Image Integration process failing after recent update

[zkohlman]

As the subject states, I am attempting to integrate 50 .fit file dark frames taken with an ASI1600MM-Cool in Sequence Generator Pro.  I am using the same parameters according to Inside PixInsight as well as how I've previously always accomplished this task.  After it 'reads' through a few files, I get a winder stating that PixInsight Core has stopped working and Windows is closing the program.  This never happened before and the only thing I can think of is it might have something to do with the recent update to this process within the application. 

[zkohlman]

Was able to catch it before it quit working again.  This is what appears in the Process Console viewer:

C:/Users/Zach/Pictures/ZWO ASI 1600MM/Calibration Frames/Dark Frames/Gain 139 Offset 21/-20C/45s/Target 1_45sec_1x1__frame26.fit
Reading FITS image: 16-bit integers, 1 channel(s), 4656x3520 pixels:
*** Error: PCL FITS Format Support: FITS file read error: C:/Users/Zach/Pictures/ZWO ASI 1600MM/Calibration Frames/Dark Frames/Gain 139 Offset 21/-20C/45s/Target 1_45sec_1x1__frame26.fit
CFITSIO error message stack:
01 : C:\Users\Zach\Pictures\ZWO ASI 1600MM\Calibration Frames\Dark Frames\Gain 139 Of
02 : Error reading data buffer from file:
03 : fset 21\-20C\45s\Target 1_45sec_1x1__frame26.fit
04 : Error reading elements 1 thru 4656 from image (ffgcld).

[001] C:/Users/Zach/Pictures/ZWO ASI 1600MM/Calibration Frames/Dark Frames/Gain 139 Offset 21/-20C/45s/Target 1_45sec_1x1__frame3.fit

C:/Users/Zach/Pictures/ZWO ASI 1600MM/Calibration Frames/Dark Frames/Gain 139 Offset 21/-20C/45s/Target 1_45sec_1x1__frame27.fit
Reading FITS image: 16-bit integers, 1 channel(s), 4656x3520 pixels:
*** Error: PCL FITS Format Support: FITS file read error: C:/Users/Zach/Pictures/ZWO ASI 1600MM/Calibration Frames/Dark Frames/Gain 139 Offset 21/-20C/45s/Target 1_45sec_1x1__frame27.fit
CFITSIO error message stack:
01 : Error reading data buffer from file:
02 : fset 21\-20C\45s\Target 1_45sec_1x1__frame27.fit
03 : C:\Users\Zach\Pictures\ZWO ASI 1600MM\Calibration Frames\Dark Frames\Gain 139 Of
04 : Error reading elements 1 thru 4656 from image (ffgcld).

I've never had issues before, and at the same time I have never used .fit files before either.  New to SGP and new ASI1600...

[Juan Conejero]

This problem has nothing to do with ImageIntegration or any updates.

04 : Error reading elements 1 thru 4656 from image (ffgcld).

Here is the problem: the FITS files are incorrect. Please use NASA FITS verifier service to check that the files you've got are correct:

https://fits.gsfc.nasa.gov/fits_verify.html

I get a winder stating that PixInsight Core has stopped working and Windows is closing the program.

There can be many causes for this on Windows, including other applications, hardware/driver related problems and of course a bug in PixInsight. We have not seen any similar problem with the latest version of ImageIntegration. Can you provide a minimal set of (correct) files where we can reproduce this problem?

[zkohlman]

Thank you for the quick reply, I ran one file and here is the outputted information from the scan:

