Documentation Index Fetch the complete documentation index at: https://docs.prefect.io/llms.txt
Use this file to discover all available pages before exploring further.
prefect.server.database.interfaceClasses DBSingleton Ensures that only one database interface is created per unique key
PrefectDBInterface An interface for backend-specific SqlAlchemy actions and ORM models.
The REST API can be configured to run against different databases in order maintain
performance at different scales. This interface integrates database- and dialect-
specific configuration into a unified interface that the orchestration engine runs
against.
Methods: Agent Agent( self ) -> type[orm_models.Agent]
An agent model
Artifact Artifact( self ) -> type[orm_models.Artifact]
An artifact orm model
ArtifactCollection ArtifactCollection( self ) -> type[orm_models.ArtifactCollection]
An artifact collection orm model
Automation Automation( self ) -> type[orm_models.Automation]
An automation model
AutomationBucket AutomationBucket( self ) -> type[orm_models.AutomationBucket]
An automation bucket model
AutomationEventFollower AutomationEventFollower( self ) -> type[orm_models.AutomationEventFollower]
A model capturing one event following another event
AutomationRelatedResource( self ) -> type[orm_models.AutomationRelatedResource]
An automation related resource model
Base Base( self ) -> type[orm_models.Base]
Base class for orm models
BlockDocument BlockDocument( self ) -> type[orm_models.BlockDocument]
A block document model
BlockDocumentReference BlockDocumentReference( self ) -> type[orm_models.BlockDocumentReference]
A block document reference model
BlockSchema BlockSchema( self ) -> type[orm_models.BlockSchema]
A block schema model
BlockSchemaReference BlockSchemaReference( self ) -> type[orm_models.BlockSchemaReference]
A block schema reference model
BlockType BlockType( self ) -> type[orm_models.BlockType]
A block type model
CompositeTriggerChildFiring CompositeTriggerChildFiring( self ) -> type[orm_models.CompositeTriggerChildFiring]
A model capturing a composite trigger’s child firing
ConcurrencyLimit ConcurrencyLimit( self ) -> type[orm_models.ConcurrencyLimit]
A concurrency model
ConcurrencyLimitV2 ConcurrencyLimitV2( self ) -> type[orm_models.ConcurrencyLimitV2]
A v2 concurrency model
Configuration Configuration( self ) -> type[orm_models.Configuration]
An configuration model
CsrfToken CsrfToken( self ) -> type[orm_models.CsrfToken]
A csrf token model
Deployment Deployment( self ) -> type[orm_models.Deployment]
A deployment orm model
DeploymentSchedule DeploymentSchedule( self ) -> type[orm_models.DeploymentSchedule]
A deployment schedule orm model
Event Event( self ) -> type[orm_models.Event]
An event model
EventResource EventResource( self ) -> type[orm_models.EventResource]
An event resource model
Flow Flow( self ) -> type[orm_models.Flow]
A flow orm model
FlowRun FlowRun( self ) -> type[orm_models.FlowRun]
A flow run orm model
FlowRunInput( self ) -> type[orm_models.FlowRunInput]
A flow run input model
FlowRunState FlowRunState( self ) -> type[orm_models.FlowRunState]
A flow run state orm model
Log Log( self ) -> type[orm_models.Log]
A log orm model
SavedSearch SavedSearch( self ) -> type[orm_models.SavedSearch]
A saved search orm model
TaskRun TaskRun( self ) -> type[orm_models.TaskRun]
A task run orm model
TaskRunState TaskRunState( self ) -> type[orm_models.TaskRunState]
A task run state orm model
TaskRunStateCache TaskRunStateCache( self ) -> type[orm_models.TaskRunStateCache]
A task run state cache orm model
Variable Variable( self ) -> type[orm_models.Variable]
A variable model
WorkPool WorkPool( self ) -> type[orm_models.WorkPool]
A work pool orm model
WorkQueue WorkQueue( self ) -> type[orm_models.WorkQueue]
A work queue model
Worker Worker( self ) -> type[orm_models.Worker]
A worker process orm model
create_db Create the database
dialect dialect( self ) -> type[sa.engine.Dialect]
drop_db Drop the database by removing all tables directly.
This reflects the actual database schema and drops every table rather
than running all Alembic downgrade migrations in reverse. Running
downgrades is fragile because individual migration downgrade steps may
fail on real-world data (e.g. re-adding a foreign key constraint when
orphaned references exist). Dropping tables directly is both faster
and more robust.
Reflection is used instead of Base.metadata.drop_all() so that
tables created by migrations but not tracked in the ORM (e.g.
deployment_version, alembic_version) are also removed.
engine engine( self ) -> AsyncEngine
Provides a SqlAlchemy engine against a specific database.
is_db_connectable is_db_connectable( self ) -> bool
Returns boolean indicating if the database is connectable.
This method is used to determine if the server is ready to accept requests.
run_migrations_downgrade run_migrations_downgrade( self , revision: str = '-1' ) -> None
Run all downgrade migrations
run_migrations_upgrade run_migrations_upgrade( self ) -> None
Run all upgrade migrations
session session( self ) -> AsyncSession
Provides a SQLAlchemy session.
session_context session_context( self , begin_transaction: bool = False , with_for_update: bool = False )
Provides a SQLAlchemy session and a context manager for opening/closing
the underlying connection.
Args:
begin_transaction: if True, the context manager will begin a SQL transaction.
Exiting the context manager will COMMIT or ROLLBACK any changes. 
