What a ping test measures

Ping sends small ICMP (or equivalent probe) packets to a host and measures round-trip time. It answers a basic question: is the host reachable from our server’s network path, and how long does an echo take? High latency or packet loss often appears before a full outage is obvious in the browser.

When to run ping

• Confirming a server is up after a deploy or reboot
• Comparing latency to a CDN origin vs an apex host
• Separating “DNS wrong” from “host unreachable”
• Documenting baseline RTT before and after a network change

How to interpret latency and loss

Stable low RTT with zero loss is healthy for that path. Spiky latency can mean congestion or overloaded middleboxes. 100% loss may mean the host is down — or that ICMP is firewalled while HTTPS still works. Always confirm with an HTTP check if the site serves web traffic but blocks ping.

Common mistakes

• Assuming ping failure equals website down (ICMP often filtered)
• Comparing RTT from different geographic probes as if they were identical
• Ignoring that mobile and corporate networks may rate-limit echo requests

Example scenario

Users in one region reported a “slow site.” Ping to the origin showed 40 ms locally but 280 ms with intermittent loss toward the region. The traceroute companion tool showed the delay starting at a transit hop — evidence for the ISP ticket, not a false alarm about application code.

Limitations

Probes originate from our infrastructure, not from your laptop. ICMP may be disabled on the target. For hop-by-hop path detail, use Traceroute; for page-level timing, use Speed Test.

Related resources

Traceroute · Speed Test · IP Geolocation · Website Health Check