[OMPI users] Fwd: problem in cluster

Discussion:

Ankita m

2018-04-25 06:03:05 UTC

While using the open mpi got this error. Can you please tell why so

---------- Forwarded message ---------
From: Ankita m <***@gmail.com>
Date: Tue, 24 Apr 2018, 12:55 pm
Subject: Re: problem in cluster
To: sagar mcp <***@mpcl.in>, Krishna Singh <***@gmail.com>

while using openmpi- 1.4.5 the program ended by showing this error

teamviewer id 565 248 412
password jfu477
my contact number 7830622816

Share your contact details
*Thanks & Regards,*
*Sagar Vijay Naik*
*Sr. Customer Support Engineer*
Address: 17/18 Navketan Estate | Opp. Onida House | Mahakali Caves Road
|Andheri ( East) | Mumbai - 400 093.
Description: download][image: Description: Facebook]
P Please don't print this e-mail unless you really need to.
*Sent:* 24 April 2018 10:52
*Subject:* Fwd: problem in cluster
---------- Forwarded message ----------
Date: Mon, Apr 23, 2018 at 4:18 PM
Subject: problem in cluster
Hello Sir
I am Ankita Maity from Mechanical Department IIT Roorkee.
I am facing problem while submitting a job . all the programs
automatically are going to either queue or status is showing "H". Please
help sir .
my program folder is Home/ankitamed/MarineTurbine1/
Team viewer id password are
565 248 412
jfu477
Regards
Ankita

John Hearns via users

2018-04-25 06:54:28 UTC

Permalink

Ankita, this is problem with your batch queuing system. Do you know which
batch system you are using on this cluster?
Can you share with us what command you use to submit a job?

Also please do not share your teamviewer password with us. I doubt this is
of much use to anyone, but...

Post by Ankita m
While using the open mpi got this error. Can you please tell why so
---------- Forwarded message ---------
Date: Tue, 24 Apr 2018, 12:55 pm
Subject: Re: problem in cluster
while using openmpi- 1.4.5 the program ended by showing this error

teamviewer id 565 248 412
password jfu477
my contact number 7830622816

_______________________________________________
users mailing list
https://lists.open-mpi.org/mailman/listinfo/users

Ankita m

2018-04-25 09:10:09 UTC

Permalink

while using openmpi- 1.4.5 the program ended by showing this error file
(in the attachment)

I am Using PBS file . Below u can find the script that i am using to run
my program

John Hearns via users

2018-04-25 09:27:14 UTC

Permalink

I do not see much wrong with that.
However nodes=4 ppn=2 makes 8 processes in all.
You are using mpirun -np 64

Actually it is better practice to use the PBS supplied environment
variables during the job, rather than hard-wiring 64
I dont have access to a PBS cluster from my desk at the moment.
You could also investigate using mpiprocs=2 Then I think with openmpi if
it has compiled in PBS support all you would have to do is
mpirun

Are you sure your compute servers only have two cores ??

I also see that you are commenting out the module load openmpi-3.0.1 I
would guess you want the default Opnempi, which is OK

First thing I would do, before the mpirun line in that job script:

which mpirun (check that you are picking up an Openmpi version)

ldd ./cgles (check you are bringing in the libraries that you should)

Also run mpirun with the verbose flag -v

Post by Ankita m

while using openmpi- 1.4.5 the program ended by showing this error file
(in the attachment)

I am Using PBS file . Below u can find the script that i am using to run
my program
_______________________________________________
users mailing list
https://lists.open-mpi.org/mailman/listinfo/users

Ankita m

2018-04-25 09:43:30 UTC

Permalink

i have 16 cores per one node. I usually use 4 node each node has 16 cores
so total 64 processes.

On Wed, Apr 25, 2018 at 2:57 PM, John Hearns via users <

Post by John Hearns via users
I do not see much wrong with that.
However nodes=4 ppn=2 makes 8 processes in all.
You are using mpirun -np 64
Actually it is better practice to use the PBS supplied environment
variables during the job, rather than hard-wiring 64
I dont have access to a PBS cluster from my desk at the moment.
You could also investigate using mpiprocs=2 Then I think with openmpi if
it has compiled in PBS support all you would have to do is
mpirun
Are you sure your compute servers only have two cores ??
I also see that you are commenting out the module load openmpi-3.0.1 I
would guess you want the default Opnempi, which is OK
which mpirun (check that you are picking up an Openmpi version)
ldd ./cgles (check you are bringing in the libraries that you should)
Also run mpirun with the verbose flag -v

Post by Ankita m

while using openmpi- 1.4.5 the program ended by showing this error file
(in the attachment)

