Flapping

Flapping describes a service, link or monitor that switches quickly between healthy and failing. A server may respond to some checks and drop others, so the monitor reports DOWN, UP, DOWN, UP within minutes.

Why it matters

Each state change can trigger a notification, so flapping floods channels and causes alert fatigue. It also makes uptime figures hard to read and can hide the real cause. Common causes include a load balancer sending traffic to one unhealthy node, an overloaded server crossing a timeout threshold, an unstable network link, or a process restarting in a loop.

Example

A check at one-minute intervals produces:

09:00 UP
09:01 DOWN
09:02 UP
09:03 DOWN

Monitoring tools handle this with flap detection or hysteresis, for example requiring three consecutive failures to go DOWN and three consecutive successes to go UP again.

Common mistakes

• Muting the alerts without investigating. The underlying instability still affects some users.
• Setting thresholds so tight that normal variation crosses them.
• Treating it as a monitoring issue when one backend behind a load balancer is failing.
• Ignoring it because the average uptime still looks high.

Related

• False positive
• Alert fatigue
• Load balancer
• Threshold alert

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.