QuantumQuadrate / CsPyController

GNU Lesser General Public License v3.0
3 stars 3 forks source link

Stalled running state #57

Open prhuft opened 4 years ago

prhuft commented 4 years ago

Anytime an experiment is running and stalls (there are multiple cases in which this can happen, but one example is if a connection to the PXI server is lost and "Pause after error" was unchecked, which it is by default, which is a terrible initialization choice), the experiment ends up in a state where it is still "running" but not doing anything. Trying to "reset and run" results in being told to halt the experiment first. After halting the experiment, "reset and run" results in stalling after "This is the pre-experiment save event. Origin reruns its save event at the end." is logged. See image below. This is incredibly frustrating, and often necessitates quitting cspy and restarting to fix.

image

c-poole commented 4 years ago

I'm familiar with the need to "stop" and then "reset and run" workflow which could be adjusted. I've never actually had the program stall after doing this though. Do you need to just restart cspy or the whole computer to regain functionality?

prhuft commented 4 years ago

When this happens quitting and relaunching CsPy is sufficient.

c-poole commented 3 years ago

I attempted to replicated this with the following steps:

run experiment with pause error unchecked during an experiment, shut down the pxi server. wait for cspy to error restart pxi server hit reset and run to get told to hit stop first hit stop hit reset and run

This does not result in halting for me but seems to be the conditions you mentioned so this may be specific to Rb. Did I make a mistake in my steps?

prhuft commented 3 years ago

I could not reproduce it either this morning. Evidently the procedure I describe above was not solely responsible for causing this behavior. I'll update this when I am able to cause it deterministically.