FLINT: The GUI can crash a continuous scan
Feedback from ID15.
On ID15 changing the sinogram colormap in FLint sometimes takes a lot of time (no idea why)
During a continuous scan, it looks like BLISS continuously send RPC commands to Flint.
As result, if the event loop is not released on FLint, the RPC request from BLISS receive a timeout and fail the scan.
There is at least 2 problems here. And both should be fixed.
- Why it takes so much time to update a colormap
- Why the scan trig Flint during the acquisition
- What if we use a thread to handle RPC requests?