Check Retries and Alert Thresholds

Why False Positives Happen

A single failed run does not always mean your service is down. A transient network blip, a slow handshake, or a one-off timeout can make a healthy endpoint look broken for a moment. LoadFocus has two ways to keep that from paging you, depending on the type of check: retries, and alert after N consecutive failures.

Both live in the Retries before alerting section of the check form on the New API Check page.

Retries: API, Multistep, TCP and DNS Checks

When one of these checks fails, LoadFocus runs it again before reporting the result. If a retry passes, the run is recorded as passed and nothing alerts. You can set:

• Retries: how many extra attempts, from 0 to 5. New checks start at 2; 0 turns retries off.
• Interval (s): the base wait between attempts, from 1 to 60 seconds.
• Backoff: Fixed, Linear or Exponential spacing between attempts.
• Max retry time (s): a cap on how long all retries together may take.

A retry always runs from the same location as the failed attempt. How the waits grow and how the time cap works is explained in Retries Before Alerting.

Alert After N Consecutive Failures: Browser Checks

Browser checks do not retry. In the same section the field becomes Alert after N consecutive failures, and your check frequency is the spacing between runs. LoadFocus alerts when N failed runs happen in a row, from 0 to 5. A value of 0 or 1 alerts on the first failed run.

A recovery notification is only sent when an alert was sent, so a short blip that never reached N does not produce a "recovered" message.

Alert on Every Failure

Under Advanced notification options, the toggle every time a check fails sends an alert for each failing run instead of only when the check starts failing. For a browser check with N above 1, it starts alerting once N failed runs in a row are reached and then alerts on every further failed run. keep reminding me while a check stays down sends repeat reminders on a schedule you choose.

For the shared check settings, see How to Create a New API Check. To choose where alerts are delivered, see Alert Channels, and learn more about Uptime Monitoring.