Confirmation Check

A confirmation check is an automatic re-test that happens after a monitor first sees a failure. Only if the retry also fails does the system open an incident and notify anyone. Some services call this a recheck, retry count, or multi-location confirmation.

Why it matters

The internet is noisy. A single dropped packet, a momentary DNS hiccup, or a restart of a few seconds can fail one check even when the site is fine. Without confirmation, those blips cause false positives और alert fatigue, and people begin ignoring real alerts.

Example

Sequence for a 5-minute monitor with confirmation:

1. 10:00:00 check from probe A fails (timeout).
2. 10:00:20 recheck from probe B fails (timeout).
3. Incident opens, alert is sent at 10:00:20.

If the recheck succeeds, the first failure is logged but nobody is paged. The trade-off is detection time: waiting for two or three failed checks at a 5-minute interval can delay an alert by 10 minutes or more. Quick rechecks seconds apart add safety with much less delay than waiting for the next interval.

Common mistakes

• Requiring too many consecutive failures on a long interval, which makes real outages slow to detect.
• Rechecking from the same location, which repeats the same local fault.
• Using the same timeout for the check and the recheck.

Related

• False positive
• Flapping
• Probe location
• MTTD

यह पेज सामान्य जानकारी है, जिसे हमारी संपादकीय टीम ने लिखा है और दस्तावेज़ीकरण बदलने पर अपडेट किया जाता है। सर्वर और प्रदाता अलग-अलग होते हैं, इसलिए बदलाव पहले सुरक्षित तरीके से जाँचें। देखें: संपादकीय नीति और अस्वीकरण.