`nest.ResetKernel()` resets installed modules

[alberto-antonietti]

Hello!

Passing from NESTML 7.0 and NEST 3.6 to current master NESTML and NEST 3.7 introduced this change in the behaviour.

If you execute nest.ResetKernel() after having loaded (nest.Install()) your NESTML-compiled module makes the module uninstalled.

E.g., you can't use the neuron models installed with the new module unless you re-run nest.Install()

Before this was not happening, the model was staying installed and if you tried to re-run nest.Install() you got the "Module '...' is loaded already error.

You can test running for instance:

nest.Install(your_module)
nest.ResetKernel()
nest.Install(your_module)

Is this wanted or not? I am not sure if this is more a NESTML or a NEST thing.

[alberto-antonietti]

CC @rbonometto

[clinssen]

Hi Alberto, this is indeed true! It is related to the change in NEST Simulator (nest/nest-simulator#3103) and was introduced in NESTML 7.0.1 (see the release notes at https://github.com/nest/nestml/releases). Sorry, but your scripts will need to be adjusted going forward.

[alberto-antonietti]

Hi Charl, perfect, thanks for the info! Will take this into account and adapt our scripts!