Datetime dependent error when creating pipeline

[GregorWautischer]

Describe the bug

Trying to create a job using az ml job craete -f hello_pipeline.yaml where hello_pipline.yaml contains (modified from an example found here):

$schema: https://azuremlschemas.azureedge.net/latest/pipelineJob.schema.json
display_name: hello_pipeline
type: pipeline
inputs:
  somevariable: "2023-01-01T00:00:00Z"
jobs:
  hello_job:
    command: echo "hello"
    environment: azureml:AzureML-sklearn-1.0-ubuntu20.04-py38-cpu@latest
    compute: computename
  world_job:
    command: echo "world"
    environment: azureml:AzureML-sklearn-1.0-ubuntu20.04-py38-cpu@latest
    compute: computename
settings:
  default_datastore: datastorename

the below error is thrown even though the created pipeline runs successfully in Azure Machine Learning. The error disappears when the value of somevariable is changed to "2023-01-01 00:00:00Z" with the pipeline still running successfully. This behaviour is independent of the name of the variable (somevariable here).

Related command

az ml job create -f pipeline.yaml

Errors

(UserError) A job with this name already exists. If you are trying to create a new job, use a different name. If you are trying to update an existing job, the existing job's Settings, Inputs, Outputs, Jobs, InputBindings, OutputBindings cannot be changed. Only description, tags, displayName, properties, and isArchived can be updated.
Code: UserError
Message: A job with this name already exists. If you are trying to create a new job, use a different name. If you are trying to update an existing job, the existing job's Settings, Inputs, Outputs, Jobs, InputBindings, OutputBindings cannot be changed. Only description, tags, displayName, properties, and isArchived can be updated.
Additional Information:Type: ComponentName
Info: {
"value": "managementfrontend"
}Type: Correlation
Info: {
"value": {
"operation": "XXXXX",
"request": "XXXX"
}
}Type: Environment
Info: {
"value": "SOMELOCATION"
}Type: Location
Info: {
"value": "SOMELOCATION"
}Type: Time
Info: {
"value": "2023-07-10T12:25:55.5257915+00:00"
}Type: InnerError
Info: {
"value": {
"code": "Immutable",
"innerError": {
"code": "JobPropertyImmutable",
"innerError": null
}
}
}Type: MessageFormat
Info: {
"value": "A job with this name already exists. If you are trying to create a new job, use a different name. If you are trying to update an existing job, the existing job's{property} cannot be changed. Only description, tags, displayName, properties, and isArchived can be updated."
}Type: MessageParameters
Info: {
"value": {
"property": " Settings, Inputs, Outputs, Jobs, InputBindings, OutputBindings"
}
}

Issue script & Debug output

I did not add debug output since it is really long and I don't want to go through the whole thing to remove sensitive information at this point.

Expected behavior

Creation of the pipeline should work when using the date time format "YYYY-mm-ddThh:mm:ssZ". However, why are variables under inputs parsed for datetime formats irrespective of their variable name?

Environment Summary

azure-cli 2.50.0

core 2.50.0
telemetry 1.0.8

Extensions:
ml 2.18.0

Dependencies:
msal 1.22.0
azure-mgmt-resource 23.1.0b2

Python location '/opt/az/bin/python3'
Extensions directory '/opt/az/extensions'

Python (Linux) 3.10.10 (main, Jun 29 2023, 11:09:34) [GCC 9.4.0]

Additional context

No response

[azure-client-tools-bot-prd]

Hi @GregorWautischer,

This is not the official Azure CLI published by Microsoft.

How to tell if the installed Azure CLI is unofficial:

• Unofficial Azure CLI deb packages are distributed by Linux repositories #19640 (comment)
• Unofficial Azure CLI deb packages are distributed by Linux repositories #19640 (comment)

Please follow https://docs.microsoft.com/en-us/cli/azure/install-azure-cli-linux to install Microsoft official Azure CLI.

