Pandas library failing to import in Automation Account

[molten nacelle]

Why not use a service that natively includes Pandas like Azure Databricks or Synapse?

[twin cradle]

Need it in a runbook to run everyday @molten nacelle

[molten nacelle]

You can do that in either Syanpse or ADB.

[twin cradle]

So my use case is sending email alerts based on approaching compute quotas across our entire company wide AZ account. We have it working to where it imports to log analytics through log collector API and then created an Alert rule off the num results for the custom log/table. Is there a better way of doing this?

We spoke with a MS rep who said what we’re doing is the best way

But if they’re wrong I’d love to know a better way. Note: we can’t use sendgrid.

[molten nacelle]

that sounds like the right path, where does the function + pandas fit in?

[twin cradle]

Okay sorry. That’s a sep problem. My mind is everywhere this morning.

I got pandas working.

We are running a script to check for tags on all resources created in the last 24 hours. And then emailing which ones don’t fall under compliance

However I can only get it to run locally

When I run locally with ResourceGraph.ResourceGraphClient(AzureCliCredential) it works but when trying to use DefaultAzureCredential in a runbook it doesn’t work and throws an error

@molten nacelle

[molten nacelle]

can you provide the error? So pandas imports fine its just the auth step that fails?

[twin cradle]

I closed this issue and opened a new one call Authenticate ResourceGraph