https://ploomber.io/ logo
Join Slack
Powered by
Hello here :wave: I have subscribed Ploomber pro t...
# general
u
Hello here 👋 I have subscribed Ploomber pro today. I would like to update the ressources of my app. In particular, I would like to increase the memory from 
ploomber-cloud.json
. How can I proceed ?
I am currently not able to deploy my app using Docker, but I faced with an issue about lack of memory
y
Hey @Urban AI, thanks for joining Ploomber and welcome! You can modify the resources of an existing project by running:
Copy code
pc resources
Then, simply deploy it with:
Copy code
pc deploy
For more details, check out our documentation. Let me know if there's anything else I can help you with!
u
Thx for your quick answer ! Yes that's what I did. Here are the ressources I set :
Copy code
"resources": {
        "cpu": 4.0,
        "ram": 8,
        "gpu": 0
    }
However it turned out that I am not able to deploy unfortunately
It still fails due to insufficient memory during this step :
Copy code
2025-05-25T09:59:02.907000+00:00 - INFO Taking snapshot of full filesystem...
Error: Internal server error. Please contact support: <mailto:contact@ploomber.io|contact@ploomber.io>
If you need help solving this issue, send us a message: <https://ploomber.io/community>
after downloading the dependencies
y
Got it. let me check this and I'll get back to you shortly.
u
Screenshot 2025-05-25 at 12.11.05.png
i
Are you using pytorch? It looks like it fails during the env build stage.
u
Yeah unfortunately... HuggingFace integration from langchain has changed so that I should install this unnecessary package. What is the error precisely ? 🙏
In parallel, I am trying to investigate more. At least, thx for raising this problem
y
Hey @Urban AI I'm still looking into this. I see you are using a specific version of torch. Please see this FAQ about deploying apps with PyTorch
u
Thanks a lot for investigating. It will be very useful. I have finally succeeded to remove torch 5 minutes ago because I found another workaround without using torch / pytorch. However it fails again when re-deploying the same application 
lingering-moon-2237
Should I re-create a new application ?
i
Try using torch-cpu that should probably fix it.
u
Ok thanks for your advice
i
Were you able to deploy? What are you trying to do exactly? It’d be helpful to get context so we can help you out.
u
I was not able to deploy yet. I changed a bit the logic so that I don't have to use pytorch (because it increases the Docker image size). However, even without this package, I faced with the same issue. Currently, I am ensuring to take the minimal packages I need. Then I am going to deploy a new app. I hope it will solve my issue
i
Makes sense, let us know if you need more help, it sounds like you’re on the right track 👍
u
Yes sure ! Normally it should be fine hopefully 🤞
Well it works as expected 🎉
Open in Slack
PreviousNext
Powered by