1

MarkLogic publishes its DB to Docker Hub.

I would like to explore how to run ML docker hub images on ACI.

I try to follow below link to do it. (It works with that Sample Microsoft aci-helloword image deployment. I assume it should also work for ML.) However I got below error message.

{"code":"DeploymentFailed","message":"At least one resource deployment operation failed. Please list deployment operations for details. Please see https://aka.ms/DeployOperations for usage details.","details":[{"code":"InaccessibleImage","message":"The image 'store/marklogicdb/marklogic-server:10.0-8.1-centos-1.0.0-ea2' in container group 'ml-container' is not accessible. Please check the image and registry credential."}]}

Create Container Instance Detail

Network Detail

Improve this question

2 Answers 2

Reset to default
1

The issue is probably with image type. It's free, but since you're required to subscribe, it's not private. Try with private image type and you'll likely need to authenticate against Docker Hub. We have a detailed example at https://github.com/marklogic/marklogic-docker but I'm not sure how you can setup private image access on Azure.

Improve this answer
Sign up to request clarification or add additional context in comments.

3 Comments

Per a side conversation that I had with @vitaly, I wasn't able to get it to work with ACI even with it set to a private image and setting the credentials correctly. We will look into what the issue might be but in the meantime, a workaround is to pull the image locally and then push to your own repo. I tested that and it works fine.
James, may I confirm whether the ML DB docker images work with ACI? Do I need to mount an Azure file share to ACI to make it work? (learn.microsoft.com/en-us/azure/container-instances/…). I guess it is a good way to set up non Production ML with ACI.
Follow James' suggestion, it indeed works to deploy to ACI once I push the ML image to my own repo. However ML keeps asking for the login. The behavior is different from Docker Desktop. (I will need to input multiple login informations in order to run any XQuery.) What is wrong? i.imgur.com/Hz9z7Gj.png
0

I think Azure Container Instances Linux is based on ubuntu and there are some known limitations. That may be the reason why there are still some compatibility issues of deploying ML docker image to run in ACI.

It is pretty straightforward and cost-effective to follow the below guide to run ML in Azure. https://www.marklogic.com/resources/deploy-on-azure/

ML regularly update its azure images in the marketplaces. https://azuremarketplace.microsoft.com/en-in/marketplace/apps/marklogic.marklogic-developer-10?tab=Overview As a result, there is no significate benefit of enabling ML to run in ACI.

Improve this answer

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.