SQL Warehouse fails to start — RESOURCE_EXHAUSTED Error

[sharath007]

I'm unable to start my SQL warehouse (Serverless) due to a RESOURCE_EXHAUSTED error.

• Error message: Clusters are failing to launch. Cluster launch will be retried. Request to create a cluster failed with an exception: RESOURCE_EXHAUSTED: Cannot create the resource, please try again later.

I did try creating a new Warehouse but the issue remains the same.

Any recommendations or solutions would be helpful

[balajij8]

Databricks workspace's compute resources will be shut down and unavailable for the rest of the day (a month in extreme cases) if you exceed the quota. While the compute will be unavailable, Data and settings will not be deleted. You can resume the work when the limit resets. Free Edition accounts may not be used for commercial purposes

More details here

[sharath007]

I do have Premium plan and the SQL Warehouse was working fine till the last week. 

[balajij8]

It indicates a capacity constraint. Serverless quota is enforced at the regional level for all workspaces in your account. When you reach your serverless quota in a region, workspaces cannot launch new serverless SQL warehouses. You can request additional quota to help@databricks.com with the workspace, region & all other details.

[balajij8]

You can also make the request in the support portal if you have support plan.

[Ashwin_DSA]

Hi @sharath007,

Just checked internally. This specific RESOURCE_EXHAUSTED: Cannot create the resource, please try again later message for a serverless SQL warehouse normally indicates that the backing serverless compute pool has run out of capacity (either in your workspace’s quota or in the region), rather than an issue with your warehouse configuration.

Creating a new warehouse or changing its size typically won’t help, because all serverless warehouses in the same workspace/region draw from the same underlying pool.

A few things to try:

• Confirm which cloud + region the workspace is in, and check the Databricks status page for any active incidents in that region. The only active issue I can see is this.
• As a temporary workaround, try running queries on a pro/classic SQL warehouse or a regular all-purpose cluster instead of serverless.

If you’re on a paid Premium (or higher) plan and this has been happening for more than a short period, please open a Databricks Support ticket from within the workspace, including:

• Workspace URL (redact any secrets),
• Cloud + region,
• Approximate time range when you see the failures,
• The full error text from the SQL warehouse UI.

Support can then check whether your workspace has hit a serverless quota or encountered a regional capacity issue, and work with engineering on a fix or a quota increase.

If this answer resolves your question, could you mark it as “Accept as Solution”? That helps other users quickly find the correct fix.