Pipeline Configuration

Clauses and defaults that define pipelines and their schedules, from lifecycle to cron and approval workflow.

Pipeline Clauses

• APPROVAL REQUIREDRequire manual approval before each scheduled or triggered run executes.
• DEFAULTSDefault parameter values injected into the pipeline SQL. Parameters use $name syntax.
• DESCRIPTIONHuman-readable description of the pipeline's purpose.
• FAIL_FASTStop pipeline execution on the first statement failure. If false, continues executing remaining statements.
• LIFECYCLEDevelopment lifecycle stage for governance and promotion workflows.
• PIPELINE (name)Pipeline name used as the identifier in the catalog.
• SCHEDULECron expression for automatic execution, or a named schedule reference.
• SLAService Level Agreement target in hours. Triggers notification if exceeded.
• STATUSOperational status of the pipeline.
• TAGSComma-separated tags for organizing and filtering pipelines.
• TIMEZONETimezone for cron schedule evaluation.

Pipeline Lifecycle

• ActivePipeline runs on schedule. Transitions from Draft, Paused, or Disabled via activate.
• ArchivedRead-only final state. Only reachable from Disabled. Cannot be re-activated.
• DeprecatedMarked for removal. Should not be used for new workflows.
• DevelopmentInitial development stage (alias: dev). Experimental, in-progress work.
• DisabledPermanently stopped until re-enabled. No scheduled runs. Can transition to Archived.
• DraftInitial editing state. Pipeline SQL can be modified. Not executable by the scheduler.
• PausedTemporarily halted. Scheduled runs are blocked. Resume returns to Active.
• ProductionCertified and live (alias: prod). Full SLA enforcement.
• StagingReady for production review. Pre-release validation stage.
• TestingUnder validation and QA (alias: test).

Schedule Clauses

• ACTIVE / INACTIVEWhether the schedule is enabled. Inactive schedules do not fire.
• CATCHUPBackfill missed cron firings when a schedule is re-enabled after being inactive.
• CRONStandard 5-field cron expression: minute hour day-of-month month day-of-week. Supports ranges (1-5), lists (1,3,5), steps (*/15), and wildcards (*).
• DESCRIPTIONHuman-readable description of the schedule.
• MAX_CONCURRENTMaximum number of concurrent pipeline runs. 0 means unlimited. Excess runs are queued.
• NOTIFYEmail addresses for run notifications (completion, failure, SLA breach).
• PRIORITYQueue priority when multiple pipelines are waiting. Higher values execute first.
• RETRIESMaximum number of retry attempts when a pipeline run fails.
• RETRY_DELAYDelay in seconds between retry attempts.
• SCHEDULE (name)Schedule name used as the identifier.
• TARGET_NODESCompute node targeting. ALL broadcasts to every healthy node. Specific node IDs route to named nodes. Default: any single healthy node.
• TIMEOUTMaximum execution time per pipeline run in seconds. Run is cancelled if exceeded.
• TIMEZONETimezone for evaluating the cron expression.
• WEBHOOKWebhook URLs that receive HTTP POST notifications on run events.

Schedule Defaults

• Approval WorkflowWhen enabled, scheduled runs enter a pending state and wait for manual approval. Approve via API to proceed; reject reverts pipeline to Draft status.
• Compute Node TargetingDefault: any single healthy node picks up the run. ALL: broadcast to every healthy node in parallel. Specific IDs: route to named nodes only.
• Cron Expression SyntaxStandard 5-field cron. Fields: minute (0-59), hour (0-23), day-of-month (1-31), month (1-12), day-of-week (0-7, 0 and 7 = Sunday).
• Parameter SyntaxParameters use $name syntax in pipeline SQL. Declared via DEFAULTS clause, overridable at trigger time via API.
• Statement Execution OrderPipeline SQL statements execute sequentially in file order. Each statement completes before the next begins. FAIL_FAST controls whether execution...