Azure / Azure/azure-cli-extensions
`az ml compute show` is returning incorrect and more limited info than previously available (e.g. ssh connection info)
- Dominant language
- Python
- Stars
- 454
- Forks
- 1.7k
- Avg merge
- 2d 19h
- Merged PRs (30d)
- 64
Description
### Extension name (the extension in question)
ml (v2)
### Description of issue (in as much detail as possible)
There is no way to obtain ssh connectivity information once upgraded to cli v2 (verbose option is missing)
Re command:
```
az ml compute show -n
```
This command incorrectly reflects if ssh/public ip is enabled for a computeinstance, and no longer offers the -v option that was available in `az ml computetarget show` (v1) to return the ip address and port connectivity info to connect to a compute instance.
### information returned by v2
*** extension version: "ml": "2.0.1a5" ***
command:
```
az ml compute show -n myDS3 --verbose
```
Response:
```
{
"admin_username": "azureuser",
"applications": [
{
"display_name": "Jupyter",
"endpoint_uri": "https://myds3.uksouth.instances.azureml.ms"
},
{
"display_name": "Jupyter Lab",
"endpoint_uri": "https://myds3.uksouth.instances.azureml.ms/lab"
},
{
"display_name": "RStudio",
"endpoint_uri": "https://myds3-8787.uksouth.instances.azureml.ms"
}
],
"enable_public_ip": false, ## <- INCORRECT
"id": "/subscriptions/cf48479b-e351-4599-8588-4867340afcce/resourceGroups//providers/Microsoft.MachineLearningServices/workspaces//computes/myDS3",
"last_operation": {
"operation_name": "Create",
"operation_status": "Succeeded",
"operation_time": "2021-08-19T15:47:01.251Z"
},
"location": "uksouth",
"name": "myDS3",
"provisioning_state": "Succeeded",
"resourceGroup": "",
"size": "STANDARD_DS3_V2",
"ssh_public_access": "False", ## <- INCORRECT
"state": "Running",
"tags": {},
"type": "computeinstance"
}
```
### information returned by v1 for comparison
*** version azure-cli-ml 1.33.0***
command:
```
az ml computetarget show -g -w -n myDS3 -v
```
Response:
```
{
"id": "/subscriptions/cf48479b-e351-4599-8588-4867340afcce/resourceGroups//providers/Microsoft.MachineLearningServices/workspaces//computes/myDS3",
"location": "uksouth",
"name": "myDS3",
"properties": {
"computeLocation": "uksouth",
"computeType": "ComputeInstance",
"description": null,
"properties": {
"applications": [
{
"displayName": "Jupyter",
"endpointUri": "https://myds3.uksouth.instances.azureml.ms"
},
{
"displayName": "Jupyter Lab",
"endpointUri": "https://myds3.uksouth.instances.azureml.ms/lab"
},
{
"displayName": "RStudio",
"endpointUri": "https://myds3-8787.uksouth.instances.azureml.ms"
}
],
"connectivityEndpoints": {
"privateIpAddress": "10.0.0.4",
"publicIpAddress": "51.105.11.201"
},
"errors": [],
"personalComputeInstanceSettings": null,
"sshSettings": {
"adminPublicKey": "",
"adminUserName": "azureuser",
"sshPort": 50000,
"sshPublicAccess": "Enabled"
},
"subnet": {
"id": null
},
"vmSize": "STANDARD_DS3_V2"
},
"provisioningErrors": null,
"provisioningState": "Succeeded",
"resourceId": null,
"status": {
"createdBy": {
"userName": "User Name",
"userObjectId": "***8a96",
"userTenantId": "***db47"
},
"creationTime": "2021-08-19T15:46:55.328165+00:00",
"errors": [],
"modifiedTime": "2021-08-19T15:48:56.764333+00:00",
"state": "Running",
"vmSize": "STANDARD_DS3_V2"
}
},
"resourceGroup": "",
"tags": {}
}
```
## Question:
How do I get the sshSettings and connectivityEndpoints objects from the command line once upgraded to ml v2 please?
-----
Contributor guide
Assessment
This issue has not been assessed yet.