You shouldn’t need to set the entrypoint. The agent will do that for you. I think the issue here is more of activating conda is tricky. Do you need conda? The container should already provide isolation
h
Horatiu Bota
05/17/2022, 9:38 PM
conda is active in the container (i followed the article you linked above 😅), so running:
docker run -it image-tag
locally starts a shell with the conda env activated
k
Kevin Kho
05/17/2022, 9:41 PM
Ah ok. There must be something wrong with Flow spinup. Check this thread and there is a code example you can use to add logs to the Flow to get better error messages
h
Horatiu Bota
05/17/2022, 9:45 PM
thanks! already read through that and have logs enabled - the only log message i'm getting is
/bin/sh: /bin/sh: cannot execute binary file
k
Kevin Kho
05/17/2022, 9:48 PM
Ok I re-read again. I know what you are saying now. I am actually confused where
Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.