<@ULVA73B9P>, I am using prefect cloud 2. In a win...
# ask-marvin
m
@Marvin, I am using prefect cloud 2. In a windows virtual machine I have created a venv, in which I launch the following operation to start a worker: ‘’’ set VIRTUAL_ENV=C:\PREFECT_GITHUB\venv set PATH=%VIRTUAL_ENV%\Scripts;%PATH% prefect cloud login -k ** --workspace prometeia/production prefect config set PREFECT_API_URL= "****" prefect worker start --pool "workpool-production" ‘’’ the created workpool is of type process. When the worker starts it gives the following error ‘’’ | ERROR | GlobalEventLoopThread | prefect._internal.concurrency - Service ‘EventsWorker’ failed with 1 pending items. ‘’’ from a discord report i read it was a python certification issue, but even with pip install --upgrade certifi it worked does anyone know anything? do you know how to help me?
m
The error
Service 'EventsWorker' failed with N pending items
seems to be a known issue. Here are some steps to potentially resolve it or mitigate its impact: 1. Configuration Check: Ensure that your environment and worker configuration are correctly set. Here are the essentials:
Copy code
shell
    set VIRTUAL_ENV=C:\PREFECT_GITHUB\venv
    set PATH=%VIRTUAL_ENV%\Scripts;%PATH%
    prefect cloud login -k **** --workspace prometeia/production
    prefect config set PREFECT_API_URL="****"
    prefect worker start --pool "workpool-production"
2. Certifi Update: If updating
certifi
didn't help, ensure that all your packages are up to date:
Copy code
shell
    pip install --upgrade certifi
    pip install --upgrade prefect
3. Windows Event Loop Issue: Since the error involves the event loop and concurrency, it might be specific to async handling on Windows. Check if there are any anomalies in
asyncio
related libraries. 4. GitHub Issues: This specific failure has been discussed on GitHub. The reported issue suggests it doesn't affect flow runs and may be more of a nuisance than a blocker. However, you can track and contribute to it here. 5. Initialize Event in Context: Prefect's internal updates address initializing certain components in the right context:
Copy code
python
    asyncio.Event in EventPersister
Make sure you are using a Prefect version that incorporates these fixes. For further reference on these points: - Global Concurrency - Prefect Worker Process - GitHub Issue If the issue persists, consider opening a new issue with detailed context and logs. Considering how inevitable problems seem and the rather bleak assurance that they don't affect flow runs, I'd suggest keeping a close watch and updating your setup frequently. Good luck. sigh