Watch anything with an IP address.
Ping monitors send ICMP echo requests to anything with an address: routers, firewalls, VMs, the NAS in the office. When a host stops answering, you hear about it on the channels you use.
Watch hosts without a web server.
Most monitoring starts at HTTP, but half of your infrastructure never speaks it. Switches, firewalls, hypervisors and storage boxes still fail, they fail without a status endpoint to poll.
A ping monitor needs an address and an interval. Nothing gets installed on the target and no port has to be exposed.
- ICMP echo, no agent on the target
- Hostname or IP address, either works
- Paste a URL and Checkmate extracts the host
Chart round-trip times on every check.
Reachability is binary, latency is not. Every check records the round-trip time, so a saturated uplink or a struggling VPN tunnel shows up as a climbing chart before it becomes an outage.
Ping monitors get the same detail page as every other monitor: uptime percentage, response times and the incident history.
- Round-trip time recorded on every check
- Uptime and incident history per host
Ping from where your users are.
Like HTTP monitors, ping monitors support distributed checks through the GlobalPing network. The same echo request runs from 6 continents, so you can tell a dead host from a broken route.
Reachable from the office but dark from Singapore is a real failure mode. Distributed ping is how you see it.
- Distributed checks from 6 continents
- Per-region reachability and latency
For the person who owns the network.
Homelab operators
The NAS, the Proxmox nodes, the Pi running DNS. Know a box dropped off the network before you are standing in front of it wondering why the lights are off.
IT teams with hardware
Switches, access points, branch-office links and printers on one dashboard. The oldest diagnostic tool in networking, run on a schedule instead of when someone remembers.
MSPs and agencies
Client networks you cannot install agents on. Ping needs nothing from the far end, which makes it the check you can roll out across client sites on day one.
See what it covers and what it doesn't.
Covered
- ICMP echo reachability on your schedule
- Round-trip time recorded per check
- Hostname or IP address targets
- Distributed checks from 6 continents
- Hosts on private networks, since checks run from your instance
- Alerts on 13 notification channels
Out of scope
- Whether the service on the host works: a machine can answer ping while the app is down, pair it with an HTTP or port monitor
- Hosts that block ICMP: some clouds and firewalls drop echo requests, use a TCP port monitor there
- Path diagnostics: no per-hop traceroute or packet-loss analysis
For the technically curious.
Plain ICMP echo
Each check sends a standard echo request and records whether a reply came back, plus the round-trip time. A reply means up, silence past the timeout means down.
No agent and no open port
The target needs nothing installed and nothing exposed. If the host's network stack answers ping, monitoring works, which is why this check fits hardware the other monitor types cannot cover.
Private networks included
Checks run from your own Checkmate instance, so 10.x and 192.168.x space is as monitorable as the public internet. Hosted ping services cannot reach those addresses at all.
Distributed echo
Ping is one of the 2 monitor types that support GlobalPing distributed checks, so reachability can be measured from 6 continents instead of only from your server's rack.
Frequently asked questions
Ping monitoring sends ICMP echo requests to a host on a schedule and alerts you when the host stops replying. Checkmate records the result and round-trip time of every check, building an uptime and latency history per host.
Anything with an IP address: servers, routers, switches, firewalls, NAS boxes, VMs and IoT devices. Ping is the right check for hardware that has no HTTP endpoint to poll.
Some cloud providers and firewalls drop echo requests, so a healthy server can look down to ping. For those hosts use a TCP port monitor or an HTTP monitor instead, both are one dropdown away in the same create dialog.
Yes. Checkmate is self-hosted, so checks originate from your own instance. Anything your instance can route to, including RFC 1918 space behind your firewall, can be monitored.
You pick the interval per monitor. Latency-sensitive links can be checked every few seconds while quiet hardware gets a relaxed schedule, and each monitor keeps its own history either way.
Checkmate is one. It is open source under AGPL-3.0, so you self-host it and monitor as many hosts as you like with no per-monitor pricing.
