Build Logic Apps connector configs for HTTP, SQL, Blob Storage, Office 365, and Teams with triggers, actions, and retry policies.
Build Logic Apps connector configs for HTTP, SQL, Blob Storage, Office 365, and Teams with triggers, actions, and retry policies.
Required Fields
workflowNametrigger.typeconnectorsconnectors[0].nameconnectors[0].typeOutput will appear here...The builder validates that workflowName, trigger.type, connectors, and the first connector's name/type all resolve before accepting the JSON as a valid combined workflow definition, the fields Logic Apps needs to know the trigger and the chain of connector actions; it can't verify referenced connection names (like sql-contoso-prod) are actually configured and authorized, or that Workflow Definition Language expressions reference real prior-action outputs, those are only surfaced when the workflow actually executes.
Build an Azure Logic App Standard workflow (kind: Stateful, meaning execution history and state persist and are queryable after each run, versus Stateless which trades that durability for lower latency) chaining ApiConnection-based connectors (SQL, Blob Storage, Office 365) with raw HTTP actions, each action able to reference prior actions' outputs via Workflow Definition Language expressions like @{body('sql-lookup-customer')}. The HTTP action's retryPolicy with type exponential and bounded minimumInterval/maximumInterval only applies to that specific HTTP action, ApiConnection-based connectors have their own separate, connector-specific retry behavior that isn't controlled by this same retryPolicy block, so a workflow mixing both connector types needs to understand each action type's actual retry semantics rather than assuming one retryPolicy setting governs the whole workflow uniformly.
Don't assume retryPolicy configured on one action type (HTTP) automatically extends to other connector types in the same workflow, verify and configure retry behavior per action type based on its own actual retry semantics.
Choose Stateful specifically when audit history or resumability genuinely matters for the workflow's use case, defaulting to Stateful everywhere adds persistence cost and overhead that a high-throughput, fire-and-forget workflow doesn't need.
Explicit error-handling actions (like the Teams notification on failure) need to be wired to every action whose failure should actually trigger notification, a workflow with error handling on only some steps has blind spots where a failure in an unwired step fails silently.
No, retryPolicy as configured in the example is scoped to that specific HTTP action, ApiConnection-based connectors (SQL, Blob Storage, Office 365) have their own connector-specific retry behavior, which may have different defaults and isn't controlled by copying the same retryPolicy block onto them. Check each connector type's own documented retry behavior rather than assuming one workflow-wide retry setting governs every action uniformly.
Stateful workflows persist their full execution history (every action's inputs, outputs, and status) for later querying, useful for auditing, debugging a specific run, or resuming a long-running workflow. Stateless workflows skip persisting that history for lower latency and cost, appropriate for high-throughput, short-lived workflows where per-run audit history isn't needed, and where losing in-flight state on a restart is an acceptable tradeoff for the performance gain.
It needs explicit configuration, the runAfterFailure pattern is defined as an action whose execution condition is tied to a prior action's failure state, it's not an automatic, workflow-wide catch-all unless every action in the chain has been deliberately wired to run conditionally on failure. A workflow missing this explicit wiring on a critical action simply fails without any notification path at all.
Was this tool helpful?
Disclaimer: This tool runs entirely in your browser. No data is sent to our servers. Always verify outputs before using them in production. AWS, Azure, and GCP are trademarks of their respective owners.