Mailman 3 YT on NASA Pleiades - yt-users

YT on NASA Pleiades

James Guillochon

17 Nov 2011 17 Nov '11

2 a.m.

I am getting a failure in the "load" statement when trying to run YT on NASA Pleiades. It seems that it is having problems with the numpy package. Here is the traceback: http://paste.yt-project.org/show/1947/ My script is just: from yt.mods import *

...

load('filename')

Any ideas? -- James Guillochon Department of Astronomy & Astrophysics University of California, Santa Cruz jfg@ucolick.org

Attachments:

attachment.html (text/html — 902 bytes)

Show replies by thread

Sam Skillman

17 Nov 17 Nov

6:26 p.m.

Hi James, This somewhat difficult to see what is going on. Could you show us the results of h5ls filename? I don't have a FLASH dataset to test things out on. Also, is this a cosmological sim? Finally, do you have a dataset similar to that one on different machine that does load? I've got some ideas but it'll be tough to test. If you're around and want to work through this a bit on IRC, i'll be around off and on today. Sam On Wed, Nov 16, 2011 at 7:00 PM, James Guillochon wrote:

...

I am getting a failure in the "load" statement when trying to run YT on NASA Pleiades. It seems that it is having problems with the numpy package. Here is the traceback:

http://paste.yt-project.org/show/1947/

My script is just:

from yt.mods import *

...
load('filename')

Any ideas?

-- James Guillochon Department of Astronomy & Astrophysics University of California, Santa Cruz jfg@ucolick.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

John ZuHone

6:44 p.m.

James, Sam, I've seen this before too on Pleiades but have never figured it out. I am not running cosmological simulations. I would be willing to try to reproduce it and help debug it, but I'm not sure if I can get to it today. Best, John Z On Nov 17, 2011, at 1:26 PM, Sam Skillman wrote:

...

Hi James,

This somewhat difficult to see what is going on. Could you show us the results of h5ls filename? I don't have a FLASH dataset to test things out on. Also, is this a cosmological sim? Finally, do you have a dataset similar to that one on different machine that does load? I've got some ideas but it'll be tough to test.

If you're around and want to work through this a bit on IRC, i'll be around off and on today.

Sam

On Wed, Nov 16, 2011 at 7:00 PM, James Guillochon wrote: I am getting a failure in the "load" statement when trying to run YT on NASA Pleiades. It seems that it is having problems with the numpy package. Here is the traceback:

http://paste.yt-project.org/show/1947/

My script is just:

from yt.mods import * load('filename')

Any ideas?

-- James Guillochon Department of Astronomy & Astrophysics University of California, Santa Cruz jfg@ucolick.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

Sam Skillman

8:45 p.m.

Hi John, That'd be great, as I neither have access to Pleiades or a representative FLASH dataset. Please let us know how this ends up. Also, James, if you could still send over the results of h5ls filename and a whether or not it's a cosmological sim, that'd be awesome. Thanks, Sam On Thu, Nov 17, 2011 at 11:44 AM, John ZuHone < jzuhone@milkyway.gsfc.nasa.gov> wrote:

...

James, Sam,

I've seen this before too on Pleiades but have never figured it out. I am not running cosmological simulations. I would be willing to try to reproduce it and help debug it, but I'm not sure if I can get to it today.

Best,

John Z

On Nov 17, 2011, at 1:26 PM, Sam Skillman wrote:

Hi James,

This somewhat difficult to see what is going on. Could you show us the results of h5ls filename? I don't have a FLASH dataset to test things out on. Also, is this a cosmological sim? Finally, do you have a dataset similar to that one on different machine that does load? I've got some ideas but it'll be tough to test.

If you're around and want to work through this a bit on IRC, i'll be around off and on today.

Sam

On Wed, Nov 16, 2011 at 7:00 PM, James Guillochon wrote:

...
I am getting a failure in the "load" statement when trying to run YT on NASA Pleiades. It seems that it is having problems with the numpy package. Here is the traceback:

http://paste.yt-project.org/show/1947/

My script is just:

from yt.mods import *

...
load('filename')

Any ideas?

-- James Guillochon Department of Astronomy & Astrophysics University of California, Santa Cruz jfg@ucolick.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

James Guillochon

8:55 p.m.

Hi Sam, So I just copied over a dataset that is working with YT on our local cluster, and I get the same result when trying to load that file on NASA Pleiades. The simulation is not cosmological. Here are the results of h5ls: http://paste.yt-project.org/show/1948/ On Thu, Nov 17, 2011 at 12:45 PM, Sam Skillman wrote:

...

Hi John,

That'd be great, as I neither have access to Pleiades or a representative FLASH dataset. Please let us know how this ends up. Also, James, if you could still send over the results of h5ls filename and a whether or not it's a cosmological sim, that'd be awesome.

Thanks, Sam

On Thu, Nov 17, 2011 at 11:44 AM, John ZuHone < jzuhone@milkyway.gsfc.nasa.gov> wrote:

...
James, Sam,

I've seen this before too on Pleiades but have never figured it out. I am not running cosmological simulations. I would be willing to try to reproduce it and help debug it, but I'm not sure if I can get to it today.

Best,

John Z

On Nov 17, 2011, at 1:26 PM, Sam Skillman wrote:

Hi James,

This somewhat difficult to see what is going on. Could you show us the results of h5ls filename? I don't have a FLASH dataset to test things out on. Also, is this a cosmological sim? Finally, do you have a dataset similar to that one on different machine that does load? I've got some ideas but it'll be tough to test.

If you're around and want to work through this a bit on IRC, i'll be around off and on today.

