PGI User Forum
 SearchSearch   MemberlistMemberlist     RegisterRegister   ProfileProfile    Log inLog in 

CUDA-x86.

fail with test input of gaussian
Goto page 1, 2  Next
 
Post new topic   Reply to topic    PGI User Forum Forum Index -> Programming and Compiling
View previous topic :: View next topic  
Author Message
jasonshih



Joined: 03 Aug 2004
Posts: 32

PostPosted: Tue Nov 30, 2004 8:42 am    Post subject: fail with test input of gaussian Reply with quote

Hi mkcolg,

I tried to compile gaussian on fedora core2 (intel p4) with latest PGI version. Though all executables are be built out, but turning to the test inputs, i encounter the following errors:

------------------
Erroneous write during file extend. write 1248 instead of 4096
Probably out of disk space.
Write error in NtrExt1: No such file or directory
------------------

I've found this is quite a common problem existing on Fedora OS. But I still have no clue how to fix this.

Thanks in advance.

BR,
J
Back to top
View user's profile
mkcolg



Joined: 30 Jun 2004
Posts: 6129
Location: The Portland Group Inc.

PostPosted: Tue Nov 30, 2004 1:05 pm    Post subject: Reply with quote

Hi Jason,

Have you run out of disk space as the message suggests?

- Mat
Back to top
View user's profile
jasonshih



Joined: 03 Aug 2004
Posts: 32

PostPosted: Tue Nov 30, 2004 6:09 pm    Post subject: Reply with quote

No. There still have prenty of disk space (about 5 G). I really have no idea why.


BR,
J
Back to top
View user's profile
mkcolg



Joined: 30 Jun 2004
Posts: 6129
Location: The Portland Group Inc.

PostPosted: Thu Dec 02, 2004 12:34 pm    Post subject: Reply with quote

Hi Jason,


I asked our application engineer that works with Gaussian to run it on on Fedora Core 2 and a SuSE 9.1 system. In both cases it worked fine. The scratch directory's disk usage topped out at about 500Mb. Granted, your problem set could be larger.

Things I would check are:

1) Does your system have disk quotas enabled and are you reaching your maximum useage?
2) Where is your GAUSS_SCRDIR scratch directory? Is it on its own smaller partition? Are you filling up this directory?
3) How big is your system's swap space? Perhaps it has reached its maximum.

If this isn't it, perhaps the someone at Gaussian has seen this error?

- Mat
Back to top
View user's profile
jasonshih



Joined: 03 Aug 2004
Posts: 32

PostPosted: Mon Dec 06, 2004 3:21 am    Post subject: Reply with quote

Hi Mat,

thanks for the feedback, and the application engineer's effort. :-)

regarrding your checklist:
1. we dont applied quota at present working parition.
2. though GAUSS_SCRDIR is not exported before encountering the error, but on older RH version, say 7.1, I dont need to setup this env variable before running the testcases. After reset this parameter (5 GB available on the disk), problem remains.
3. I've notice that our swap space is about 2GB (the same as physical memory).

What do you think result in this error? have you modified any parameter in gaussian's source? or just compile it straighforward.

BR,
J
Back to top
View user's profile
Display posts from previous:   
Post new topic   Reply to topic    PGI User Forum Forum Index -> Programming and Compiling All times are GMT - 7 Hours
Goto page 1, 2  Next
Page 1 of 2

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © phpBB Group