "number of
intervals" other than 0 and see if it works.
# Try specifying the ipaddress of "PBRFPNM1" in the mpiexec command
(instead of "PBRFPNM1").
Let us know the results.
Regards,
Jayesh
----- Original Message -----
From: "Celestino Abrego" < Celestino.Abrego at parsons.com >
To: mpich-discuss at mcs.anl.gov
Sent: Monday, May 7, 2012 11:37:18 AM
Subject: [mpich-discuss] MPICH2 Errors running MCNPX-SMPDU connect
failed?
I am in the process of trying to run a version of software that has is
widely distributed and was compiled to be run with MPICH2. I have been
running around in circles trying to figure out why I am not having
success running the software with multiple processors and I though this
group might be able to help me. Below are the results of a series of
commands that I copied and pasted from the DOS window. Everything seems
to be working as expected until I try to execute the MCNPX software. CPI
examples ramps up to nearly 100% CPU usage.
I have tried uninstalling and reinstalling MPICH2, using an admin
command prompt. SMPD and MPIEXEC.exe have been added as exceptions
through the windows firewall although I am not sure how much windows is
actually blocking because my company uses Trend MICRO Officescan and it
has its own firewall. I have found that adding SMPD and MPIEXEC as
exceptions through the windows firewall does not have an effect on the
errors I am receiving. I am running low on ideas right now, and any
help?
Here are the results from the command prompt. Thanks in advanced!
________________________________________________________________________
_______
c:\mcnp\mcnp5\bin>testmcnpxrunmpi_27
c:\mcnp\mcnp5\bin>rem ++++++Validates MPICH2 USERNAME AND PW
registration+++++++
+
c:\mcnp\mcnp5\bin>"C:\Program Files\MPICH2\bin\mpiexec.exe" -validate
SUCCESS
c:\mcnp\mcnp5\bin>rem +++++Renamed MPICH2 example Cpi and moved to
Desktop for T
esting++++++
c:\mcnp\mcnp5\bin>"C:\Program Files\MPICH2\bin\mpiexec.exe" -hosts 1
PBRFPNM1 20
C:\Documents and Settings\p0038192\Desktop\Testcpi
Enter the number of intervals: (0 quits) 0
c:\mcnp\mcnp5\bin>rem ++++++++Test MCNP Run with HOST name using MCNPX
Win32_MPI
++++++++
c:\mcnp\mcnp5\bin>"C:\Program Files\MPICH2\bin\mpiexec.exe" -hosts 1
PBRFPNM1 2
C:\MCNP\mcnp5\bin\MCNPX27mpi_Win32.exe i=beam0
[01:7788]..ERROR:Error creating a socket, A system call that should
never fail h
as failed. (10107)
[01:7788]..ERROR:Error creating a socket
SMPDU_Sock_post_connect failed.
[1] PMI_ConnectToHost failed: unable to post a connect to
PBRFPNM1.Parsons.com 1
92.168.1.3 192.168.79.1 192.168.102.1 :3130, error: Unknown error class,
error s
tack:
(unknown)(): Invalid buffer pointer
uPMI_ConnectToHost returning PMI_FAIL
[1] PMI_Init failed.
Fatal error in MPI_Init: Other MPI error, error stack:
MPIR_Init_thread(388):
MPID_Init(107).......: channel initialization failed
MPID_Init(367).......: PMI_Init returned -1
[01:484]..ERROR:Error creating a socket, A system call that should never
fail ha
s failed. (10107)
[01:484]..ERROR:Error creating a socket
SMPDU_Sock_post_connect failed.
[0] PMI_ConnectToHost failed: unable to post a connect to
PBRFPNM1.Parsons.com 1
92.168.1.3 192.168.79.1 192.168.102.1 :3137, error: Unknown error class,
error s
tack:
(unknown)(): Invalid buffer pointer
uPMI_ConnectToHost returning PMI_FAIL
[0] PMI_Init failed.
Fatal error in MPI_Init: Other MPI error, error stack:
MPIR_Init_thread(388):
MPID_Init(107).......: channel initialization failed
MPID_Init(367).......: PMI_Init returned -1
c:\mcnp\mcnp5\bin>pause
Press any key to continue . . .
c:\mcnp\mcnp5\bin>path
PATH=C:\MCNP\mcnp5\bin;C:\Program Files\Common Files\Microsoft
Shared\Windows Li
ve;C:\WINDOWS\system32;C:\WINDOWS;C:\WINDOWS\System32\Wbem;C:\PROGRA~1\I
BM\CLIEN
T~1;C:\PROGRA~1\IBM\CLIENT~1\Shared;C:\PROGRA~1\IBM\CLIENT~1\Emulator;C:
\Program
Files\Common Files\Roxio
Shared\DLLShared\;C:\WINDOWS\system32\WindowsPowerShel
l\v1.0;C:\Program Files\Windows Imaging\;C:\MCNP\bin;C:\Program
Files\MATLAB\R20
10b\bin;c:\Program Files\Microsoft SQL Server\90\Tools\binn\;C:\Program
Files\Qu
ickTime\QTSystem\;C:\Program Files\7-Zip;C:\Program
Files\MPICH2\bin;C:\Program
Files\Common Files\Microsoft Shared\Windows Live;C:\Program Files\7-Zip
c:\mcnp\mcnp5\bin>ipconfig
Windows IP Configuration
Ethernet adapter VMware Network Adapter VMnet8:
Connection-specific DNS Suffix . :
IP Address. . . . . . . . . . . . : 192.168.102.1
Subnet Mask . . . . . . . . . . . : 255.255.255.0
Default Gateway . . . . . . . . . :
Ethernet adapter VMware Network Adapter VMnet1:
Connection-specific DNS Suffix . :
IP Address. . . . . . . . . . . . : 192.168.79.1
Subnet Mask . . . . . . . . . . . : 255.255.255.0
Default Gateway . . . . . . . . . :
Ethernet adapter Local Area Connection:
Connection-specific DNS Suffix . :
IP Address. . . . . . . . . . . . : 192.168.1.3
Subnet Mask . . . . . . . . . . . : 255.255.255.0
Default Gateway . . . . . . . . . : 192.168.1.1
Celestino Abrego
_______________________________________________
mpich-discuss mailing list mpich-discuss at mcs.anl.gov
To manage subscription options or unsubscribe:
https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
_______________________________________________
mpich-discuss mailing list mpich-discuss at mcs.anl.gov
To manage subscription options or unsubscribe:
https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
_______________________________________________
mpich-discuss mailing list mpich-discuss at mcs.anl.gov
To manage subscription options or unsubscribe:
https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
_______________________________________________
mpich-discuss mailing list mpich-discuss at mcs.anl.gov
To manage subscription options or unsubscribe:
https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
_______________________________________________
mpich-discuss mailing list mpich-discuss at mcs.anl.gov
To manage subscription options or unsubscribe:
https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
_______________________________________________
mpich-discuss mailing list mpich-discuss at mcs.anl.gov
To manage subscription options or unsubscribe:
https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
_______________________________________________
mpich-discuss mailing list mpich-discuss at mcs.anl.gov
To manage subscription options or unsubscribe:
https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
More information about the mpich-discuss
mailing list