Ping monitoring

Is the host even there? Ask it every 30 seconds.

A ping monitor sends an ICMP echo to a hostname or IP on a schedule and records the round-trip time. When replies stop, you find out. When a host blocks ping, we knock on a port instead.

Free for one monitor. No card needed.

edge.tidepool.ioReachable24 ms
$ ping -c 1 203.0.113.10PING 203.0.113.10 (203.0.113.10): 56 data bytes64 bytes from 203.0.113.10: icmp_seq=0 ttl=54 time=24.1 ms--- 203.0.113.10 ping statistics ---1 packets transmitted, 1 packets received, 0.0% packet loss
  1. 09:41:02Echo replyicmp24 ms
  2. 09:40:31Echo replyicmp25 ms
  3. 09:40:01Echo replyicmp24 ms
  4. 09:39:32Echo replyicmp23 ms
  5. 09:39:02Echo replyicmp26 ms

If the host drops ICMP, the check knocks on ports 443, 80 and 22 instead. A refused connection still proves the host is up.

30s
Fastest interval
ICMP
Echo request
3
Fallback ports
ms
Round-trip kept

What ping tells you, and what it doesn't

Ping is the simplest question you can ask a machine: are you there? It works below HTTP and TCP, so it reaches routers, VPN gateways, bare servers and office connections that have no web page to check. The reply also carries a round-trip time, which is the cleanest measure of the network path between us and you.

What ping cannot tell you is whether the service on that machine works. A server can answer pings with its database on fire. That is why most teams pair a ping monitor on the host with an HTTPS or TCP monitor on the service, and let the status page show both.

Hosts that block ping still get checked

Plenty of cloud firewalls drop ICMP by default. When there is no echo reply, the check tries a TCP connection to ports 443, 80 and 22. A connection, or even a refusal, proves the host is up. Only silence on all three counts as down.

  • Hostnames are resolved once per check, and private addresses are refused
  • Each result records which method answered, ICMP or a port
  • Round-trip times are kept for 30 days, then rolled up daily
Monitors3 up · 1 down
  • edge.tidepool.io
    Ping · every 30s
    24 ms
  • vpn.tidepool.io
    Ping · every 1 min
    31 ms
  • 203.0.113.42
    Ping · every 1 min
    No reply
  • office-router.tidepool.io
    Ping · every 5 min
    48 ms

Alerts when the replies stop

Set the number of failed checks that count as down and the number of passes that count as recovered. Link the monitor to a component if customers should see it, or keep it team-only for infrastructure nobody outside needs to know about.

  • Team alerts on Slack, Discord, Telegram, Teams, email, SMS or webhook
  • Team-only monitors never touch the public page unless you link them
  • Latency is on the monitor's detail page, so a rising trend is visible before a failure
203.0.113.42Outage · 14 min
  1. 11:20:04

    No ICMP reply from 203.0.113.42 within 10 s. TCP 443, 80 and 22 did not answer either.

  2. 11:21:04

    Still nothing. Threshold reached, host marked down.

  3. 11:21:05

    Telegram and email alerts sent to the network team

  4. 11:34:30

    Echo reply in 29 ms, twice in a row

  5. 11:34:31

    Host back to Operational after 14 minutes

Setting one up

01

Enter a host

A hostname or an IP address, IPv4 or IPv6. Ports and paths are ignored; ping does not use them.

02

Choose an interval

Every 30 seconds for the edge, every 5 minutes for the office router.

03

Decide who hears about it

Alert the team, link a component, or both.

The other checks

Ping monitoring questions

Know when a host disappears.

One monitor and a status page, free. No card needed.

Add a ping monitor