[mpich-discuss] MPICH2 test not running properly

Alejandro Allievi alejandro.allievi at gmail.com
Fri Mar 2 11:42:43 CST 2012


Robert. Please see below lines starting with capitals. Thanks. Alex

On Thu, Mar 1, 2012 at 7:05 PM, McBroom, Robert C <McBroomRC at oro.doe.gov>wrote:

>  Alex****
>
> ** **
>
> **A)     **For you, just an example of an code to be installed.****
>
> **B)      **When you installed did you right click on the command window
> icon and click on “run as administrator”?  On the configuration settings
> there is a checkbox for hosts, checking it and the next line will let you
> enter your computer name.  There is another set of checkboxes  for
> localonly where you enter “yes”.
>
>
>
YES I DID uses "Run as Administrator". However, when I use the command
"msiexec /I mpich2-1.4.1p1-win-x86-64.msi" a series of windows come up.
They are shown in attached file MPICH2_Install.pdf. As you can see, at no
time there is a configuartion setting option. Also, when I run the MPICH2
examples they do fine, see attached MPICH2_Example.pdf . I do not see any
configuartion settings. Am I missing something here??

> **C)      **“Ctrl alt delete” and click on task manager,  start the
> graphic display of system usage and there is a button for display a window
> for each processor.
>
>
>
OK , there are 8 windows (quad core, two threads each). The INITIAL
percentage of the total CPU for 2/n, 4/n and 8/n are 5.27, 3.44 and 2.39
respectvely. After a little while they all show 12% regardless of n. Are
these results correct??

> **D)     **Typically there has to be a compiler option that points to the
> include files not just the path.  Depends on your compiler.
>
>
>
I USE GCC. When I run my Helloworld_MPI.c example I run into problems, see
attached Helloworld_MPI.pdf . I have run my code in Unix using mpicc and it
runs just fine. Any suggestions?? What compiler do you use to be able to
include mpi.h??

> **E)      **“User Access Control” is doing something to you.
>
>
>
THE MESSAGE  "The page cannot be loaded" is still there when I click on
"Allow a program or feature to go through a firewall". COULD not solve it
even by completely removing Windows Firewall. Have no way to access
"Allowed Programs and Features".

>
>
> ** **
>
> *From:* mpich-discuss-bounces at mcs.anl.gov [mailto:
> mpich-discuss-bounces at mcs.anl.gov] *On Behalf Of *Alejandro Allievi
> *Sent:* Thursday, March 01, 2012 12:59 PM
> *To:* mpich-discuss at mcs.anl.gov
> *Subject:* Re: [mpich-discuss] MPICH2 test not running properly****
>
> ** **
>
> Hi Robert,****
>
>  ****
>
> Thank you for your reply. I'm not accessing a network, I'm only trying to
> run MPI code on my quad core machine. Have some questions regarding your
> attached file****
>
>  ****
>
> A) Points (1) and (2) of your file. I did not do the creation of MCNPDATA
> and MCNP, I just followed instructions in MPICH2 installer's guide, page 20
> (Windows version). However, all example seem to run correctly.****
>
> B) Point (7) of your file. I attached a file with "MPICH2 configuration
> settings", something does not look right with my computer name "MPICH2 not
> installed....". Will that be related to (E) below??****
>
> C) Point (9) of your file, I managed to successfully run the example for
> cpi.exe. I do not understand what you mean by total CPU = 2/total number of
> bars showing in taskbar.****
>
> D) When I compile my own HelloWorld MPI code the compile error message is
> "Helloworld_MPI.c:20:50: fatal error: mpi.h: No such file or directory". My
> path DOES point at MPICH2/bin and MPICH2/include .****
>
> E) My Windows Firewall under Control Panel says "The page cannot be
> loaded" when I click on "Allow a program or feature to go through a
> firewall". I do have Administrator rights.****
>
>  ****
>
> Thanks again for your help.****
>
>  ****
>
> Alex****
>
> On Thu, Mar 1, 2012 at 1:06 PM, McBroom, Robert C <McBroomRC at oro.doe.gov>
> wrote:****
>
> Alex****
>
> There are some tricks involved.  The attached file will give you some
> clues.  The scheme works for both the 32 bit install and the 64 bit with
> the appropriate source file changes.  Note the installation as
> administrator and the use of the command line options on the installer.***
> *
>
>  ****
>
> Robert McBroom****
>
>  ****
>
> *From:* mpich-discuss-bounces at mcs.anl.gov [mailto:
> mpich-discuss-bounces at mcs.anl.gov] *On Behalf Of *Alejandro Allievi
> *Sent:* Wednesday, February 29, 2012 5:26 PM
> *To:* mpich-discuss at mcs.anl.gov
> *Subject:* [mpich-discuss] MPICH2 test not running properly****
>
>  ****
>
> Hi. ****
>
>  ****
>
> I'm new to this list and would appreciate some help with this
> beginners issue. I just installed MPICH2 and am running the recommended
> test. The output is as follows. ****
>
>  ****
>
> D:\mpiexec -n 2 cpi.exe
> Error while connecting to host, No connection could be made because the
> target machine actively refused it. (10061)
> Connect on sock (host=ACE-eggnog.wds.mun.ca, port=8676) failed, exhaused
> all end points
> Unable to connect to 'ACE-eggnog.wds.mun.ca:8676',
> sock error: Error = -1
> ****
>
> Has anybody encountered this problem before??****
>
>  ****
>
> Thank you.****
>
>  ****
>
> Alex****
>
>  ****
>
>
>  ****
>
>  ****
>
>
> _______________________________________________
> 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****
>
>
>
>
> -- ****
>
> *Dr Alejandro Allievi, PhD, PEng*****
>
>  ****
>
> *"Tenez, mon ami, si vous y pensez bien, vous trouverez qu'en tout, notre
> véritable sentiment n'est pas celui dans lequel nous n'avons jamais
> vacillé; mais celui auquel nous sommes le plus habituellement revenus".
> Denis Diderot.*****
>
>  ****
>
> *This email and its attachments are for the sole use of the addressee and
> may contain information which is confidential and/or legally privileged.
> This email and its attachments are subject to copyright and should not be
> partly or wholly reproduced without the consent of the copyright owner. If
> you receive this email in error, please immediately delete it from your
> system and notify the sender by return email.*****
>
> * *****
>
> ** **
>
> _______________________________________________
> 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
>
>


-- 
*Dr Alejandro Allievi, PhD, PEng*

*"Tenez, mon ami, si vous y pensez bien, vous trouverez qu'en tout, notre
véritable sentiment n'est pas celui dans lequel nous n'avons jamais
vacillé; mais celui auquel nous sommes le plus habituellement revenus".
Denis Diderot.*
 **
*This email and its attachments are for the sole use of the addressee and
may contain information which is confidential and/or legally privileged.
This email and its attachments are subject to copyright and should not be
partly or wholly reproduced without the consent of the copyright owner. If
you receive this email in error, please immediately delete it from your
system and notify the sender by return email.*
* *
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/mpich-discuss/attachments/20120302/232a5a12/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MPICH2_Example.pdf
Type: application/pdf
Size: 244423 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/mpich-discuss/attachments/20120302/232a5a12/attachment-0003.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MPICH2_Install.pdf
Type: application/pdf
Size: 304463 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/mpich-discuss/attachments/20120302/232a5a12/attachment-0004.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MPICH2_Helloworld_MPI.pdf
Type: application/pdf
Size: 417125 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/mpich-discuss/attachments/20120302/232a5a12/attachment-0005.pdf>


More information about the mpich-discuss mailing list