search

kslong / kred

Tools for analysing narrow band imaging with DECam
GNU General Public License v3.0
1 stars 0 forks source link

MefPrep.pf failure field LMC_c38 #15

Closed SeanPoints closed 2 days ago

SeanPoints commented 1 year ago

Branch: back Commit: b6f3289e4daedc51aea4f44a7e1980c8ea27cbb6

Began processing file c4d_190108_063145_ooi_N662_v1 Runtime error: decompression error: hit end of compressed byte stream

Checked the MEF fits image and it gave message: "Fitsy++ rice bad inflate" and only showed 3 detectors. It may be a bad FITS file from the archive. Going to move it out of the MEF directory when I can.

kslong commented 1 year ago

@SeanPoints I looked, and I may have seen this error also. I will try running this mañana, also. I want to figure out how to throw an error that we see, but that get's past it.

kslong commented 1 year ago

I have run MefSum on this file. There are lots of warnings, about keywords, but it runs successfully. However, if one looks at the det.tab file shows a number of extension for which the statistics that were problems. It's odd however that the keywords which are searched for seemed to be in the det.tab file. I confirm that MefPrep fails, with the error that @SeanPoints indicated.

I think what should have happened is the MefSum should have recognized that something was wrong and put this in the bad.tab file, which it does if the RAs and DECs are missing

kslong commented 1 year ago

I am testing the fix that I suggested above, on all of LMC_c38

kslong commented 12 months ago

@SeanPoints Are you still seeing this error. If you are not could you please close the issue. If you are still seeing it, please provide a bit more information.

kslong commented 3 days ago

To check whether this is really solved. One needs to have the MEF file above in the DECam_MED directory. For the recent data that is no longer the case, so it may not be surprising that I am not seeing the error anymore. Presumably the file is in one of the old versions that is on Box.

kslong commented 2 days ago

I downloaded the oldest version of the data that is on Box, and ran the current version of kred on LMC_c48 through the MefPrep stages and the file that was associated with this exposure (and all of the other files run fine). These files files have the extension _req, instead of v1. My guess is that something did fail in the original processing of the data, that was corrected when the data were reprocessed. One could verify this by re-retrieving the data from the NOIRLAB archive, but this seems not worth the trouble. I am closing the issue; if a new problem appears, we can deal with it then.