Error Handling#

Comprehensive guide to error codes, debugging, and troubleshooting under the APIM-first security model.

Overview#

Ocp-Apim-Subscription-Key

Base API URL for clients: https://api.pay.ledger1.ai/portalpay

• Health check path:
  markup

  GET /portalpay/healthz

  (no subscription required)
• All API routes:
  markup

  /portalpay/api/*

  (APIM rewrites to backend
  markup

  /api/*

  )

cb_auth_token

HTTP Status Codes#

Error Response Format#

All errors follow this structure:

json
{
  "error": "error_code",
  "message": "Human-readable error description"
}

resetAt

Common Error Codes#

Authentication & Authorization Errors#

markup

unauthorized

#

• HTTP: 401
• Message: "Missing or invalid subscription key" (developer APIs)
• Cause: Missing/invalid
  markup

  Ocp-Apim-Subscription-Key

  header
• Solution: Include a valid APIM subscription key on every developer API request

cURL:

bash
# Wrong (no key)
curl -X GET "https://api.pay.ledger1.ai/portalpay/api/inventory"

# Correct (with APIM key)
curl -X GET "https://api.pay.ledger1.ai/portalpay/api/inventory" \
  -H "Ocp-Apim-Subscription-Key: $APIM_SUBSCRIPTION_KEY"

For admin-only operations (PortalPay UI):

• HTTP: 401
• Message: "JWT authentication failed"
• Cause: Missing/expired admin session cookie
  markup

  cb_auth_token

• Solution: Re-authenticate through the web interface

markup

forbidden

#

• HTTP: 403
• Message: "Insufficient scope or not allowed"
• Causes:
  • APIM subscription does not include the required scope (e.g.,
    markup

    orders:create

    ,
    markup

    inventory:write

    )
  • Attempt to access or mutate resources without proper role/ownership (admin-JWT paths)
• Solutions:
  • Ensure your APIM product/subscription grants the required scopes
  • For admin routes, confirm you are logged in with appropriate roles

x-edge-secret

Business Logic Errors#

markup

split_required

#

• HTTP: 403
• Message: "Split contract not configured for this merchant"
• Cause: Creating orders without configuring split first
• Solution: Configure your split in the PortalPay Admin UI (Settings → Payments → Split), then retry

markup

inventory_item_not_found

#

• HTTP: 400
• Message: "Product not found in inventory"
• Cause: Referencing a SKU or ID that doesn't exist
• Solution: Verify SKU/ID exists in your inventory; list inventory and confirm before ordering

typescript
// Check inventory first
const items = await listProducts();
const exists = items.some(item => item.sku === 'ITEM-001');
if (!exists) throw new Error('inventory_item_not_found');

// Then create order
await createOrder([{ sku: 'ITEM-001', qty: 1 }]);

markup

items_required

#

• HTTP: 400
• Message: "At least one item is required"
• Cause: Creating an order with an empty items array
• Solution: Include at least one item in the order

Validation Errors#

markup

invalid_input

#

• HTTP: 400
• Message: "Invalid request parameters"
• Cause: Missing required fields or invalid data types
• Solution: Review endpoint docs for required parameters; validate types and value ranges

Common causes:

• Missing required fields (sku, name, price, etc.)
• Invalid data types (string instead of number)
• Out of range values (negative prices, invalid stock quantity)

typescript
// Wrong
{
  "sku": "ITEM-001",
  // Missing name
  "priceUsd": "invalid",  // Should be number
  "stockQty": -10         // Should be >= -1
}

// Correct
{
  "sku": "ITEM-001",
  "name": "Product Name",
  "priceUsd": 25.00,
  "stockQty": 100
}

Rate Limiting Errors#

markup

rate_limited

#

• HTTP: 429
• Message: "Rate limit exceeded"
• Response: Includes
  markup

  resetAt

  timestamp (Unix ms)
• Cause: Too many requests in the time window
• Solution: Implement backoff and retry after
  markup

  resetAt

Example payload:

json
{
  "error": "rate_limited",
  "message": "Rate limit exceeded",
  "resetAt": 1698765432000
}

