topic Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target in Data Engineering

Databricks asset bundles and Dashboards - pass parameters depending on bundle target

erigaud — Mon, 02 Dec 2024 15:08:32 GMT

Hello everyone !

Since Databricks Asset Bundles can now be used to deploy dashboards, I'm wondering how to pass parameters so that the queries for the dev dashboard query the dev catalog, and the dashboard in stg query the stg catalog etc.

Is there any way to do this ?

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

Alberto_Umana — Mon, 02 Dec 2024 16:07:55 GMT

Hi @erigaud,

Here’s how you can achieve this:

Define Variables in the Bundle Configuration: You can define custom variables in your databricks.yml file to hold the catalog names for different environments. For example:

variables:

catalog_name:

description: "The catalog name to use for the environment"

default: "dev_catalog"

Use Targets to Override Variables: You can specify different values for these variables in the targets section of your bundle configuration. This allows you to set different catalog names for different environments (e.g., dev, stg, prod).

targets:

dev:

default: true

variables:

catalog_name: "dev_catalog"

stg:

variables:

catalog_name: "stg_catalog"

prod:

variables:

catalog_name: "prod_catalog"

Reference Variables in SQL Queries: In your SQL queries within the dashboards, you can reference these variables to dynamically set the catalog name. For example

SELECT * FROM ${var.catalog_name}.schema_name.table_name

Please refer to: https://docs.databricks.com/en/dev-tools/bundles/settings.html

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

erigaud — Tue, 03 Dec 2024 07:17:32 GMT

Hello, I tried modifying my query in the lvdash.json file corresponding to my dashboard with

"query": "SELECT * FROM mytable WHERE my_col = ${var.testvar}"

But it does not seem to be working and the variable is not replaced. I am doing something wrong ?

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

selva092 — Mon, 27 Jan 2025 13:39:53 GMT

@erigaud : Did you find any solution for this. I'm also trying the same as yours. Please share if you have any solution.

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

erigaud — Thu, 30 Jan 2025 16:05:56 GMT

Hi @selva092, I was able to contact someone from Databricks that explained that it is currently not supported, however it is in the backlog so hopefully it wiill come soon ! We just have to monitor the release notes for asset bundle until it's available

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

Bram123 — Thu, 12 Jun 2025 11:20:06 GMT

HI @erigaud, do you perhaps know if it is already possible at the moment?

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

JonA — Thu, 11 Sep 2025 08:04:06 GMT

Hello, do you know if there is a solution?

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

BramLust — Thu, 11 Sep 2025 09:37:18 GMT

Hi, it was discussed yesterday in a champions call. They confirmed that it is something that they will end but they don't know yet when the feature will be available. They recommended that if you really need it, to use terraform.

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

jelmer — Wed, 05 Nov 2025 06:44:04 GMT

Has this been added yet ? having dashboards in asset bundles without support for parameterization is borderline broken

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

chris_koester — Thu, 06 Nov 2025 15:27:49 GMT

There is a limited workaround contained in the repo folder linked below. See the README for details.

https://github.com/databricks-solutions/databricks-dab-examples/tree/main/knowledge-base/metric-views

Product and engineering are aware of this issue and are working on a solution.

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

Coffee77 — Thu, 06 Nov 2025 16:28:21 GMT

What I did as a workaround. It works pretty fine but you'll need to duplicate Dashboard JSON code per environment and then, replace catalog names 🙄 It is not the perfect solution but the only way I could find to include these deployment in my Databricks Asset Bundles (DAB).

1) Include dashboards in DAB:

2) Define dashboards in there similar to this one. I usually create or update dashboards directly in Databricks UI and then, import JSON source code:

3) Then, include the JSON files associated to your dashboards in proper paths depending on environments. In my case, I just need to replace catalog names per environment. It takes me less than one minute although waiting for a better solution without duplicating code...

I hope this helps

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

protmaks — Mon, 05 Jan 2026 10:01:35 GMT

In the new version v0.281.0, catalog and schema parameterization for the Databricks Dashboard finally works. I tested it and wrote examples - https://medium.com/@protmaks/dynamic-catalog-schema-in-databricks-dashboards-b7eea62270c6

Re: Databricks asset bundles and Dashboards - pass parameters depending on bundle target

masoomeh_gh — Tue, 10 Feb 2026 09:39:08 GMT

This worked for me, once I upgraded the CLI to v0.287.0. I am using two schemas in the lvdash.json file but only one catalog. What I did was to only parametrize the catalog name in the yml file,

variables: warehouse_id: description: SQL Warehouse ID default: xxxx catalog_name: description: Catalog name default: dev_gold resources: dashboards: dashboard_example: display_name: 'Dashboard_Demo' file_path: ../../dashboards/my_dashboard.lvdash.json warehouse_id: ${var.warehouse_id} dataset_catalog: ${var.catalog_name} targets: dev: mode: development default: true workspace: host: https://adb-xxx.azuredatabricks.net variables: catalog_name: dev_gold warehouse_id: xxx prod: mode: production workspace: host: https://adb-yyy.azuredatabricks.net variables: catalog_name: gold warehouse_id: yyy

and once I deployed, it could pick up the right schema names. In the lvdash.json file, I only omitted the catalog name and left the schema names, as follows:

"queryLines": [ "SELECT * FROM schema_1.table1" ] ... "queryLines": [ "SELECT * FROM schema_2.table2" ]