Ping Monitoring & ICMP Uptime Alerts

HTTP checks are great for websites — but sometimes you just need to know whether a host is reachable. UpDog pings your host and alerts when it stops responding (and when it recovers).

Start free Back to features
Ping monitoring dashboard showing latency and an alert

What is ping monitoring?

Ping monitoring uses ICMP echo requests (the classic “ping”) to test whether a host can be reached over the network. It’s a fast, lightweight way to answer a simple question: can we reach this machine right now?

It’s important to treat ping as a reachability signal — not an application health check. A host can reply to ICMP while your web server is down, and a site can serve pages while ICMP is blocked. That’s why ping works best as part of a small set of complementary checks.

What UpDog measures

UpDog runs ICMP ping checks on the schedule you choose and records results over time. For each target, you’ll see:

  • Host up/down: whether the host is reachable or unreachable.
  • Latency (RTT): round-trip time from the ping check, tracked over time.

UpDog does not report packet loss on the ping monitoring page. If you need packet-loss percent specifically, you’ll typically pair monitoring with network tooling — and still keep ping alerts for fast “something changed” signal.

What common failures look like

  • Timeouts: the request goes out but no reply comes back in time (often packet drops, filtering, congestion, or a host that’s overloaded).
  • Unreachable: the network can’t route to the host (often a routing change, gateway issue, or host/network down).
  • High RTT: the host is reachable but latency is elevated (often early warning of saturation or regional network issues).

When to use ping vs HTTP monitoring

Use ping monitoring when you care about network/host reachability. Use HTTP monitoring when you care about service availability — status codes, redirects, and (optionally) content checks.

Ping monitoring (ICMP)

  • Routers, gateways, and VPN endpoints
  • Bare servers and internal hosts
  • Private infrastructure where HTTP doesn’t exist (or shouldn’t)

HTTP monitoring

  • Websites and landing pages
  • APIs (health endpoints + real user paths)
  • “Is it working?” checks: status codes + response time

Practical pattern: monitor both. Ping tells you when a host becomes unreachable. HTTP tells you when your service is actually failing — even if the host is up.

Catch network issues early

Ping checks won’t diagnose every incident, but they do surface the “something changed” moments quickly — especially when the problem is below the application layer.

  • ISP outage / upstream routing issue: the host looks unreachable from the outside, even though it’s still powered on.
  • Firewall rules changed: a security group update drops ICMP or blocks return traffic.
  • Host powered off / hypervisor problem: the machine disappears and doesn’t respond at all.
  • DDoS mitigation / rate limiting: some defenses drop ICMP, which can create false negatives (ping down while the service still works).
  • Regional connectivity issues: latency rises or timeouts appear intermittently, often before you see total failure.

Set smart alerts (avoid flukes)

Alerting on a single missed ping is an easy way to create noise. UpDog reduces flukes by requiring multiple consecutive failed checks before sending a down alert.

You’ll also get alerted on recovery, so you know when the host becomes reachable again — without having to keep refreshing dashboards.

A simple mental model

Treat ping alerts as “network reachability changed.” If the host is down, you can quickly pivot to HTTP/port checks to answer “is the service down too, or is ICMP blocked?”

How to set up ping monitoring

  1. Enter an IPv4 address or hostname for the host you want to monitor.
  2. Choose a check interval that matches how quickly you need to know (UpDog uses a fixed timeout under the hood).
  3. Pick alert channels — email and SMS — so the right people get “down” and “recovered” notifications.

Best practices

  • Prefer hostnames when appropriate: if IPs change (cloud instances, failovers), a hostname can keep monitoring stable.
  • Whitelist ICMP when needed: many networks drop ping by default — open it intentionally (or rely on HTTP/port checks).
  • Monitor both reachability and service health: pair ping (host/network) with HTTP monitoring and/or port monitoring (service).

FAQ

Yes. UpDog uses ICMP echo requests (what most people mean by “ping”) to test whether a host is reachable. It’s a network-level signal — not an application-level check.

Ping monitoring is best for network and host reachability: routers, gateways, VPN endpoints, servers, and private infrastructure where an HTTP endpoint isn’t the right signal. It answers “can we reach it?” quickly, so you can decide what to investigate next.

Some hosts and networks block or rate-limit ICMP, even while HTTP/HTTPS traffic is allowed. That can make ping look “down” while the service is still reachable in a browser. Pair ping with HTTP monitoring to distinguish reachability from application availability.

Ping only tells you the host is reachable at the network level. Your web server, app, database, TLS, or DNS can still be failing while the machine replies to ICMP. Use HTTP monitoring (and optionally port monitoring) to verify the service itself.

Yes — you can monitor an IPv4 address directly. If the IP may change over time (cloud instances, failover, load balancers), a hostname can be a better long-term target.

Today, UpDog ping monitoring supports IPv4 targets. If you run dual-stack infrastructure, a common approach is to monitor IPv4 reachability via ping and monitor the IPv6-facing service via HTTP (or a port check) until IPv6 ping is available.

Ping checks run on the interval you choose. Shorter intervals detect outages faster; longer intervals reduce noise. For most infrastructure targets, start with a practical interval and tighten it once you’ve seen stable behavior.

Transient packet drops, upstream routing changes, and ICMP rate limiting can all cause missed replies. UpDog uses multiple consecutive checks before alerting to reduce flukes, but the best mitigation is to pair ping with HTTP/port monitoring so one signal can confirm the other.

Yes. UpDog records round-trip time (RTT) from ping checks and displays it over time. That makes it easier to spot rising latency, correlate with incidents, and answer “when did it start?”

Yes. Many security groups and firewalls drop ICMP by default. If you want ping monitoring, allow ICMP echo requests/replies for your host (ideally scoped as tightly as you can). If you can’t open ICMP, use HTTP or port monitoring instead.

Ping monitoring checks whether the host is reachable over the network. Port monitoring checks whether a specific service is accepting connections on a specific port (like 443, 22, or 5432). If ping is down, port checks will usually fail too — but if ping is up and the port check fails, that often points to the service itself.

UpDog ping monitoring focuses on reachability and RTT latency. It doesn’t display packet loss percentage on this page. If loss is a key metric for you, use ping alerts to catch incidents quickly, and supplement with network tooling that calculates loss across a window.

Related features

Build a simple, layered monitoring setup that tells you both “reachable?” and “healthy?”

Get ping alerts without the noise

Add a host, pick an interval, and get “down” + “recovered” alerts when reachability changes — with RTT history to help you debug what happened.

Start Monitoring for Free Explore features