Srikanth Chekuri
Srikanth Chekuri
@sumit-158 this may not really be the good first issue so I am go remove that tag so I would suggest picking some different good first issue to get you...
I am not familiar with pylance but mypy [checking](https://github.com/open-telemetry/opentelemetry-python/blob/main/tox.ini#L174) should be working with `opentelemetry-api`.
@aceberle This might help you https://mypy.readthedocs.io/en/stable/command_line.html#cmdoption-mypy-namespace-packages
We didn't add `force_flush` on metric reader interface. I think @aabmass suggested we will just invoke `collect`.
>I believe the solution here would be the processor receiving a callable that instantiates the exporter, so that it can create a new exporter after a fork. Currently, it receives...
>A fork only forks the thread that calls the fork function What do you mean it _only forks the thread that calls the fork_? > So, in my opinion, implementing...
> The concern belongs to the processor to make everything it needs work in the forked environment it is creating. Just want make sure we are on same page, is...
My understanding of the issue so far is, that `AgentClientUDP` will possibly end up in an inconsistent state if forking happens at the right time in the export pipeline. This...
> But let's say that there's a process using the exporter, and it calls something that begins a thrift struct, and forks before calling the function that ends the struct....