Explanation of container app target and image build? #1445
-
We are working on various samples for Azure Container Apps, and we're struggling to understand how azd is able to deploy an image from the app code to ACR and reference that image for ACA in a single deploy. That seems to be what the TODO ACA apps do, as you can do a deploy and see the ACAs reference an image stored in the newly created ACR, but we don't understand what magic helps that happen. Relevant files: Do we need to set particular environment variable names? Do we need to structure the Bicep in a particular order, to make sure image exists first? Thanks! |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 5 replies
-
|
Beta Was this translation helpful? Give feedback.
-
Dang, could this be added to the documentation? I had similar questions about this seemingly magical behavior so that I could add new services to my project and modify the bicep and azure.yaml accordingly. I spent quite a bit of time scouring the docs, watching vids, and just tinkering...all to no avail, and finally stumbled across this issue only after I had given up all hope! |
Beta Was this translation helpful? Give feedback.
@pamelafox
SERVICE_<service name>_IMAGE_NAME
.azure.yaml
. azure.yaml, bicep examples. Note that this bicep file does not need to be referenced bymain.bicep
.parameters.json
for your ACA deployment that usesSERVICE_<service name>_IMAGE_NAME
as a parameter value substitution. Example.