If you feel that further discussion is needed, please add a comment with the text /clibot unresolve to reopen the issue.

[yonzhan]

Thank you for opening this issue, we will look into it.

[GregorWautischer]

/clibot unresolve

Thanks for the feedback! We are routing this to the appropriate team for follow-up. cc @azureml-github.

Issue Details

---

Describe the bug

Trying to create a job using az ml job craete -f hello_pipeline.yaml where hello_pipline.yaml contains (modified from an example found here):

$schema: https://azuremlschemas.azureedge.net/latest/pipelineJob.schema.json
display_name: hello_pipeline
type: pipeline
inputs:
  somevariable: "2023-01-01T00:00:00Z"
jobs:
  hello_job:
    command: echo "hello"
    environment: azureml:AzureML-sklearn-1.0-ubuntu20.04-py38-cpu@latest
    compute: computename
  world_job:
    command: echo "world"
    environment: azureml:AzureML-sklearn-1.0-ubuntu20.04-py38-cpu@latest
    compute: computename
settings:
  default_datastore: datastorename

the below error is thrown even though the created pipeline runs successfully in Azure Machine Learning. The error disappears when the value of somevariable is changed to "2023-01-01 00:00:00Z" with the pipeline still running successfully. This behaviour is independent of the name of the variable (somevariable here).

Related command

az ml job create -f pipeline.yaml

Errors

(UserError) A job with this name already exists. If you are trying to create a new job, use a different name. If you are trying to update an existing job, the existing job's Settings, Inputs, Outputs, Jobs, InputBindings, OutputBindings cannot be changed. Only description, tags, displayName, properties, and isArchived can be updated.
Code: UserError
Message: A job with this name already exists. If you are trying to create a new job, use a different name. If you are trying to update an existing job, the existing job's Settings, Inputs, Outputs, Jobs, InputBindings, OutputBindings cannot be changed. Only description, tags, displayName, properties, and isArchived can be updated.
Additional Information:Type: ComponentName
Info: {
"value": "managementfrontend"
}Type: Correlation
Info: {
"value": {
"operation": "XXXXX",
"request": "XXXX"
}
}Type: Environment
Info: {
"value": "SOMELOCATION"
}Type: Location
Info: {
"value": "SOMELOCATION"
}Type: Time
Info: {
"value": "2023-07-10T12:25:55.5257915+00:00"
}Type: InnerError
Info: {
"value": {
"code": "Immutable",
"innerError": {
"code": "JobPropertyImmutable",
"innerError": null
}
}
}Type: MessageFormat
Info: {
"value": "A job with this name already exists. If you are trying to create a new job, use a different name. If you are trying to update an existing job, the existing job's{property} cannot be changed. Only description, tags, displayName, properties, and isArchived can be updated."
}Type: MessageParameters
Info: {
"value": {
"property": " Settings, Inputs, Outputs, Jobs, InputBindings, OutputBindings"
}
}

Issue script & Debug output

I did not add debug output since it is really long and I don't want to go through the whole thing to remove sensitive information at this point.

Expected behavior

Creation of the pipeline should work when using the date time format "YYYY-mm-ddThh:mm:ssZ". However, why are variables under inputs parsed for datetime formats irrespective of their variable name?

Environment Summary

azure-cli 2.50.0

core 2.50.0
telemetry 1.0.8

Extensions:
ml 2.18.0

Dependencies:
msal 1.22.0
azure-mgmt-resource 23.1.0b2

Python location '/opt/az/bin/python3'
Extensions directory '/opt/az/extensions'

Python (Linux) 3.10.10 (main, Jun 29 2023, 11:09:34) [GCC 9.4.0]

Additional context

No response

Author:	GregorWautischer
Assignees:	-
Labels:	bug, Service Attention, Machine Learning, customer-reported, needs-team-attention, Auto-Assign, Auto-Resolve
Milestone:	-