Telegraf Ping Metric does not trigger alert in VCF Operations
Article ID: 449963
Updated On:
Products
Issue/Introduction
The Telegraf ping metric in VCF Operations fails to trigger an alert when ping connectivity is lost.
Symptoms observed:
When configuring the ping metric to monitor connectivity for a specified IP address or FQDN, the Packet Loss % metric collects as expected and reports 0 when the destination is reachable.
When the monitored virtual machine is rebooted or experiences packet loss, the Packet Loss % metric does not update or reflect the failure, preventing the configured alert from triggering.
Environment
VCF Operations 9.1.x
Cause
Data collection in VCF Operations operates on a default 5-minute cycle. If a connectivity loss event or virtual machine reboot occurs and resolves entirely within a single collection window (5-minute interval), the system does not capture the transient failure. As a result, the Packet Loss % metric remains unchanged at 0 for that interval, and the corresponding alert is not triggered. This behavior is by design.
Resolution
This behavior is by design when connectivity loss occurs within a single collection interval.
To manage or account for transient ping events in VCF Operations:
Verify the data collection interval for the Telegraf agent in VCF Operations (default is 5 minutes).
If short-duration failures must be detected, evaluate lowering the collection frequency or configuring threshold policies to account for transient states. (See Additional Information)
Ensure connectivity loss extends beyond the collection interval window when testing alerting rules.
Additional Information
Feedback
