Hey everyone! I'm currently working on a project that relies on Python 3.7, and we've recently transitioned from using Docker images on EC2 to an AWS Graviton instance. Unfortunately, my code is now failing because it relies on Python 3.7 and its associated packages. We tested everything in the DEV environment, and I'm stuck at a crossroads. Here are my main options I'm considering:
1. Go back to using Docker images.
2. Roll back from the Graviton instance.
3. Upgrade my project to Python 3.10, but that would mean a lot of coding effort since it's been stable in production for quite a while.
I'm looking for a better solution, any suggestions?
3 Answers
What made you switch to Graviton if you're facing all these challenges with your existing code? This seems more like a downgrade. Upgrading from Python 3.7 to 3.10 shouldn't be too complicated, unlike moving from Python 2 to 3, which was a whole different ball game.
Yes, you can definitely create a virtual environment on Graviton instances! I've been using ARM64 and haven't encountered any major issues. Can you provide more details on the errors you're facing? Are you having trouble with installing the requirements, or is it a runtime error? Remember, just upgrading to Python 3.10 won't fix existing issues since Python 3.7 should still work on Graviton.
I'm running into errors while installing the numba and llvmlite packages.
Honestly, Python 3.7 is outdated and no longer supported as of June 2023. It might be a tough call, but upgrading now could save you from bigger headaches in the future—security vulnerabilities, build issues, you name it. Trust me, you'll thank yourself later for the extra work now!
I get that, but this project is complex and really stable in production right now. We just don’t have the bandwidth to enhance it at this moment.
Could you share some steps so I can help resolve this?