Nexus writer feedback: round 1
I did a tour today of all BLISS beamlines to get feedback on the Nexus writer. These are the recurring comments so far:
- Default plots in the data file (plotselect for writer and flint #1358 (closed))
- Scans should not stop when something goes wrong with the writing: I would say at least the scan should never be stopped due to a timeout on the check-writer call to the Tango device (!2205 (merged)). If the writer is in FAULT state (data is not being written) it depends on who you ask. I'm wondering whether the writer should not mark the writing state of a scan in Redis? Not sure that's a good idea but then BLISS does not need to make Tango calls to check the writer status.
- a2scan NXdata plot should be a curve and not a scatter plot: goes together with plotselect because one of the motors should be selected as the X-axis
So @valentin.valls we'll have to continue our plotselect efforts :-).
Edited by Wout De Nolf