_______________________________________________
users mailing list
https://lists.open-mpi.org/mailman/listinfo/users

Ankita m

2018-04-25 09:44:51 UTC

Permalink

Can you please tell me whether to use mpicc compiler ar any other compiler
for openmpi programs

Post by Ankita m
i have 16 cores per one node. I usually use 4 node each node has 16 cores
so total 64 processes.
On Wed, Apr 25, 2018 at 2:57 PM, John Hearns via users <

Post by John Hearns via users
I do not see much wrong with that.
However nodes=4 ppn=2 makes 8 processes in all.
You are using mpirun -np 64
Actually it is better practice to use the PBS supplied environment
variables during the job, rather than hard-wiring 64
I dont have access to a PBS cluster from my desk at the moment.
You could also investigate using mpiprocs=2 Then I think with openmpi
if it has compiled in PBS support all you would have to do is
mpirun
Are you sure your compute servers only have two cores ??
I also see that you are commenting out the module load openmpi-3.0.1 I
would guess you want the default Opnempi, which is OK
which mpirun (check that you are picking up an Openmpi version)
ldd ./cgles (check you are bringing in the libraries that you should)
Also run mpirun with the verbose flag -v

Post by Ankita m

while using openmpi- 1.4.5 the program ended by showing this error file
(in the attachment)

I am Using PBS file . Below u can find the script that i am using to
run my program
_______________________________________________
users mailing list
https://lists.open-mpi.org/mailman/listinfo/users

_______________________________________________
users mailing list
https://lists.open-mpi.org/mailman/listinfo/users

John Hearns via users

2018-04-25 10:16:30 UTC

Permalink

Ankita, please read here: https://www.open-mpi.org/faq/?category=mpi-apps

Post by Ankita m
Can you please tell me whether to use mpicc compiler ar any other compiler
for openmpi programs

Post by Ankita m
i have 16 cores per one node. I usually use 4 node each node has 16 cores
so total 64 processes.
On Wed, Apr 25, 2018 at 2:57 PM, John Hearns via users <

Post by John Hearns via users
I do not see much wrong with that.
However nodes=4 ppn=2 makes 8 processes in all.
You are using mpirun -np 64
Actually it is better practice to use the PBS supplied environment
variables during the job, rather than hard-wiring 64
I dont have access to a PBS cluster from my desk at the moment.
You could also investigate using mpiprocs=2 Then I think with openmpi
if it has compiled in PBS support all you would have to do is
mpirun
Are you sure your compute servers only have two cores ??
I also see that you are commenting out the module load openmpi-3.0.1 I
would guess you want the default Opnempi, which is OK
which mpirun (check that you are picking up an Openmpi version)
ldd ./cgles (check you are bringing in the libraries that you should)
Also run mpirun with the verbose flag -v

Post by Ankita m

Post by Ankita m
while using openmpi- 1.4.5 the program ended by showing this error
file (in the attachment)

I am Using PBS file . Below u can find the script that i am using to
run my program
_______________________________________________
users mailing list
https://lists.open-mpi.org/mailman/listinfo/users

_______________________________________________
users mailing list
https://lists.open-mpi.org/mailman/listinfo/users

John Hearns via users

2018-04-25 10:19:18 UTC

Permalink

That fine. But in your job script ppn=2

Also check ldd cgles on the compute servers themselves.
Are all the libraries available in your path?

Post by Ankita m
i have 16 cores per one node. I usually use 4 node each node has 16 cores
so total 64 processes.
On Wed, Apr 25, 2018 at 2:57 PM, John Hearns via users <

Post by John Hearns via users
I do not see much wrong with that.
However nodes=4 ppn=2 makes 8 processes in all.
You are using mpirun -np 64
Actually it is better practice to use the PBS supplied environment
variables during the job, rather than hard-wiring 64
I dont have access to a PBS cluster from my desk at the moment.
You could also investigate using mpiprocs=2 Then I think with openmpi
if it has compiled in PBS support all you would have to do is
mpirun
Are you sure your compute servers only have two cores ??
I also see that you are commenting out the module load openmpi-3.0.1 I
would guess you want the default Opnempi, which is OK
which mpirun (check that you are picking up an Openmpi version)
ldd ./cgles (check you are bringing in the libraries that you should)
Also run mpirun with the verbose flag -v

Post by Ankita m

while using openmpi- 1.4.5 the program ended by showing this error file
(in the attachment)

I am Using PBS file . Below u can find the script that i am using to
run my program
_______________________________________________
users mailing list
https://lists.open-mpi.org/mailman/listinfo/users

_______________________________________________
users mailing list
https://lists.open-mpi.org/mailman/listinfo/users