IP & Network
ICMP ping
Measure round-trip latency to any host or IP, live.
Host or IP, packet count
IPv4, IPv6 or domain. Public addresses only.
About this tool
Ping sends ICMP echo requests to a host and measures how long each reply takes. It is the simplest, oldest reachability test on the internet — and still the fastest way to confirm whether a server is up, how far away it sits in network-time, and whether the path between you and it is stable.
Our tool sends 1 to 30 echoes from our Frankfurt server, streams each reply to your browser as it arrives, and finishes with the standard min/avg/max/mdev summary plus packet-loss percentage. Time values come straight from the kernel, so what you see is the actual round-trip on the wire — not request-handling overhead.
When to use this tool
- Confirming a server is reachable. A clean run with 0% loss means the host is up and ICMP is allowed end-to-end.
- Comparing latency to several locations. Run the tool against the same target from different regions to see which datacentre is closest in network-time.
- Spotting an unstable path. Steady RTTs are healthy; jitter (high
mdev) and intermittent loss point to a congested or flapping link somewhere along the way. - Sanity-checking a deploy. Right after a DNS or load-balancer change, ping the new IP to verify it is responsive before sending real traffic.
How to read the result
seq is the echo number (1, 2, 3 …). ttl is the remaining time-to-live in the reply — it tells you roughly how many hops the packet took to come back. time is the round-trip in milliseconds. The summary at the bottom gives you the minimum, average, maximum, and mean deviation across all successful echoes, plus how many packets were lost. A few percent of loss is usually noise; sustained loss above 5–10% is worth investigating.
Frequently asked questions
Why does the host show 100% packet loss when I can browse it?
Many hosts (especially cloud load balancers, CDN edges, and home routers behind NAT) drop ICMP echo by policy while still serving HTTP, SSH, or other TCP traffic. 100% ping loss is not a reliable "down" signal for those — try a TCP-level check (Port Checker) before assuming the service is unreachable.
What is mdev and why does it matter?
mdev is the mean deviation of round-trip times — a measure of jitter. Low mdev (under a few ms) means the path is stable; high mdev relative to the average means RTT is bouncing around, which is bad news for VoIP, gaming, and any latency-sensitive protocol. Audio call quality starts degrading around 30–50ms of jitter regardless of average latency.
Are private IPs and link-local addresses allowed?
No. To prevent abuse and accidental leaks, the tool refuses RFC 1918 private ranges (10/8, 172.16/12, 192.168/16), loopback, link-local (169.254/16), CGN (100.64/10), and cloud metadata endpoints (169.254.169.254 etc.). Pinging your own LAN should happen from your own machine, not from a public service.
Can I increase the number of echoes?
Yes — pick 5, 10, 20, or 30 from the dropdown. More echoes give you a tighter average and a better sense of jitter, but each echo waits 1 second by default, so 30 echoes take roughly 30 seconds. For a quick reachability check 5 is plenty; for diagnosing intermittent loss 20+ is more useful.
How is this different from running ping in my terminal?
The ICMP probes leave from our Frankfurt server, not your machine, so absolute RTTs reflect the network distance from us to the target — useful for checking whether the target is reachable from outside your own network, but not the same as your local latency. Run the same ping locally to compare; if our numbers are good and yours are not, the issue is in your connection, not the destination.