BentoML

Thanks for reporting. 

<@U0122472WCV> <@U01BMCJ29LN> can you guys help take a look

<@UKH494PHP> Hi, any chance that the team has been able to look at this? Thank you!

Hi <@UKH494PHP> just wanted to close the loop and let you know that I solved the issue. It was a lack of disk space. The default disk space is 10 GB, and this was too small. I modified the `<http://main.tf|main.tf>` file by adding `size = 20` in the `boot_disk { initialize_params {` section, and it deployed without issue very quickly after that.

hey <@U046TQDUBAT> qq what is the size of the bento and docker image ur deploying? Maybe we should bump up the default size to 20Gi

Hi <@U0122472WCV> that's a good question, when containerizing and uploading, it was 3.3 GB, but because I was using the huggingface pretrained model, which was also afterwards downloading from HF, I assume that that might have tipped the scale. I only thought of expanding the disk size because the logs kept stating that they couldn't write the errors due to lack of disk space.

got it, Thanks for sharing the solution and the detailed explanation. We should probably make explicit in the terraform template. Let me know if you are interesting in making that change or I'll add it too

<@U0122472WCV> Would love to give it a shot!