Environment Variables

Main configuration

Name	Description
`DEKART_MAPBOX_TOKEN`	Mapbox Token to show a map
`DEKART_POSTGRES_DB`	Database name. Dekart needs Postgres Database to store query meta information. Example: `dekart`
`DEKART_POSTGRES_HOST`	Example: `localhost`
`DEKART_POSTGRES_PORT`	Example: `5432`
`DEKART_POSTGRES_USER`	Example: `postgres`
`DEKART_POSTGRES_PASSWORD`	Example: `******`
`DEKART_PORT`	Example: `8080`
`DEKART_POSTGRES_URL` version >= 0.13	Alternatively to specify `DEKART_POSTGRES_DB`, `DEKART_POSTGRES_HOST`, `DEKART_POSTGRES_PORT`, `DEKART_POSTGRES_USER`, `DEKART_POSTGRES_PASSWORD`, configure PostgreSQL by passing the connection string. If both specified `DEKART_POSTGRES_URL` is used. Example: `postgres://user:pass@hostname:5432/dekart?sslmode=verify-full`
`DEKART_DATASOURCE=BQ` version >= 0.8	Which datasource to use: Values `BQ` BigQuery, default `ATHENA` AWS Athena `SNOWFLAKE` Snowflake version >= 0.12 `PG` Postgres version >= 0.16
`DEKART_STORAGE=GCS` version >= 0.8	Which storage backend to use for storing queries and query results Values `GCS` Google Cloud Storage, default, works only with BigQuery data source `S3` AWS S3, works with BigQuery and AWS Athena
`DEKART_CLOUD_STORAGE_BUCKET`	Google Cloud Storage or AWS S3 bucket name where Dekart Query results will be stored. Example: `dekart-bucket` If value is empty, users will be able to define storage bucket via UI. Supported datasource `DEKART_DATASOURCE`: `BQ` BigQuery from version >= 0.15
`DEKART_CORS_ORIGIN=` version >= 0.10	CORS Origin to be allowed by Dekart backend and set in `Access-Control-Allow-Origin` header. If not set or set incorrectly, warning will appear in logs. If set incorrectly. Example: `https://dekart.example.com`

AWS

Dekart support started AWS SDK environment variables. Required to query AWS Athena and use AWS S3.

Name	Description
`AWS_REGION` version >= 0.8	The AWS SDK compatible environment variable that specifies the AWS Region to send the request to
`AWS_ACCESS_KEY_ID` version >= 0.8	Specifies an AWS access key associated with an IAM user or role.
`AWS_SECRET_ACCESS_KEY` version >= 0.8	Specifies the secret key associated with the access key. This is essentially the “password” for the access key.

AWS Athena

Name	Description
`DEKART_ATHENA_CATALOG` version >= 0.8	Data source (group of databases) for AWS Athena to reference when executing queries. Default value is usually `AwsDataCatalog`. Example: `my-athena-catalog`
`DEKART_ATHENA_S3_OUTPUT_LOCATION` and then copied to `DEKART_CLOUD_STORAGE_BUCKET`. Example: `athena-results`
`DEKART_ATHENA_S3_OUTPUT_LOCATION` version >= 0.8	Amazon S3 query result location required by Athena SDK. This is different from `DEKART_CLOUD_STORAGE_BUCKET`. First query results are stored in `DEKART_ATHENA_S3_OUTPUT_LOCATION` and then copied to `DEKART_CLOUD_STORAGE_BUCKET`. Example: `athena-results`
`DEKART_ATHENA_WORKGROUP` version >= 0.13	AWS Athena workgroup to use when executing Athena queries. If not specified, the default `primary` workgroup will be used. Example: `my-athena-workgroup`

Google Cloud

Required to query BigQuery and use Cloud Storage

Name	Description
`GOOGLE_APPLICATION_CREDENTIALS`	Credentials for Google Cloud API Example: `/.../service-account-123456.json`

BigQuery

Name	Description
`DEKART_BIGQUERY_PROJECT_ID`	Unique identifier for your Google Cloud project with BigQuery API Enabled. Example: `my-project` version >= 0.15 If value is empty, users will be able to define project ID via UI.
`DEKART_BIGQUERY_MAX_BYTES_BILLED` version >= 0.7	Sets `maximumBytesBilled` in BigQuery Job Configuration to implement Best Practices for Controlling Query Cost. If not set warning message will appear in logs.
`DEKART_GCP_EXTRA_OAUTH_SCOPES` version >= 0.14	Set additional scopes for the GCP OAuth token when connecting to BigQuery. The value is interpreted as a comma-delimited list. E.g., in order to query a BigQuery table backed by a Google Sheet in Google Drive, the value needs to be set to `https://www.googleapis.com/auth/drive`.

Snowflake

Name	Description
`DEKART_SNOWFLAKE_ACCOUNT_ID` version >= 0.12	Snowflake Account Identifier Example: `orgname-account_name`
`DEKART_SNOWFLAKE_USER` version >= 0.12	Snowflake user with default warehouse configured Example: `example_user`
`DEKART_SNOWFLAKE_PASSWORD` version >= 0.12	Snowflake user password Example: `******`

Postgres (as a data source)

Postgres can be used as a data source for Dekart. Do not confuse with Dekart’s Postgres database, which is used to store query meta information.

