[mpich-discuss] MPICH2 Error

Jayesh Krishna jayesh at mcs.anl.gov
Tue Apr 19 11:27:46 CDT 2011


Hi,
 Why do you want to launch a parallel program from another parallel program ?
 You can run MPMD programs using mpiexec (mpiexec -n 3 A.exe : -n 2 B.exe). If you don't want the programs to be part of the same job why not launch two jobs ?

-Jayesh

----- Original Message -----
From: "dxingya" <dxingya at 163.com>
To: "Jayesh Krishna" <jayesh at mcs.anl.gov>
Sent: Tuesday, April 19, 2011 1:58:41 AM
Subject: Fwd:Re:Fwd: [mpich-discuss] MPICH2 Error

It is not the problem of network connectivity because I only run on a computer with 4 cores. Furthermore, the firewall has been turned down. 
I used MPIEXEC wrapper to run the main parallel program A. In program A, a command was added to call another parallel program B. This is done by command as "C:\Program Files\MPICH2\bin\mpiexec.exe" -n 3 B.exe . However, program B will not be executed. I can not understand what is wrong with mpich2. However, if I use a serial program to call parallel program B, it is executed normally. 

At 2011-04-19 01:56:56,"Jayesh Krishna" < jayesh at mcs.anl.gov > wrote:

>FYI
>
>----- Forwarded Message -----
>From: "Jayesh Krishna" < jayesh at mcs.anl.gov >
>To: mpich-discuss at mcs.anl.gov >Sent: Monday, April 18, 2011 12:56:25 PM
>Subject: Re: [mpich-discuss] MPICH2 Error
>
>Hi,
> This looks like a network connectivity issue between the machines. Can you ping one machine from the other ? If there are any firewalls running on any of the machines try after turning them off on both the machines.
>
>Regards,
>Jayesh
>
>----- Original Message -----
>From: "dxingya" < dxingya at 163.com >
>To: mpich-discuss at mcs.anl.gov >Sent: Monday, April 18, 2011 8:49:10 AM
>Subject: [mpich-discuss] MPICH2 Error
>
>
>
>I found a problem when i used mpich2 in windows. First, I use mpich2 to run a parallel program A. Then, program A will call another parallel program B using mpich2 too. However, program B will not be excuted which show a error code of 10061. Below are the error informations. Could you help me fix this problem? thank you Da Xingya 
>[01:5080]..ERROR:Connect on sock (host=PC2010070315tdx ::1 169.254.133.133 182.128.140.243 , port=2953) failed, exhaused all end points
>SMPDU_Sock_post_connect failed.
>[0] PMI_ConnectToHost failed: unable to post a connect to PC2010070315tdx ::1 169.254.133.133 182.128.140.243 :2953, error: Undefined dynamic error code
>uPMI_ConnectToHost returning PMI_FAIL
>[0] PMI_Init failed.
>Fatal error in MPI_Init: Other MPI error, error stack:
>MPIR_Init_thread(394): Initialization failed
>MPID_Init(103).......: channel initialization failed
>MPID_Init(374).......: PMI_Init returned -1 
>
>
>体验网易邮箱2G超大附件,轻松发优质大电影、大照片,提速3倍! 
>
>
>体验网易邮箱2G超大附件,轻松发优质大电影、大照片,提速3倍! 
>_______________________________________________
>mpich-discuss mailing list
> mpich-discuss at mcs.anl.gov >https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
>_______________________________________________
>mpich-discuss mailing list
> mpich-discuss at mcs.anl.gov >https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss 


体验网易邮箱2G超大附件,轻松发优质大电影、大照片,提速3倍! 


体验网易邮箱2G超大附件,轻松发优质大电影、大照片,提速3倍!


More information about the mpich-discuss mailing list