Snowflake Database
Deploys a Snowflake database with configurable Time Travel retention, transient mode, Iceberg table defaults, and task execution parameters. All database-level settings are managed declaratively through the spec, including collation, logging, and user task configuration.
What Gets Created
When you deploy a SnowflakeDatabase resource, OpenMCF provisions:
- Snowflake Database — a
snowflake_databaseresource with the specified name and all configured parameters including Time Travel retention, collation, and Iceberg settings - Snowflake Provider — configured using explicit credentials from provider config or environment variables
Prerequisites
- Snowflake credentials configured via environment variables (
SNOWFLAKE_ACCOUNT,SNOWFLAKE_USER,SNOWFLAKE_PASSWORD) or OpenMCF provider config - A Snowflake account with permissions to create databases
- An external volume if configuring Iceberg table defaults via
externalVolume
Quick Start
Create a file snowflake-database.yaml:
apiVersion: snowflake.openmcf.org/v1
kind: SnowflakeDatabase
metadata:
name: my-database
labels:
openmcf.org/provisioner: pulumi
pulumi.openmcf.org/organization: my-org
pulumi.openmcf.org/project: my-project
pulumi.openmcf.org/stack.name: dev.SnowflakeDatabase.my-database
spec:
name: MY_DATABASE
Deploy:
openmcf apply -f snowflake-database.yaml
This creates a standard (non-transient) Snowflake database named MY_DATABASE with default settings.
Configuration Reference
Required Fields
| Field | Type | Description | Validation |
|---|---|---|---|
name | string | Snowflake identifier for the database. Must be unique within the account. Avoid characters: |, ., (, ), ". | — |
Optional Fields
| Field | Type | Default | Description |
|---|---|---|---|
catalog | string | "" | Default catalog for Iceberg tables. Sets the CATALOG database parameter. |
comment | string | "" | Comment attached to the database. |
dataRetentionTimeInDays | int | 0 | Number of days for Time Travel actions (CLONE and UNDROP). Also sets the default retention for all schemas in the database. |
defaultDdlCollation | string | "" | Default collation specification for all schemas and tables. Can be overridden at schema or table level. |
dropPublicSchemaOnCreation | bool | false | Drops the PUBLIC schema on database creation. Has no effect if changed after creation. |
enableConsoleOutput | bool | false | Enables stdout/stderr fast path logging for anonymous stored procedures. |
externalVolume | string | "" | Default external volume for Iceberg tables. Sets the EXTERNAL_VOLUME database parameter. |
isTransient | bool | false | Creates a transient database. Transient databases have no Fail-safe period, reducing storage costs but removing Fail-safe data recovery. |
logLevel | string | "" | Severity level for event table ingestion. Valid values: TRACE, DEBUG, INFO, WARN, ERROR, FATAL, OFF. |
maxDataExtensionTimeInDays | int | 0 | Maximum days Snowflake can extend data retention to prevent streams from becoming stale. |
quotedIdentifiersIgnoreCase | bool | false | When true, the case of quoted identifiers is ignored. |
replaceInvalidCharacters | bool | false | Replaces invalid UTF-8 characters with the Unicode replacement character in Iceberg table query results. Only applies to tables using an external Iceberg catalog. |
storageSerializationPolicy | string | "" | Storage serialization policy for Iceberg tables using Snowflake as catalog. Valid values: COMPATIBLE (interoperable with third-party engines), OPTIMIZED (best performance within Snowflake). |
suspendTaskAfterNumFailures | int | 0 | Consecutive task failures before automatic suspension. 0 disables auto-suspending. |
taskAutoRetryAttempts | int | 0 | Maximum automatic retry attempts for user tasks. |
traceLevel | string | "" | Controls trace event ingestion into the event table. Valid values: ALWAYS, ON_EVENT, OFF. |
userTask.managedInitialWarehouseSize | string | "" | Initial warehouse size for managed warehouses when no execution history exists. |
userTask.minimumTriggerIntervalInSeconds | int | 0 | Minimum seconds between triggered task executions. |
userTask.timeoutMs | int | 0 | User task execution timeout in milliseconds. |
Examples
Transient Database with Short Retention
A cost-optimized transient database with 1-day Time Travel retention, suitable for staging or ephemeral workloads:
apiVersion: snowflake.openmcf.org/v1
kind: SnowflakeDatabase
metadata:
name: staging-db
labels:
openmcf.org/provisioner: pulumi
pulumi.openmcf.org/organization: my-org
pulumi.openmcf.org/project: my-project
pulumi.openmcf.org/stack.name: staging.SnowflakeDatabase.staging-db
spec:
name: STAGING_DB
isTransient: true
dataRetentionTimeInDays: 1
comment: "Staging environment - transient, short retention"
Production Database with Extended Retention
A production database with 14-day Time Travel retention, extended data retention for stream staleness prevention, and debug-level logging:
apiVersion: snowflake.openmcf.org/v1
kind: SnowflakeDatabase
metadata:
name: prod-analytics
labels:
openmcf.org/provisioner: pulumi
pulumi.openmcf.org/organization: my-org
pulumi.openmcf.org/project: my-project
pulumi.openmcf.org/stack.name: prod.SnowflakeDatabase.prod-analytics
spec:
name: PROD_ANALYTICS
dataRetentionTimeInDays: 14
maxDataExtensionTimeInDays: 28
logLevel: DEBUG
traceLevel: ON_EVENT
defaultDdlCollation: "en-ci"
comment: "Production analytics database"
Full-Featured Database with Iceberg and Task Configuration
A database configured for Iceberg table workloads with task execution settings, suitable for data lake integration:
apiVersion: snowflake.openmcf.org/v1
kind: SnowflakeDatabase
metadata:
name: data-lake-db
labels:
openmcf.org/provisioner: pulumi
pulumi.openmcf.org/organization: my-org
pulumi.openmcf.org/project: my-project
pulumi.openmcf.org/stack.name: prod.SnowflakeDatabase.data-lake-db
spec:
name: DATA_LAKE_DB
dataRetentionTimeInDays: 7
maxDataExtensionTimeInDays: 14
catalog: "my_iceberg_catalog"
externalVolume: "my_external_volume"
storageSerializationPolicy: COMPATIBLE
replaceInvalidCharacters: true
logLevel: INFO
traceLevel: ON_EVENT
suspendTaskAfterNumFailures: 5
taskAutoRetryAttempts: 3
userTask:
managedInitialWarehouseSize: "XSMALL"
minimumTriggerIntervalInSeconds: 60
timeoutMs: 3600000
comment: "Data lake integration database with Iceberg support"
Stack Outputs
After deployment, the following outputs are available in status.outputs:
| Output | Type | Description |
|---|---|---|
id | string | Provider-assigned unique ID for the database resource |
name | string | Fully-qualified name of the created database |
owner | string | Owner role of the database |
created_on | string | Timestamp when the database was created |
is_transient | string | Whether the database is transient ("true" or "false") |
data_retention_time_in_days | string | Configured data retention time in days |
Related Components
No other OpenMCF components have direct foreign key references to SnowflakeDatabase. This component is typically deployed as a standalone resource within a Snowflake account.
Next article
Examples
Examples This section provides ready-to-use manifest examples for deploying infrastructure with OpenMCF. Each example is a complete, valid manifest that you can copy, customize, and deploy. How to Use These Examples Every example follows the same workflow: Copy the manifest YAML into a file (e.g., my-resource.yaml) Customize the fields for your environment — resource names, regions, credentials, and configuration values Set provisioner labels — choose pulumi or tofu as your provisioner and...