You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, for a single process, python interpreter doesn’t perform dlclose after opening any cython library until the process terminates. This produces issues when the variables have global visibility, which is the default for pybind. We can set visibility for these variables within only the objects scope by passing appropriate flags in the pybind module.
My specific problem has to do with the Node creation. I want to instantiate multiple Nodes in isolated network in a single process. Currently due to the mentioned problem, this is impossible because the Node will stay with the env vars found at first import. We cannot delete references to the module from sys.modules and obtain desired behaviour.
I would like to work on this as well, to contribute to the project.
Alternatives considered
Implementation suggestion
Additional context
Multiple isolated Nodes in a single Python process.
The text was updated successfully, but these errors were encountered:
Desired behaviour
Currently, for a single process, python interpreter doesn’t perform dlclose after opening any cython library until the process terminates. This produces issues when the variables have global visibility, which is the default for pybind. We can set visibility for these variables within only the objects scope by passing appropriate flags in the pybind module.
My specific problem has to do with the Node creation. I want to instantiate multiple Nodes in isolated network in a single process. Currently due to the mentioned problem, this is impossible because the Node will stay with the env vars found at first import. We cannot delete references to the module from sys.modules and obtain desired behaviour.
I would like to work on this as well, to contribute to the project.
Alternatives considered
Implementation suggestion
Additional context
Multiple isolated Nodes in a single Python process.
The text was updated successfully, but these errors were encountered: