I need to customize the container that gets deployed to dagster cloud a bit. I need it to run

meltano install

after adding the code to the image in the

Dockerfile

. I was thinking I could: 1. Build the standard image using

dagster-cloud serverless build

2. Tag the image created something like

dagsterstandard

3. Create a new Dockerfile that looks like:

FROM dagsterstandard
RUN cd meltano && meltano install

4. Tag that custom image with the same tag as the standard image and push that to the dagster serverless registry. Think that will work? Any gotchas?

@prha is the expert on this

I fixed the exact same problem just 1 hour ago. I made some adjustments to the Dockerfile template in our own fork, which gives you the possibility to run your shell script, if it exists. In this file I install all required dependencies and after that install meltano. But this was all really cumbersome. It would be great to have better support for these custom actions.

ok we’re going to clean this up this week and enable support for custom base images

So you had to fork the Dockerfile template? Which one specifically?

For now I forked the `dagster-cloud-action`image. And then pushed this image as

<http://ghcr.io/|ghcr.io/><our-org>/dagster-cloud-action

. And used this image in the Github action. You can take a look here: https://github.com/quantile-development/dagster-cloud-action But it's all a bit hacky. It might be better to wait for official support 🙃

I added support in the

dagster-cloud serverless

CLI to provide a custom base image, as well as support for running a

dagster_cloud_pre_install.sh

/

dagster_cloud_post_install.sh

script inside of the target directory if they exist. I’m in the process of adding support to our GH action as well. You should be able to run:

dagster-cloud serverless deploy --base-image my_base_image <target_dir>

Great, thank you very much!

Cool, so we should be able to use any base image? Will the container still run

pip install -r requirements.txt

, or will that be assumed to have already been installed?

It will still run

pip install -r requirements.txt

if it exists