Implementation:

typescript
async function makeRequestWithRetry(fn: () => Promise<any>, maxRetries = 3) {
  for (let i = 0; i < maxRetries; i++) {
    try {
      return await fn();
    } catch (err: any) {
      if (err?.error === 'rate_limited') {
        const resetAt = err.resetAt || Date.now() + 60_000;
        const waitMs = Math.max(0, resetAt - Date.now());
        if (i < maxRetries - 1) {
          await new Promise(r => setTimeout(r, waitMs));
          continue;
        }
      }
      throw err;
    }
  }
}

Rate limit headers (if enabled at gateway):

• markup

  X-RateLimit-Limit

• markup

  X-RateLimit-Remaining

• markup

  X-RateLimit-Reset

System / Degraded Mode#

markup

cosmos_unavailable

#

• HTTP: 200 (Degraded mode)
• Response: Includes
  markup

  degraded: true

• Cause: Database temporarily unavailable
• Solution: System operates in degraded mode; data will be persisted when database recovers

json
{
  "ok": true,
  "degraded": true,
  "reason": "cosmos_unavailable",
  "data": { "...": "..." }
}

Handling:

typescript
const response = await createOrder(items);
if (response.degraded) {
  console.warn('System in degraded mode:', response.reason);
  // Inform user or queue for reconciliation
}

markup

platform_recipient_not_configured

#

• HTTP: 400
• Message: "Platform recipient address not set up"
• Cause: Server/platform configuration issue
• Solution: Contact support

Debugging Tips#

1. Use Correlation IDs#

X-Correlation-Id

cURL:

bash
curl -i "https://api.pay.ledger1.ai/portalpay/api/orders" \
  -H "Content-Type: application/json" \
  -H "Ocp-Apim-Subscription-Key: $APIM_SUBSCRIPTION_KEY" \
  -X POST -d '{...}'

# Response headers may include:
# X-Correlation-Id: 550e8400-e29b-41d4-a716-446655440000

2. Check Request Format#

typescript
console.log('Request:', { url, method: 'POST', headers, body: payload });

const response = await fetch(url, {
  method: 'POST',
  headers,
  body: JSON.stringify(payload)
});

const data = await response.json();
if (!response.ok) {
  console.error('Error:', { status: response.status, error: data });
}

3. Validate Before Sending#

typescript
function validateOrder(items: any[]) {
  if (!Array.isArray(items) || items.length === 0) throw new Error('items_required');
  for (const item of items) {
    if (!item.sku && !item.id) throw new Error('Item must have SKU or ID');
    if (!item.qty || item.qty < 1) throw new Error('Invalid quantity');
  }
}

4. Monitor Your Usage#

Track call/error/rate-limit counts and alert on spikes:

typescript
const metrics = { calls: 0, errors: 0, rateLimits: 0 };
async function tracked(fn: () => Promise<any>) {
  metrics.calls++;
  try { return await fn(); }
  catch (e: any) { metrics.errors++; if (e?.error === 'rate_limited') metrics.rateLimits++; throw e; }
}
setInterval(() => console.log('API Metrics:', metrics), 60_000);

Common Scenarios#

Scenario 1: Authentication Failure (Developer APIs)#

• Error:
  markup

  unauthorized

  (401)
• Steps:
  1. Ensure
     markup

     Ocp-Apim-Subscription-Key

     is present and valid
  2. Confirm your subscription is active
  3. Retry the request

Scenario 2: Missing Scope#

• Error:
  markup

  forbidden

  (403)
• Steps:
  1. Check required scope in the API docs (e.g.,
     markup

     orders:create

     )
  2. Verify your APIM product/subscription includes that scope
  3. Request access or upgrade if necessary

Scenario 3: Split Not Configured#

• Error:
  markup

  split_required

  (403)
• Steps:
  1. In Admin UI, configure split (Settings → Payments → Split)
  2. Retry order creation

Scenario 4: Rate Limited#

• Error:
  markup

  rate_limited

  (429)
• Steps:
  1. Read
     markup

     X-RateLimit-*

     headers and
     markup

     resetAt

  2. Back off until reset time
  3. Implement client-side throttling

Getting Help#

If you encounter persistent errors:

1. Documentation: Review the API Reference
2. Include Details in Reports:
   • X-Correlation-Id (if present)
   • Endpoint, method, headers (redact secrets)
   • Full error payload and status code
   • Steps to reproduce
3. Contact Support with the above information

Next Steps:

• API Reference
• Rate Limits
• Quick Start