Missing HDU's in sfrhist output

[Gandhali Joshi]

Hello everyone,

I'm a new sunrise user and I'm having a few problems getting the programme working properly. I'm wondering if anyone has had similar issues or has any advice. I'm using a tipsy file of a Gasoline simulation. Here are the problems I'm having:

i) If I run sfrhist with n_threads>1, I get a segmentation fault.

ii) Once I set n_threads=1, I can run sfrhist, although I still get a lot of warnings. However, here is final part of the output from it:

Calculating bolometric luminosities

Writing output file

  galaxy

 File already exists: sfrhist.fits attempting to open existing file

 File already exists: sfrhist.fits attempting to open existing file

 File already exists: sfrhist.fits attempting to open existing file

Warning: No SED defined for species 2, assuming zero

      chunk size is 28 rows

 File already exists: sfrhist.fits attempting to open existing file

Warning: No SED defined for species 3, assuming zero

      chunk size is 28 rows

 File already exists: sfrhist.fits attempting to open existing file

      chunk size is 28 rows

 File already exists: sfrhist.fits attempting to open existing file

Warning: No SED defined for species 5, assuming zero

      chunk size is 28 rows

 File already exists: sfrhist.fits attempting to open existing file

FITS Error: Cannot read HDU in FITS file: No HDU with name INTEGRATED_QUANTITIES and version 1

Writing integrated grid SED to HDU INTEGRATED_QUANTITIES

FITS Error: Cannot read HDU in FITS file: Cannot access HDU name INTEGRATED_QUANTITIES version 1

FITS Error: Cannot read HDU in FITS file: INTEGRATED_QUANTITIES with version 1

 File already exists: sfrhist.fits attempting to open existing file

Sfrhist complete

The following config keywords were unused, check that this is ok:

L_bol_tolerance 1e+100  # [] sigma L/average L allowed ;changed from 10 to fit BC 3.5

NOBJECT 0       # Number of objects contained in this snapshot

configfile      "sfrhist.config"        # Configuration file

expansion       1       # 

pwd     "/work/joshigd/example" # Working directory

sfrhist "configuration file automatically generated by mcrxrun" # 

simdir  "./"    # directory where snapshots are

use_mappings_sed        1       # Set to false to run without molecular clouds

iii) When I tried running mcrx, I got a segmentation fault almost immediately, but it turns out this might be because the previous output doesn't have some of the HDU's it should. Most importantly the GRID_STRUCTURE one seems to be empty.

I've attached the configuration files I'm using and a log of sfrhist. I would really appreciate any pointers on what I'm doing wrong.

Thank you.

Gandhali Joshi

[Chris Hayward]

Hi Gandhali,

Sorry that I missed the approval notification for this message.

1. I haven't encountered the sfrhist segfault issue that you mentioned, and I can't diagnose it without output.

2. It looks like you don't have any luminous particles and thus no SEDs are created, although it's hard to tell from the output that you sent. Can you please send the full output?

3. This is expected given the sfrhist error.

Cheers,

Chris

--
You received this message because you are subscribed to the Google Groups "Sunrise" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sunrisemcrx...@googlegroups.com.
To post to this group, send email to sunri...@googlegroups.com.
Visit this group at http://groups.google.com/group/sunrisemcrx.
For more options, visit https://groups.google.com/d/optout.

--

Chris Hayward

Moore Postdoctoral Scholar in Theoretical Astrophysics

California Institute of Technology
Office: Cahill 326

Phone: 626-395-6987

Mailing: Caltech 350-17

1200 E. California Blvd.

Pasadena, CA 91125

http://www.cfa.harvard.edu/~chayward

[Giovanni]

Hi Gandhali,

I had the same problem. Be sure that you install the Blitz libraries with --enable-threadsafe

Cheers,

Giovanni

[Gandhali Joshi]

Hello,

I'm so sorry, I didn't have my notifications set up properly, so I just saw these replies today. Thank you for getting back to me.

Chris: The files were too big to attach here, so I've uploaded the input and output files as well as the configuration files to this link:

http://we.tl/u2lFGFQAjn

Thanks again for helping me with this.

Giovanni: Thanks for the tip, I'll definitely try that and hopefully it should fix the nthreads issue.

Cheers

Gandhali

[Chris Hayward]

Hi Gandhali,

I just now tried to look at this but the outputs are gone. Can you please repost them? All that I need are the output messages from Sunrise and your parameter files, not the Sunrise FITS output or input simulation files.

Cheers,

Chris

[Gandhali Joshi]

Hi Chris,

Oh, sorry about that, the link might have expired after a week. I've reattached the input parameter files and a log of the output messages for sfrhist. It looks like the grid I realized that I didn't have save_cellcodes set to true, so the single column in the GRIDSTRUCTURE HDU makes sense now, but I'm not sure why the segmentation fault is happening in the mcrx stage. I've attached the mcrx configuration files as well, in case they're the problem. Thanks again for looking into this.

Cheers

Gandhali