Monitor the Status of Lakehouse¶

In DataOS, the status of a Resource indicates its current life-cycle state, such as active, error, or deleted. Monitoring status allows teams to detect state transitions (e.g., from active to deleted or error) that may impact downstream dependencies, trigger configuration issues, or reflect access problems.

DataOS CLI¶

The status of a Lakehouse can be monitored using the DataOS CLI by executing the following command, replacing the placeholder with the workspace in which the user is working. This command will list all the Lakehouse Resources created by the user.

dataos-ctl get -t lakehouse -w ${{workspace-name}}

Example Usage:

dataos-ctl get -t lakehouse -w public
INFO[0000] 🔍 get...                                     
INFO[0001] 🔍 get...complete                             

        NAME       | VERSION |      TYPE       | WORKSPACE | STATUS | RUNTIME |    OWNER     
-------------------|---------|-----------------|-----------|--------|---------|--------------
  lakehouse      | v1alpha |   lakehouse     |  public   | active |         | iamgroot

In the example above, the STATUS column indicates the current state of the Lakehouse active in this case, which confirms that the Lakehouse is available and usable.

Metis UI¶

To monitor the status of Lakehouse on the Metis Catalog UI, follow the steps below:

Open the Metis Catalog.

Open the Metis Catalog
Search for a Lakehouse by name.

Search the Lakehouse in Metis
Click on the Lakehouse that needs to be monitored and check the Status, which is represented as State.

Lakehouse details in Metis

The active state indicates that the Lakehouse is currently accessible and usable. This is the expected state for any Lakehouse Resource.

Operations App¶

To monitor the status of Lakehouse on the Operations app, follow the steps below:

Open the Operations app.

Open the Operations app
Under the User space → type → Lakehouse, search for the Lakehouse by name.

Operations > User Space > Lakehouse

The active status indicates that the Resource is valid and available for use within DataOS. This is the expected state for a healthy and usable Resource.
On clicking the Lakehouse, its builder state can also be monitored.

🗣 The Builder Stage reflects the internal progress of a DataOS Resource as it is being reconciled and provisioned by the platform. This stage is managed by the Poros controller, which is responsible for ensuring the system’s actual state matches the desired state defined in the Resource manifest (YAML).

When a user applies a Resource YAML (dataos-ctl resource apply -f resource.yaml), the builder workflow begins. Poros orchestrates this by comparing the input state (what the user requested) with the current cluster state (what already exists), and attempts to reconcile the two.

If a Resource enters an error state during this stage, it means something failed while setting it up. A Resource in an error state at the Builder Stage is considered not fully created and should not be treated as active, even if it appears in the UI list.

Lakehouse details in Operations

Monitoring the Builder Stage is recommended when the Resource status shows an error.
- If the status is error and the Builder Stage is still building, it indicates that the issue occurred after the building stage.
- If the status is error and the Builder Stage also shows error, it means the issue happened during the building stage itself.
This helps in identifying whether the problem lies within the building phase or after the Resource was built.

Status alerts¶

To automatically track critical state transitions, users can configure a Monitor and Pager to send alerts when the status of a Lakehouse changes to values like error or deleted. This enables teams to respond immediately to resource failures, misconfigurations, or unexpected deletions that may impact dependent components. Click here to view the steps to set up alerts for status changes.