[GitHub] [airflow] potiuk commented on issue #17310: Got `subprocess.TimeoutExpired` when run `./breeze --python 3.8 buildl-docs` on Apple M1 chip

[GitBox <gi...@apache.org>]

potiuk commented on issue #17310:
URL: https://github.com/apache/airflow/issues/17310#issuecomment-889808293


   Also you can try to increase the number of CPUs that are available to docker engine. The Build will take advantage of as many CPUS as many you give it to and it will build the documentation in parallel. It looks like the timeout happens because some of the builds were scheduled to run (via paralell) and waited too long, so i f you increase number of CPUs, they will run much faster in general. 
   
   (Please let us know how it works).
   
   Root caus of it is that M1 uses emulation (our Docker image is x86 one) so it will run much slower than native execution.
   
   Also another work is try to run it  locally on Mac - as long as  you have locally instlaled "airflow[doc]" virtualenv, building docs should work with your natively installed python packages - hopefully all sphinx-related libraries can run natively on Mac M1, so you can run it there (it will not take advantage of parallell runs though).
   
   Yet another workaround is that you can just build a single package using for example `./brees build-docs -- --package-filter 'apache-airflow'` to only build airflow documentation.
   
   Taking into account a number of workarounds available and the fact that there is another ticket to implement native M1 docker https://github.com/apache/airflow/issues/15635  I close this ticket as "won't fix".


-- 
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.

To unsubscribe, e-mail: commits-unsubscribe@airflow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org