Serverless warehouse suddenly stops to start up.

Question

Hey All.

From today, suddenly we are getting below error while starting a serverless warehouse.

Details for the latest failure: Error: Cluster launch timeout. Type: SERVICE_FAULT Code: K8S_DBR_CLUSTER_LAUNCH_TIMEOUT

Warehouse details:
Type: Serverless

Cluster size: X-Small

Auto stop: After 5 minutes of inactivity

Channel: Current(v 2024.25)

There was no such issue till today. Can anyone guide us on how to resolve this? Thanks in advance.

Answer 1

@Bhargav Shir - Thanks for the question and using MS Q&A platform.

It seems like you are facing an issue with creating a Databricks cluster. The error message you provided indicates that there was a cluster launch timeout. This error can occur when the cluster takes too long to start up, and the launch process times out.

To resolve this issue, you can try the following steps:

1. This could be an intermittent issue, could you please retry to start the serverless warehouse.
2. Check if there are any issues with your network connectivity. You can do this by checking if you are able to access other resources on the internet.

Verify that you have egress/outbound network connectivity from your VPC to Databricks Control plane. Most likely VPC firewall is blocking egress communication. You can check if there are any firewall rules that are blocking outbound traffic from your VPC to Databricks Control plane.

Make sure that you have a network route set for VPC to communicate with Databricks control plane. You can check if there are any network routes that are missing or misconfigured.

Make sure that an egress appliance like Cloud NAT is attached to subnets used by Databricks. You can check if there are any egress appliances that are missing or misconfigured.

For more details, refer to Azure databricks - Troubleshooting.

If none of these step's work, you can open a support ticket for further assistance.

I hope this helps! Let me know if you have any other questions.