Sam

On Wed, Nov 16, 2011 at 7:00 PM, James Guillochon wrote:

...
I am getting a failure in the "load" statement when trying to run YT on NASA Pleiades. It seems that it is having problems with the numpy package. Here is the traceback:

http://paste.yt-project.org/show/1947/

My script is just:

from yt.mods import *

...
load('filename')

Any ideas?

-- James Guillochon Department of Astronomy & Astrophysics University of California, Santa Cruz jfg@ucolick.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

-- James Guillochon Department of Astronomy & Astrophysics University of California, Santa Cruz jfg@ucolick.org

Sam Skillman

9:05 p.m.

Thanks, and then what about h5ls filename/logical\ runtime\ parameters It seems like it's getting an int instead of a string when searching through the parameter list. If you have time and can log on to IRC I can talk you through some debugging with ipython that will probably speed this up quite a bit. Thanks and sorry for the trouble, Sam On Thu, Nov 17, 2011 at 1:55 PM, James Guillochon wrote:

...

Hi Sam,

So I just copied over a dataset that is working with YT on our local cluster, and I get the same result when trying to load that file on NASA Pleiades. The simulation is not cosmological. Here are the results of h5ls:

http://paste.yt-project.org/show/1948/

On Thu, Nov 17, 2011 at 12:45 PM, Sam Skillman wrote:

...
Hi John,

That'd be great, as I neither have access to Pleiades or a representative FLASH dataset. Please let us know how this ends up. Also, James, if you could still send over the results of h5ls filename and a whether or not it's a cosmological sim, that'd be awesome.

Thanks, Sam

On Thu, Nov 17, 2011 at 11:44 AM, John ZuHone < jzuhone@milkyway.gsfc.nasa.gov> wrote:

...
James, Sam,

I've seen this before too on Pleiades but have never figured it out. I am not running cosmological simulations. I would be willing to try to reproduce it and help debug it, but I'm not sure if I can get to it today.

Best,

John Z

On Nov 17, 2011, at 1:26 PM, Sam Skillman wrote:

Hi James,

This somewhat difficult to see what is going on. Could you show us the results of h5ls filename? I don't have a FLASH dataset to test things out on. Also, is this a cosmological sim? Finally, do you have a dataset similar to that one on different machine that does load? I've got some ideas but it'll be tough to test.

If you're around and want to work through this a bit on IRC, i'll be around off and on today.

Sam

On Wed, Nov 16, 2011 at 7:00 PM, James Guillochon wrote:

...
I am getting a failure in the "load" statement when trying to run YT on NASA Pleiades. It seems that it is having problems with the numpy package. Here is the traceback:

http://paste.yt-project.org/show/1947/

My script is just:

from yt.mods import *

...
load('filename')

Any ideas?

-- James Guillochon Department of Astronomy & Astrophysics University of California, Santa Cruz jfg@ucolick.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

-- James Guillochon Department of Astronomy & Astrophysics University of California, Santa Cruz jfg@ucolick.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

Sam Skillman

10:46 p.m.

Just an update, this seems to have been fixed in https://bitbucket.org/yt_analysis/yt/changeset/68227951f9b3/ Sam On Thu, Nov 17, 2011 at 2:05 PM, Sam Skillman wrote:

...

Thanks, and then what about h5ls filename/logical\ runtime\ parameters It seems like it's getting an int instead of a string when searching through the parameter list. If you have time and can log on to IRC I can talk you through some debugging with ipython that will probably speed this up quite a bit.

Thanks and sorry for the trouble, Sam

On Thu, Nov 17, 2011 at 1:55 PM, James Guillochon wrote:

...
Hi Sam,

So I just copied over a dataset that is working with YT on our local cluster, and I get the same result when trying to load that file on NASA Pleiades. The simulation is not cosmological. Here are the results of h5ls:

http://paste.yt-project.org/show/1948/

On Thu, Nov 17, 2011 at 12:45 PM, Sam Skillman wrote:

...
Hi John,

That'd be great, as I neither have access to Pleiades or a representative FLASH dataset. Please let us know how this ends up. Also, James, if you could still send over the results of h5ls filename and a whether or not it's a cosmological sim, that'd be awesome.

Thanks, Sam

On Thu, Nov 17, 2011 at 11:44 AM, John ZuHone < jzuhone@milkyway.gsfc.nasa.gov> wrote:

...
James, Sam,

I've seen this before too on Pleiades but have never figured it out. I am not running cosmological simulations. I would be willing to try to reproduce it and help debug it, but I'm not sure if I can get to it today.

Best,

John Z

On Nov 17, 2011, at 1:26 PM, Sam Skillman wrote:

Hi James,

This somewhat difficult to see what is going on. Could you show us the results of h5ls filename? I don't have a FLASH dataset to test things out on. Also, is this a cosmological sim? Finally, do you have a dataset similar to that one on different machine that does load? I've got some ideas but it'll be tough to test.

If you're around and want to work through this a bit on IRC, i'll be around off and on today.

Sam

On Wed, Nov 16, 2011 at 7:00 PM, James Guillochon wrote:

...
I am getting a failure in the "load" statement when trying to run YT on NASA Pleiades. It seems that it is having problems with the numpy package. Here is the traceback:

http://paste.yt-project.org/show/1947/

My script is just:

from yt.mods import *

...
load('filename')

Any ideas?

-- James Guillochon Department of Astronomy & Astrophysics University of California, Santa Cruz jfg@ucolick.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

-- James Guillochon Department of Astronomy & Astrophysics University of California, Santa Cruz jfg@ucolick.org

_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org

4543

Age (days ago)

4543

Last active (days ago)

List overview

Download

6 comments

3 participants

participants (3)

James Guillochon
John ZuHone
Sam Skillman

YT on NASA Pleiades

James Guillochon

Sam Skillman

John ZuHone

Sam Skillman

James Guillochon

Sam Skillman

Sam Skillman

tags

participants (3)