Reauth issue with more than 4 running containers

I have this strange issue, which I only encounter when I have a certain number of running processes.

The setup is VMware ESXi virtual machine running Debian 11 with Docker on top. I am now running devices in docker containers, which basically contains a Azure IoT Device client per container together with a bunch of logic. Three of these containers works fine, but when I startup the fourth container, it connects fine, but after 1 hour when it tries to re-auth it fails (only happens on the last started container):

```
22-11-22 10:30:02[ERROR] Exception caught in background thread. Unable to handle.
22-11-22 10:30:02[ERROR] ["azure.iot.device.common.pipeline.pipeline_exceptions.OperationCancelled: OperationCancelled('Transport timeout on connection operation')\n"]
Exception in thread Thread-410:
Traceback (most recent call last):
File "/usr/local/lib/python3.9/threading.py", line 980, in _bootstrap_inner
self.run()
File "/usr/local/lib/python3.9/threading.py", line 1306, in run
self.function(*self.args, **self.kwargs)
File "/usr/local/lib/python3.9/site-packages/azure/iot/device/common/pipeline/pipeline_stages_base.py", line 466, in retry_reauthorize
this._reauthorize()
File "/usr/local/lib/python3.9/site-packages/azure/iot/device/common/pipeline/pipeline_thread.py", line 192, in wrapper
assert (
AssertionError:
Function _reauthorize is not running inside pipeline thread.
It should be. You should use invoke_on_pipeline_thread(_nowait) to enter the
pipeline thread before calling this function. If you're hitting this from
inside a test function, you may need to add the fake_pipeline_thread fixture to
your test. (generally applied on the global pytestmark in a module)
22-11-22 10:31:02[DEBUG] OperationCancelled('Could not complete operation') caused by OperationCancelled('Transport timeout on connection operation')
```

I am running:
- `azure-iot-device==2.8.0` (Tried version 2.9.0, doesn't help)
- Docker: `Docker version 20.10.21, build baeda1f`