Details
-
Type:
Story
-
Status: Done
-
Resolution: Done
-
Fix Version/s: None
-
Component/s: obs_lsst
-
Labels:
-
Story Points:2
-
Epic Link:
-
Team:Data Release Production
-
Urgent?:No
Description
The template definitions in (many of) the policy files in obs_lsst for are identical for the icExp and icExpBackground datasets thus a "put" of one will overwrite the other. Since, in the one place these datasets can be written out (in characterizeImage here and controlled via config parameter charImage.doWriteExposure), they are both written out in sequence, the latter always overwrites the former. The icExpBackground template needs to be changed to remedy this situation.
Attachments
Activity
Jenkins is running.
Processing DC2 data with processCcd.py and including the charImage.doWriteExposure=True config override in order to get the icExp and icExpBackground datasets persisted, prior to this ticket, if you tried to read in the icExp, you'd get something like this:
What??!! Turns out that is actually the icExpBackground exposure that clobbered the icExp due to having identical templates in the policy file. With this ticket, we now get:
Aahhh...much better!
Jira did some funky timing-out when I was updating this ticket, so the comments above are out of order!
Yup, looks great 
Thanks for the speedy review! Jenkins passed, so merged and done!
Can I bug you for a quick review? I'm singling you out as I did make a change to the LATISS policy file which I'm not able to officially check...I did try poking at some data in /project/shared/auxTel/ but it seemed to hang when it got to:
But, I'm actually thinking running with charImage.doWriteExposure=True on LATISS data might have bombed since it would have defaulted to the definition in the main lsstCamMapper.yaml and would look for a raftName?