Name	Description
`DEKART_POSTGRES_DATA_CONNECTION` version >= 0.16	Postgres DB to be used as data source Example: `postgres://user:password@host:port/db`

File upload

Starting from version 0.10 Dekart supports file upload. File upload is disabled by default. Once uploaded files are stored in the same storage as query results. Both AWS S3 and Google Cloud Storage are supported. The recommended max file size is 100MB.

Name	Description
`DEKART_ALLOW_FILE_UPLOAD` version >= 0.10	Enable file upload Example value: `1`

User authorization via Google OAuth 2.0 flow

Dekart can authorize users via Google OAuth 2.0 and use users’ credentials to access BigQuery and Cloud Storage. When this option is enabled, Dekart does not require a service account and GOOGLE_APPLICATION_CREDENTIALS to be set. The user token is retrieved from Google OAuth 2.0 flow and stored in only in the browser memory. When the page is refreshed, the token is retrieved again. User short-lived token is then passed via Authorization header Dekart backend to access BigQuery and Cloud Storage.

No token is stored in the Dekart backend, database, or logs.

Each user needs to have access to BigQuery and Cloud Storage with following permissions:

BigQuery Data Viewer
BigQuery Job User
BigQuery Read Session User
Storage Object User

This option is only supported for BigQuery and Cloud Storage. It is not supported for AWS and Snowflake Data Sources.

Name	Description
`DEKART_REQUIRE_GOOGLE_OAUTH` version >= 0.15	Enables Google OAuth 2.0 flow. Requires users to be authenticated. Example value: `1`
`DEKART_GOOGLE_OAUTH_CLIENT_ID` version >= 0.15	Google OAuth 2.0 Client ID. Example value: `1234567890-abcde.apps.googleusercontent.com`
`DEKART_GOOGLE_OAUTH_SECRET` version >= 0.15	Google OAuth 2.0 Client Secret. Example value: `******`

Creating Google OAuth 2.0 Client ID and Client Secret:

Configure OAuth Consent Screen in your Google Cloud Project
Create OAuth 2.0 Client ID with Web application type
Add https://your-dekart-url.com/api/v1/authenticate to Authorized redirect URIs

User authorization via Google IAP

Dekart can read claims provided by Google IAP and authorize users to:

list and edit only their own reports
read-only access to other users reports

Name	Description
`DEKART_REQUIRE_IAP`	Enables validation Google IAP JWT. Required users to be authenticated. ENables user management policies. Example value: `1`
`DEKART_IAP_JWT_AUD`	Signed Header JWT Audience (`aud`). You can get the values for the aud string mentioned above by accessing the Cloud Console, or you can use the gcloud command-line tool. See details. Example value: `/projects/PROJECT_NUMBER/apps/PROJECT_ID`

User authorization via Amazon Load Balancer

Dekart can read claims provided by Amazon Load Balancer and authorize users to:

list and edit only their reports
read-only access to other user’s reports

Amazon Load Balancer configuration example with Terraform

Name	Description
`DEKART_REQUIRE_AMAZON_OIDC`	Enables users authorization. Requires users to be authenticated and `x-amzn-oidc-data` to be passed from Load Balancer. Requires `AWS_REGION`. Example value: `1`

User Experience

Name	Description
`DEKART_UX_HOMEPAGE`	Change URL linked from Dekart logo Default value: `/`
`DEKART_UX_DATA_DOCUMENTATION`	Allows provide URL to dataset documentation. It will appear in Dekart UI. Example value: `https://my.company/dataset/schema.html`
`DEKART_HTML_CUSTOM_CODE`	Allows to add custom HTML code to `<head>`. Can be used for adding trackers.
`DEKART_DISABLE_USAGE_STATS` version >= 0.11	By default, Dekart appends certain information to the referrer of external links. This information includes the version number, the SHA256 hash of the hostname, the name of the data source, and the total number of reports, queries, files, and authors. No other information is collected. The source code for this implementation can be found here. This behavior can be turned off by setting this variable to `1`.
`DEKART_UX_ACCESS_ERROR_INFO_HTML` version >= 0.16	Allows to provide custom HTML code to be shown on the access error page.
`DEKART_UX_NOT_FOUND_ERROR_INFO_HTML` version >= 0.16	Allows to provide custom HTML code to be shown on the not found error page.
`DEKART_UX_SAMPLE_QUERY_SQL` version >= 0.16	Allows to provide a sample SQL query to be shown in the query editor.

Development specific

Do not change for production

Name	Description
`DEKART_LOG_DEBUG`	Set Dekart log level to debug Example value: `1`
`DEKART_LOG_PRETTY`	Print pretty colorful logs in console. By default Dekart formats logs as JSON Example value: `1`
`DEKART_STATIC_FILES`	Example value: `./build`

Edit this page on GitHub

Environment Variables

Main configuration#

AWS#

AWS Athena#

Google Cloud#

BigQuery#

Snowflake#

Postgres (as a data source)#

File upload#

User authorization via Google OAuth 2.0 flow#

User authorization via Google IAP#

User authorization via Amazon Load Balancer#

User Experience#

Development specific#