Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Provide a Docker image blueprint for air-gapped environments #3618

Closed
Tracked by #3755
masci opened this issue Nov 23, 2022 · 4 comments
Closed
Tracked by #3755

Provide a Docker image blueprint for air-gapped environments #3618

masci opened this issue Nov 23, 2022 · 4 comments
Assignees
Labels

Comments

@masci
Copy link
Contributor

masci commented Nov 23, 2022

Is your feature request related to a problem? Please describe.
It's common to run Haystack behind its rest_api within a container that doesn't have access to the Internet. In these situations the solution would be creating a custom Docker image that incorporates any resource that Haystack would normally download as it's needed like models.

This approach is fully supported by the new Docker images but users have to figure out by themselves how to do this.

Describe the solution you'd like
Provide a blueprint in form of a Dockerfile showing how you would build an image capable of running a pipeline without internet access.

Describe alternatives you've considered
We could provide a CLI tool that guides the user through the process but I think we should start smaller.

Additional context
/cc @ju-gu for visibility

@mayankjobanputra
Copy link
Contributor

Two ways:

leave comment large docker image

  1. Cache HF Model
  2. Local model path - copy model inside, not mount -

How would I test? - Run the API using gunicorn, call the API through a python script?

cache_model - telemetry changes

add simpler pipeline - reader pipeline without elasticsearch

@masci masci added P2 Medium priority, add to the next sprint if no P1 available and removed P1 High priority, add to the next sprint labels Mar 9, 2023
@masci
Copy link
Contributor Author

masci commented Apr 12, 2023

@mayankjobanputra I remember you wanted to keep this issue open to track some unfinished work, is still the case?

@mayankjobanputra
Copy link
Contributor

Yesss. We can talk about it whenever you have time. :)

@masci masci added P3 Low priority, leave it in the backlog and removed P2 Medium priority, add to the next sprint if no P1 available labels May 10, 2023
@mayankjobanputra
Copy link
Contributor

Fixed in this PR: deepset-ai/haystack-demos#18

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

2 participants