Fail to use PYACTION

[phoo220]

I am trying to apply PYACTION in FLOW to run another simulation using the result from flow, read the data, go back to flow and restrict the production or injection accordingly. The problem is whenever I try run the flow with PYACTION, Timestep is chopped and showing incorrect restart time as below. How can I use PYACTION to open or shut the wells and go back to previous timestep?

================ Starting main simulation loop ===============


Report step  0/24 at day 0/730, date = 01-Jan-2015

Starting time step 0, stepsize 1 days, at day 0/31, date = 01-Jan-2015
2023-10-31 09:16:11 INFO     cliecalc        eCalc™ simulation starting. Running version '8.3.0' started at '2023.10.31 09:16:11'
2023-10-31 09:16:16 WARNING  libecalc        !EXPERIMENTAL! Maximum power constraint is an experimental feature where the syntax may change at any time.. It has not thoroughly tested and Quality Assured yet. Use at own risk. !EXPERIMENTAL!
2023-10-31 09:16:17 INFO     cliecalc        Writing run info to cache 'output/.ecalc'.
2023-10-31 09:16:17 INFO     cliecalc        Writing results to cache 'output/.ecalc'.
2023-10-31 09:16:17 INFO     cliecalc        eCalc™ simulation successful. Duration: 0:00:05.282316
[14.15407 14.15407 14.15407 14.15407 14.15407 14.15407 14.15407 14.15407
 14.15407 14.15407 14.15407 14.15407 14.15407 14.15407 14.15407 14.15407
 14.15407 14.15407 14.15407 14.15407 14.15407 14.15407 14.15407 14.15407
 14.15407 14.15407 14.15407 14.17458 14.22277]

Problem:
Timestep chopped to 0.330000 days


Starting time step 0, stepsize 0.33 days, at day 0/31, date = 01-Jan-2015
Program threw an exception: Elapsed time (0.32999999999999996 [DAYS]) must not precede previous elapsed time (1 [DAYS]). Incorrect restart time?

[blattms]

Sorry for the late reply.

Cool that you are stress testing the PYACTION stuff. We are still working on improving it.

I might be mistaking, but this looks familiar (not with PYACTION, though). Have you retried with more recent versions?

If the problem persists and you need help (and we find time...), then we need to able to replicate this. Do you have a small test case that triggers this?