1 Header-Data Units in this file.
 
=================== HDU 1: Primary Array ===================
 
*** Warning: Some CRVALi keywords appear to be missing; expected 2.
 
   1 | SIMPLE  =                    T / file does conform to FITS standard
   2 | BITPIX  =                   16 / number of bits per data pixel
   3 | NAXIS   =                    2 / number of data axes
   4 | NAXIS1  =                 4656 / length of data axis 1
   5 | NAXIS2  =                 3520 / length of data axis 2
   6 | BZERO   =                32768 / offset and data range to that of unsigned short
   7 | BSCALE  =                    1 / default scaling factor
   8 | CRPIX1  =                 2328 / reference spectrum pixel coordinate for axis 1
   9 | CRPIX2  =                 1760 / reference spectrum pixel coordinate for axis 2
  10 | CTYPE1  = 'RA---DEF'           / standard system and projection
  11 | CTYPE2  = 'DEC--DEF'           / standard system and projection
  12 | OBJECT  = 'Target 1'           / Object name
  13 | DATE-LOC= '2016-12-26T03:08:29' / Local observation date
  14 | DATE-OBS= '2016-12-26T08:08:29' / UTC observation date
  15 | IMAGETYP= 'DARK    '           / Type of frame
  16 | CREATOR = 'Sequence Generator Pro v2.5.1.17' / Capture software
  17 | INSTRUME= 'ASI Camera (1)'     / Instrument name
  18 | EXPOSURE=                  300 / Exposure time in seconds
  19 | CCD-TEMP=                  -20 / Camera cooler temperature
  20 | SET-TEMP=                  -20 / Camera cooler temperature
  21 | XBINNING=                    1 / Camera X Bin
  22 | CCDXBIN =                    1 / Camera X Bin
  23 | YBINNING=                    1 / Camera Y Bin
  24 | CCDYBIN =                    1 / Camera Y Bin
  25 | XPIXSZ  =                  3.8 / Pixel Width in microns (with binning)
  26 | YPIXSZ  =                  3.8 / Pixel Height in microns (with binning)
  27 | END     
 
 27 header keywords
 
 16-bit integer pixels,  2 axes (4656 x 3520),
 
++++++++++++++++++++++ Error Summary  ++++++++++++++++++++++
 
 HDU#  Name (version)       Type             Warnings  Errors
 1                          Primary Array    1         0     
 
**** Verification found 1 warning(s) and 0 error(s). ****

[zkohlman]

The process will crash; however, with numerous repeated attempts it will all of a sudden work and allow me to create the Master Dark file.  This is a new camera (I haven't even been out in the field with it) and my first time using .fit files as I had previously been using a Nikon DSLR, and it is my first time using SGP which so far as I can see only outputs .fit files. 

[zkohlman]

https://db.tt/P9gk1EzMVc

Here is a folder with 10 of the FITS files. 

Again, the process begins running through the 50 FIT files and all of a sudden the window shows up stating that PixInsight stopped working.  After a few, sometimes 5 or 6, negative attempts with the same outcome, all of a sudden ImageIntegration successfully accepts all of the files and creates the Master Dark I have been working on. 

I tried ImageCalibration first one time to subtract bias signal from the darks which works fine and of course outputs an .xisf file which ImageIntegration accepts.  But if I'm going straight to ImageIntegration with the .fit files, the noted problems arise.

[Juan Conejero]

Thank you for uploading the files. I can integrate these FITS files without any problem on Windows 10, OS X 10.10 and 10.11, and Linux. I have integrated them multiple times with different parameters (just to make the test more intensive, since there is basically one way to integrate dark frames) without any issues.

Since I can't reproduce the problem, I have no idea why it's happening in your case. You can disable high-level parallel processing in ImageIntegration by setting the useFileThreads parameter to false. Click the Edit instance source code button, change the parameter value, and click the commit button.

[zkohlman]

That is weird and I appreciate looking into it.  Your suggestion works as well.  Before, after all of the program closing and when it would finally work, each file name in the Process View log would be in green font.  The way you suggested has everything in white.  That is how it used to look when I was integrating .nef files so I'm assuming this is the correct way to go.  Plus, I only have to worry about this when making master calibration frames which I go directly into ImageIntegration to process.  My lights, as per the known workflows, go through ImageCalibration first which pushes out .xisf files.  Thank you again!  Happy Holidays!

[Juan Conejero]

This problem should now be fixed with an update I've just released. Please see the other thread where this has also been reported for a full description of the problem and its solution.