NIRSPEC dark parallel error for program 6634

[stscijgbot-jp]

Issue JP-3694 was created on JIRA by Michael Ridgaway:

Datasets  jw06634435001_05301_00001_nrs1 and jw06634435001_05301_00001_nrs2 failed at Level 2A with the following error:

---------------------------------------------------------------------- ERROR RUNNING STEP 'DarkPipeline': Error determining best reference for 'superbias' = No match found. ----------------------------------------------------------------------

Discussion with Hien and Rossy revealed there is no subarray for the corners of this NIRSpec dark parallel. The full discussion can be found here: https://stsci.slack.com/archives/C9KF4MVEV/p1721655419896409

[stscijgbot-jp]

Comment by Rosa Diaz on JIRA:

I took a look at the PRD and the corners for SUB32 there agree with the subarrays we have in CRDS [1399,975] for NIRIS1 and [619,1043]for NRIS2. Note that the start of the subarray in this data [1,1]is odd. Does not agree even with the start of the subarray in the detector coordinates [975,1399] for NRS1 and [975, 975]for NRS2.

Was this special dark? It is not defined in the PRD or have a superbias reference file for it. Should it be calibrated?

[stscijgbot-jp]

Comment by Christian Hayes on JIRA:

Rosa Diaz thanks for pointing this case out.  This is not a special dark and in principle it should have corners that match with current CRDS reference files.  We suspect there is an issue with how the corner keywords are being populated in the headers and have opened up a DMS ticket here to look into the issue:  JWSTDMS-1042.  It's possible this issue may occur for other new Cycle 3 NIRSpec subarray darks (until the source of the issue is fixed), so it would be useful to know if you run into this again.