cancel
Showing results for 
Search instead for 
Did you mean: 
Data Engineering
Join discussions on data engineering best practices, architectures, and optimization strategies within the Databricks Community. Exchange insights and solutions with fellow data engineers.
cancel
Showing results for 
Search instead for 
Did you mean: 

SUBNET_EXHAUSTED_FAILURE(CLOUD_FAILURE): or No more address space to create NIC within injected virtual network

Phani1
Databricks MVP

Currently we are using an all-purpose compute cluster. When we tried to allocate the scheduled jobs to job cluster, we are blocked at the following error:

SUBNET_EXHAUSTED_FAILURE(CLOUD_FAILURE): azure_error_code:SubnetIsFull,azure_error_message:No more address space to create NIC within injected virtual network. Details: {  "error": {   "code": "SubnetIsFull",   "message": "Subnet public-subnet-databricks-wwmddatabricksnonprod01 with address prefix 10.99.0.0/26 does not have enough capacity for 1 IP addresses.",   "details": []  } },databricks_error_message:Error code: SubnetIsFull, error message: No more address space to create NIC within injected virtual network. Details: {  "error": {   "code": "SubnetIsFull",   "message": "Subnet public-subnet-databricks-wwmddatabricksnonprod01 with address prefix 10.99.0.0/26 does not have enough capacity for 1 IP addresses.",   "details": []  } } 

We raised a service request to the cloud vendor, Microsoft recommended updating the CIDR range on the Databricks workspace to allocate more no. of IPs to the Network address.

Could you please help us to clarify the below questions?

1)What all are data bricks objects will impact/have to take care of during the updating CIDR range on the Databricks workspace to allocate more no. of IPs to the Network address space

2)Is it the right approach to what we are doing, or is there any alternate way?

space.

2 REPLIES 2

daniel_sahal
Databricks MVP

Answering your questions - yes, your vnet/subnet is out of non-occupied IPs and this can be fixed by allocating more IPs to your network address space.

Each cluster requires it's own IP, so if there are none available, it simply cannot start.

Thanks for your answer!!

What all objects will be impacted in data bricks while applying the changes/increasing IPs (Clusters should not be in running status, mount location, etc). so that will take appropriate action before applying the changes.