Say I have a device with two signals, one fails to connect with a timeout, the other immediately raises an error when connect is called. The latter error is not logged. If the timing out signal is subsequently fixed, then the error signal will be logged if I try to connect again.
This may be a good thing as it prevents too much log spam. It may be beneficial to work through the errors one-at-a-time, continually re-running the program and solving the next error until there are none left. Opinions on this point are welcome.
Even if we chose to leave the error un-logged, we should still fix the annoying spam that comes from asyncio because the exception is not handled.
Task exception was never retrieved
future: <Task finished name='Task-4' coro=<DummyDeviceThatTimesOutWhenConnecting.connect() done, defined at ophyd/v2/tests/test_core.py:207> exception=NotConnected('source: foo')>
Traceback (most recent call last):
File "ophyd/v2/tests/test_core.py", line 209, in connect
await asyncio.Future()
asyncio.exceptions.CancelledError
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "ophyd/v2/tests/test_core.py", line 211, in connect
raise NotConnected("source: foo")
ophyd.v2.core.NotConnected: source: foo
Say I have a device with two signals, one fails to connect with a timeout, the other immediately raises an error when connect is called. The latter error is not logged. If the timing out signal is subsequently fixed, then the error signal will be logged if I try to connect again.
This may be a good thing as it prevents too much log spam. It may be beneficial to work through the errors one-at-a-time, continually re-running the program and solving the next error until there are none left. Opinions on this point are welcome.
Even if we chose to leave the error un-logged, we should still fix the annoying spam that comes from asyncio because the exception is not handled.