Data Writing Issues with Prospero
Monday, May 13, 2019 12:46 PM
We are still encountering issues with saving the fits files. The data seems to read out fine, but then the system freezes while writing to file. The only solution has been to reboot from the Raritan keyboard in the computer room, and then reinitialize Caliban and restart prospero. The previous exposure is then lost.
I have no idea if there is any substance to my hypothesis (I've only consciously noticed it a couple times, while this issue has happened several times), but it may have something to do with changing coordinates in JSkyCalc (and/or starting a slew?) before the readout has completed. I am not sure though.
This issue has caused us to lose several calibration files in addition to some science exposures.
There could be something to this. The 1.3m TCS is a very “dumb” system, in that it can’t do two process simultaneously. As such, it may be possible that while trying to get telescope information for the header while the telescope is in motion, the data writing hangs. Likely worthwhile not slewing until the file is done reading out.