Azure / Azure/azure-cli

az ml online-deployment update Cant update/removes tag

Open
#33,834 1 comment 0 reactions 0 assignees View on GitHub
Machine Learning Service Attention
Dominant language
Python
Stars
4.6k
Forks
3.5k
Avg merge
3d 2h
Merged PRs (30d)
60

Description

**Related command**
`az ml online-deployment update --set tags.tag=tag`

**Is your feature request related to a problem? Please describe.**
Hi,
For a ML deployment I want to use a tag as a hash fingerprint so can check if the new a new pipeline run has to update the deployment or not.

Currently tags can be added when a deployment is created
`az ml online-deployment create -f "deployment.yml --set tags.tag=tag`

And these tags can be requested through
`az ml online-deployment show -n deployment-name --endpoint-name endpoint-name --workspace-name --query "[tags.tag]"`

But the update command:
` az ml online-deployment update --set tags.tag=tag`
Throws the following error:
`ERROR: Couldn't find 'tags' in ''. Available options: ['environment']`

Seems the update command cannot set any tags, but only update the environment listed to them. Beyond that if the tags part is left out, and the deployment is updated it will remove any preexisting tags.

**Describe the solution you'd like**
For the update command to support tags to pull the command inline with the create command

Contributor guide

Open the contributing guide

Research direction

Start by reproducing the az ml online-deployment update command with --set tags.tag=tag, then compare its handling with the create command and the deployment update behavior when tags are omitted. The work is done when update accepts tags and preserves existing tags; the issue does not name a source file or test.

Written by the indexing model from the issue text.

Assessment

Tech stack
azure, python
Domain
cli, cloud
Issue type
Feature
Difficulty
3/5
Estimated time
1-2 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
56/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.