Retries and Backoff

When a callback fails, AsyncQueue retries the request according to your configuration. You control the maximum number of attempts and the backoff strategy.

Retry configuration

Set retry behavior when creating a task:

curl -X POST https://api.asyncqueue.io/v1/tasks \
  -H "Authorization: Bearer your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "targetUrl": "https://api.example.com/webhook",
    "maxRetries": 5,
    "retryBackoff": "exponential"
  }'

Field	Default	Description
maxRetries	3	Maximum retry attempts (0-10). Set to 0 for no retries.
retryBackoff	"exponential"	Backoff strategy: exponential or linear

Backoff strategies

Exponential backoff (default) - delays double with each retry:

Retry	Delay
1	1 second
2	2 seconds
3	4 seconds
4	8 seconds
5	16 seconds

Linear backoff - delays increase by a fixed amount:

Retry	Delay
1	1 second
2	2 seconds
3	3 seconds
4	4 seconds
5	5 seconds

What triggers a retry

• Network errors (connection refused, DNS failure, socket timeout)
• HTTP 5xx responses (server errors)
• Request timeout (exceeding the timeout value)

What does NOT trigger a retry

• HTTP 2xx responses (treated as success)
• HTTP 3xx responses (redirects are followed automatically)
• HTTP 4xx responses (client errors are treated as completed, not retried)

After all retries fail

When all retry attempts are exhausted, the task moves to failed status with an error message. If you configured an onCompleteUrl, AsyncQueue sends a webhook notification with the failure details.

The task’s retryCount field shows how many retry attempts were made.