Skip to content
Log In
Oh Dear

Before the client calls

Ping monitoring.

See the host go unreachable before the client reports it. An ICMP check against a hostname or IP address, for the machines around your sites that have no homepage to request.

Start monitoring See how it works

10-day free trial · No credit card required · Every feature included

Capterra
4.9 (73) G2
4.7 (33)
Checked every minute Hostname, IPv4 or IPv6
  • Bitmovin
  • HBO Nordic
  • Obsidian
  • Laravel
  • Fathom Analytics
  • PHP 8
  • Stanford University
  • Takeaway.com
  • IGN
  • VRT NWS
  • spatie

The maintenance relationship

The client should not be the one who tells you.

Your agency already watches client sites. Around them sit mail hosts, database servers, gateways, staging boxes and network devices that are not websites at all.

The call comes in as "the server is down". By the time a client says that sentence out loud, the maintenance relationship has already been damaged, whatever you find when you look.

Ping is the network-layer answer to that: is the machine reachable at all? It covers the hosts that have no homepage to request, which is most of the infrastructure sitting behind a client site.

It is deliberately a small signal. A host that answers ICMP can still be running a broken application, a full disk, or a mail queue nobody is processing. What it gives you is the earliest possible confirmation that the box itself has gone.

Casey Sprague
Oh Dear detected an issue that our other uptime monitor did not!

Casey Sprague, CTO at thera-LINK

A host and a website are two different questions. This check answers the first one.

A clear boundary

Reachable is not the same as working.

Ping gives you one low-level signal. It does not replace the checks that look at a website, a port, or the state the application reports about itself.

Ping fails

The problem is below the application.

The host is not answering ICMP from the checker. Whatever else is wrong, it is not worth debugging the app until the machine is back.

Ping succeeds

The machine is up. That is all you know.

The site, the port, and the application can each still be failing. A green ping is a minimum, not a guarantee.

What it checks

  • Whether the host answers ICMP from Oh Dear's checkers.
  • Packet loss, measured against a threshold you set.
  • Average round-trip time, measured against a threshold you set.
  • Hostnames, IPv4 addresses and IPv6 addresses.

What it does not

A host that answers ICMP can still be running a broken application, a full disk, or a mail queue nobody is processing.

Operational ownership

Send the incident to the person who owns the fix.

A gateway going offline at one client is not the same person's problem as a staging box at another. Route each host to whoever is responsible for it, on the channels that team already uses.

Hosts you monitor this way sit next to the client's website checks in the same account, so one view covers both the site and the machine underneath it.

Slack
Email
Webhook
Explore notifications and routing

Agency first

Who needs this check most.

Agencies and MSPs with many client machines: the mail, database, gateway, staging and network-device hosts that sit around the sites they already look after.

One agency, unrelated clients, clear ownership.

The stacks do not have to match. Ping watches network reachability per host without an agent, and each incident stays attached to the client and the person who owns the fix.

More on how agencies run this at Oh Dear for web agencies.

Hosting providers and MSPs, same job at a different scale.

The question across a fleet of customer machines is identical: which of these stopped answering, and who needs to hear about it?

E-commerce teams use the same signal around a shop and the hosts behind it, where an offline database server shows up as a checkout failure long before anyone looks at the box.

No feature tiers

Every check.
Every plan.

Ping monitoring is on every plan. You pay for the number of hosts and sites you monitor, not for unlocking ICMP.

Start with everything on.

Try ping monitoring alongside website uptime, DNS, certificates and the rest of the website-health toolkit.

  • Ping monitoring on every plan
  • Every monitoring feature included
  • Plans scale by monitored site count
  • No credit card required
Start a free trial

10-day free trial. No credit card.

For the technically curious

How the ping check actually works.

Oh Dear sends ICMP to the hostname or address you configure, from its own checkers. The work happens outside the host, so there is no resident agent.

  • Targets. A hostname, an IPv4 address or an IPv6 address.
  • Cadence. Every minute by default, and settable per monitor anywhere from one to sixty minutes.
  • Packet settings. The number of packets sent, the interval between them, the per-packet timeout, the packet size and the maximum number of hops are all configurable per monitor.
  • Failure thresholds. Set an acceptable packet-loss percentage and an acceptable average round-trip time. New monitors start at 0% loss and 400 ms. The host is marked down when a run exceeds either one, so a host that answers but answers badly still raises an alert.
  • Two locations. Each monitor has a primary location you choose. When a run fails there, Oh Dear re-checks from a checker in a different city, on a different provider, before calling it down. If the primary fails and the second one succeeds, the run is a warning rather than downtime, so it does not dent your uptime figure.
  • The report. Each run keeps the round-trip minimum, maximum, average and standard deviation, the packets sent and received, and the raw ping output, so you can tell a host that vanished from one that has been degrading for a week.
  • Network access. The checks come from Oh Dear's own addresses. If a host should not answer ICMP from the open internet, restrict it to those addresses rather than blocking ICMP outright.
  • What it will not tell you. There is no traceroute and no per-hop detail. A failure means the host did not answer the checker, not that you know which link in between dropped the packet.

Want all the technical details?

Uptime and performance monitoring Which IPs we monitor from

Before the first check

Questions about ping monitoring.

What ICMP observes, how it differs from the neighbouring checks, and what you configure.

What is ping monitoring and when should I use it?

Ping monitoring sends an ICMP echo to a hostname or IP address and alerts you when the machine stops answering. Use it for hosts that serve no website at all, or as the network-layer "is the machine even up" signal next to a website check.

Does a ping check return HTTP 200 or "OK"?

No. This check speaks ICMP, not HTTP, so there is no status code and no response body. If you want Oh Dear to read a health status your application reports, that is application health monitoring.

How is ping different from website uptime monitoring?

Website uptime monitoring makes an HTTP(S) request to a public URL. Ping asks a narrower question one layer down: does this machine answer at all? A host can answer every ping while the site on it returns 500 to every visitor.

How is ping different from TCP port monitoring?

Ping tells you the machine answers ICMP. TCP port monitoring tells you a specific port is accepting connections. A box can ping perfectly while port 25 refuses every connection.

Can I ping an IPv6 address?

Yes. The check accepts a hostname, an IPv4 address or an IPv6 address.

Do I need to install an agent?

No. The ping comes from Oh Dear's own checkers, from outside your network, so there is nothing to install on the host. If a firewall restricts ICMP, allowlist the addresses Oh Dear checks from.

Does ping monitoring show me where the packet died?

No. It reports whether the host answered and how it answered, not the route in between. It does not produce a traceroute, so a failure tells you the host is unreachable from the checker rather than which hop dropped it.

Is ping monitoring included in every plan?

Yes. Every feature is on every plan, so you pay for the number of hosts and sites you monitor rather than for unlocking ICMP.

See all other FAQ items
The public website Monitor HTTP(S) availability The next layer down Check a specific port

Get started

Start monitoring the hosts you're responsible for.

Add a host in under a minute. Every feature on. No credit card.

Capterra
4.9 (73) G2
4.7 (33)
Start monitoring