francis-dupuis commented on issue #11507:
URL: https://github.com/apache/superset/issues/11507#issuecomment-772101000


   So I've used geckodriver and chrome on 1.0.0 and 0.38 and they've both 
worked. The problem is that a dashboard email job won't complete. You need to 
know how to debug this to get anywhere.
   
   From a surface level, it could be a couple of things that are all 
speculatory...
   
   - You're posting incorrect credentials with the webdriver
   - The webdriver process isn't launching at all (incorrect webdriver URL 
config or permission issue)
   - You're celery config is wrong
   
   You can get started debugging via celery. Most of the email reporting 
configs are via a celery beat-scheduler and a celery worker.
   
   If you're running a docker setup (which I'm assuming you are) you can try 
examining the logs in or outside of the docker container. What helped me out 
the most is launching a celery flower instance with docker-compose. Flower will 
allow you to monitor the success of these jobs as well as showing the traceback 
if a job fails. You can spin up Flower pretty easily with docker-compose. I 
configured flower to run on port 5555 (which is the default). Once Flower is 
running, navigate to it and inspect the job. Below is my compose setup. Replace 
the repo in the image name with one of your choice.
   
   `version: '3'
   services:
     redis:
       image: redis
       restart: always
       volumes:
         - redis:/data
     postgres:
       image: postgres
       restart: always
     superset:
       image: supers
       restart: always
       depends_on:
         - postgres
         - redis
       ports:
         - "8088:8088"
     worker:
       image: repo/superset
       restart: always
       depends_on:
         - postgres
         - redis
     beat_worker:
       image: repo/superset
       restart: always
       depends_on:
         - postgres
         - redis
     flower:
       image: repo/superset
       ports:
         - 5555:5555
       depends_on:
         - beat_worker
         - worker
   volumes:
     postgres:
     redis:
   `
   After everything is spun up, try sending a test email on the scheduled 
dashboard UI.
   
   ![Screen Shot 2021-02-02 at 6 09 14 
PM](https://user-images.githubusercontent.com/76928527/106674762-e14c6980-6581-11eb-8c81-fa5cad813baa.png)
   
   ![Screen Shot 2021-02-02 at 6 09 07 
PM](https://user-images.githubusercontent.com/76928527/106674759-e01b3c80-6581-11eb-9f48-5a0539576697.png)
   
   You would see a traceback if it failed. This is how I debugged originally 
when my emails wouldn't send. There's a bunch of possibilities why you
   
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to