Ping (ICMP) Monitoring

ICMP ping monitoring for network-level availability checking. Tests basic network connectivity to a host by sending ICMP Echo Request packets and measuring the response.

How It Works

Hoppla sends ICMP Echo Request (ping) packets to the target host and measures the round-trip time for each reply. This is the most fundamental connectivity check available -- if a host does not respond to ping, it is either down, unreachable, or blocking ICMP traffic.

Ping monitoring operates at the network layer (Layer 3), below TCP and HTTP. It verifies that the host is reachable on the network without testing any specific service or port.

Configuration

Setting	Description	Default
Host	Hostname or IP address to ping	Monitor URL host
Packet count (ping_count)	Number of ICMP packets to send per check (1–10)	3
Timeout	Maximum time to wait for each reply (seconds)	5
Check interval	How often to ping the host	60 seconds

Results

Field	Description
Packets sent	Number of ICMP packets transmitted
Packets received	Number of replies received
Packet loss	Percentage of packets that did not receive a reply
RTT min	Minimum round-trip time (ms)
RTT avg	Average round-trip time (ms)
RTT max	Maximum round-trip time (ms)

Example Result -- Healthy

Ping: server.example.com (203.0.113.42)
  Packets:     3 sent, 3 received, 0% loss
  RTT min:     8.2ms
  RTT avg:     12.5ms
  RTT max:     18.1ms

Example Result -- Degraded

Ping: server.example.com (203.0.113.42)
  Packets:     3 sent, 2 received, 33% loss
  RTT min:     45.3ms
  RTT avg:     67.8ms
  RTT max:     90.2ms

Example Result -- Unreachable

Ping: server.example.com (203.0.113.42)
  Packets:     3 sent, 0 received, 100% loss

Alerts

Event	Trigger	Severity
incident.created	Host is unreachable (100% packet loss, or ping output could not be parsed)	Incident
warning.degraded_performance	Partial packet loss (between 0% and 100%)	Warning

An incident is created when all ICMP packets are lost (or the ping output is unparseable), indicating the host is completely unreachable. The incident is resolved automatically when the host starts responding again.

Any partial packet loss (greater than 0% but below 100%) sets the check status to warning and can fire the warning.degraded_performance notification event, subject to your notification rules. Only 100% loss opens an incident.

Use Cases

Scenario	Why Ping
Network infrastructure	Routers, switches, and firewalls that do not run HTTP services
Bare-metal servers	Verify host is powered on and network-connected before testing services
VPN gateways	Check reachability of VPN endpoints
IoT devices	Simple availability check for devices with no web interface
Baseline latency	Track network-level round-trip time trends over time

Limitations

• Some hosts block ICMP traffic via firewall rules. If a host does not respond to ping, it may still be running and serving traffic on specific ports. Use TCP monitoring or Uptime monitoring instead.
• Cloud providers (AWS, GCP, Azure) may block ICMP by default on certain instance types. Check your security group and firewall rules.
• Ping does not test application-level health. A host can respond to ping while its services are down.

When to Use Ping vs. Other Checks

• Ping -- Network reachability only, no service-level validation
• TCP -- Specific port is open and accepting connections
• Uptime -- Full HTTP/HTTPS request-response cycle with status code validation
• Application Health -- Internal component health (database, cache, queue)

Use ping as a first line of defense for hosts where higher-level checks are not applicable, or as a complement to service-level monitors to distinguish between network issues and application issues.