traceroute

Additional Notes

traceroute shows the routers (hops) between your machine and a destination. It helps locate where latency or packet loss appears along a path by sending probes with increasing Time-To-Live (TTL) values.

Each hop is discovered by listening for ICMP "time exceeded" replies. Some networks block or rate-limit these probes, so missing hops do not always mean a broken path.

Syntax

traceroute [options] host

Parameters

• options: Flags that change how traceroute behaves.
• host: Hostname, IP address, or destination to trace.

Common Options

• -n: Do not resolve hostnames; show numeric addresses only.
• -I: Use ICMP echo probes (like ping).
• -T: Use TCP SYN probes when supported.
• -U: Use UDP probes to a destination port.
• -p PORT: Set the base destination port.
• -m HOPS: Set the maximum number of hops (default 30).
• -f HOPS: Start probing at a given hop (skip early hops).
• -w SECONDS: Wait time for each reply.
• -q N: Number of probes per hop.
• -4 / -6: Force IPv4 or IPv6.
• -z MS: Wait between probes (milliseconds).

Examples

traceroute example.com

Trace the path to a domain using default UDP probes.

traceroute -n 8.8.8.8

Trace to an IP without DNS lookups.

traceroute -I example.com

Use ICMP echo probes, which often pass where UDP is filtered.

traceroute -T -p 443 example.com

Use TCP SYN probes toward port 443 (useful through firewalls).

traceroute -m 15 -q 1 example.com

Limit to 15 hops with a single probe per hop for a faster trace.

traceroute -4 example.com

Force an IPv4 trace.

Practical Notes

• * * * means no reply for that hop, which is not always a failure; it may be filtered.
• Use ping first to confirm basic reachability before tracing.
• ICMP (-I) or TCP (-T -p 443/-p 80) probes often succeed where default UDP is blocked.
• Routing can be asymmetric; the return path may differ from the path shown.
• If installed, mtr gives continuous traceroute-style monitoring over time.