Catch Block — Examples

// Catch Block node (default config)
{
  "nodeType": "CatchBlock",
  "id": "catch_erp_call",
  "label": "Catch — ERP API Errors"
}

// Slack node connected to catch_erp_call's success port:
{
  "channel": "#workflow-errors",
  "message": ":red_circle: *Workflow Error — ERP Sync*\n\n*Type:* {{ $mem.__exception_type__ }}\n*Message:* {{ $mem.__exception_message__ }}\n*Order ID:* {{ $var.order_id }}\n*Workflow Run:* {{ $ctx.workflow_instance_id }}\n\nInvestigate: https://app.bizfirstai.com/runs/{{ $ctx.workflow_instance_id }}"
}

Outcome: Engineering receives a structured Slack alert within seconds of the failure, with enough context to immediately begin investigation. The workflow run ID is clickable for direct navigation to the execution log.

// Catch Block node
{
  "nodeType": "CatchBlock",
  "id": "catch_payment",
  "label": "Catch — Payment Errors"
}

// Database Insert node in catch scope:
{
  "table": "workflow_errors",
  "data": {
    "workflow_id": "{{ $ctx.workflow_definition_id }}",
    "run_id": "{{ $ctx.workflow_instance_id }}",
    "node_id": "http_payment_gateway",
    "error_type": "{{ $mem.__exception_type__ }}",
    "error_message": "{{ $mem.__exception_message__ }}",
    "stack_trace": "{{ $mem.__exception_stacktrace__ }}",
    "context_json": "{ \"order_id\": \"{{ $var.order_id }}\", \"amount\": {{ $var.amount }} }",
    "occurred_at": "{{ $now }}"
  }
}

Outcome: Every payment failure is logged with full context. The support team can query workflow_errors by workflow_id and date range to produce SLA breach reports and identify systematic failure patterns.

// Catch Block
{
  "nodeType": "CatchBlock",
  "id": "catch_api",
  "label": "Catch — API Errors"
}

// Switch node on $mem.__exception_type__:
// Case "TimeoutException"
//   → Set Var: retry_needed = true
//   → Delay: 30 seconds
//   → Retry HTTP Request (loop back)
//
// Case "UnauthorizedException"
//   → HTTP Request: Refresh OAuth Token
//   → Slack: "API credentials refreshed — retrying workflow"
//   → Restart workflow from auth step
//
// Default case
//   → Slack Alert: generic error notification
//   → Set Order Status: processing_failed

Outcome: Timeout errors trigger an automatic retry with a 30-second pause. Authentication errors trigger an automatic credential refresh. All other errors escalate to the operations team. Each failure type is handled optimally without manual intervention.

// Catch Block
{
  "nodeType": "CatchBlock",
  "id": "catch_critical",
  "label": "Catch — Critical Workflow Error"
}

// HTTP Request to Jira REST API in catch scope:
{
  "url": "https://yourcompany.atlassian.net/rest/api/3/issue",
  "method": "POST",
  "headers": { "Authorization": "Basic {{ $env.JIRA_TOKEN }}" },
  "body": {
    "fields": {
      "project": { "key": "OPS" },
      "issuetype": { "name": "Bug" },
      "priority": { "name": "High" },
      "summary": "Workflow Error: {{ $mem.__exception_type__ }} in {{ $ctx.workflow_definition_id }}",
      "description": {
        "type": "doc",
        "version": 1,
        "content": [{
          "type": "paragraph",
          "content": [{
            "type": "text",
            "text": "Error: {{ $mem.__exception_message__ }}\n\nWorkflow Run: {{ $ctx.workflow_instance_id }}\nContext: Order {{ $var.order_id }}\n\nStack Trace:\n{{ $mem.__exception_stacktrace__ }}"
          }]
        }]
      }
    }
  }
}

Outcome: A Jira bug ticket is created automatically with full error context, assigned to the workflow operations project. Engineers can filter OPS tickets by label or component to monitor workflow health. The ticket ID is stored in $var.jira_ticket_id and included in the Slack alert for cross-reference.

// Catch Block
{
  "nodeType": "CatchBlock",
  "id": "catch_api_response",
  "label": "Catch — API Response Errors"
}

// Set Variable node in catch scope:
{
  "error_code": "PROCESSING_ERROR",
  "error_user_message": "We were unable to process your request at this time. Please try again in a few minutes.",
  "error_reference": "{{ $ctx.workflow_instance_id }}"
}

// HTTP Response node (returns error to caller):
{
  "status_code": 500,
  "body": {
    "success": false,
    "error": {
      "code": "{{ $var.error_code }}",
      "message": "{{ $var.error_user_message }}",
      "reference": "{{ $var.error_reference }}"
    }
  }
}

Outcome: The API client receives a structured, sanitised error response with a reference ID (workflow instance ID) they can quote when contacting support. The internal exception details are logged separately, never exposed externally. The workflow run ID allows support staff to look up the full error context instantly.