Skip to main content

Traceroute online

Trace the exact network path from HostTracker's worldwide locations to any website, server, or IP address, and see the response time at every router along the way.

Examples: 300+ locations worldwide · results in about 10 seconds
{{error.message}}

Unlock the full report

30-day free trial · cancel anytime

Loading plans…
Recommended
{{ plan.name }}
{{ money(priceMo(plan)) }}/mo
  • 1 website · all monitoring types
  • {{ plan.maxTaskCount == null ? 'Unlimited' : plan.maxTaskCount }} monitored sites
  • All check locations in the report
  • TLS, headers & timing details
  • Scheduled reports
  • REST API access
Start the 30-day trial
Billed only after your trial · Compare all plans

How trace ip tool works

Tracert online sends out a sequence of packets to the destination address. This process allows to determine the tracert ip address of each router in the path by incrementally increasing the TTL value of successive batches of packets. Trace route is also measure the round-trip time of each packet, providing information about the delay introduced at each hop. Online tracert tool usually helps with finding out the source of the problem, if this source is external or network-related

Find out more

How website trace is useful

Our traceroute tool is very useful for diagnosing the problems on the network and their severity. This tool measures the time of reception and transmission of data packets from each router encountered along the way when you trace a domain or IP address. It will allow you to pinpoint a variety of problems on your network

Find out more

Online traceroute convenience

Our HostTracker online tracing tool has its own additional features that are beneficial for full-scale online diagnostics:

  • Our web traceroute utility has the variety of available access points that will allow to use a tracert tool from different regions and different ISPs to pinpoint the source of network problems
  • Performing traceroute ip address HostTracker provides its complex logging and data collecting system. It allows users to compare diagnostics results

Find out more

More about traceroute - how to read hops, latency jumps, and timeouts.

Need this checked around the clock, not just once? Set up availability monitoring

Key takeaways

  • This tool runs a traceroute to the host you enter and lists every router hop on the way, with the response time measured at each one.
  • A hop showing asterisks is usually a router that declines to answer traceroute probes while still forwarding real traffic; what matters is a latency jump that persists for every hop after it.
  • The trace runs from several HostTracker locations at once, which separates a routing problem on one path from a problem at the destination.
  • It is free, needs no login and no credit card.
  • For continuous checks, see the network path monitoring page: ping and port monitors watch reachability on a schedule, and a traceroute is the diagnosis you run afterwards.

Frequently asked questions.

Read it hop by hop: each line is one router and the times of three echo requests sent to it. Times should rise with distance - fiber adds roughly 1 ms of round trip per 100 km, and an ocean crossing adds tens of milliseconds. A jump that stays high to the final hop is distance, congestion or a detour; a jump that disappears on the next hop is a router answering pings to itself at low priority; ignore it. Private addresses on the first hops (10.x, 172.16-31.x, 192.168.x, 100.64-127.x) are normal; read from the first public address on. The same addresses alternating until hop 30 is a routing loop.
A * means a probe got no reply within 1 second. Stars on one or two hops in the middle, with normal hops after them, are harmless: the router ignores ICMP or rate-limits replies to itself while still forwarding traffic, so judge the path by the hops after it. This tool sends three probes per hop, so a lone * next to two normal times is a single lost reply.
Many servers, load balancers and cloud firewalls drop ICMP echo requests while serving web traffic normally, so the trace ends one or more hops early and the last address does not match the IP column. That is not an outage. Check the service itself with the HTTP check or the port checker on port 443, or trace over TCP from your own machine.
Compare the locations. If every location stops or slows at the same place, the problem is at or near the server: confirm with an HTTP check and send your provider the last address that answered. If only one region stalls, it is a routing or peering problem between that region and your provider; rerun with that region picked in the location menu, then send both traces to your host. If every location reaches the server but you cannot, look at your own network, VPN or ISP. For a browser error rather than a slow path, the ERR_CONNECTION_TIMED_OUT guide covers the same symptoms from the visitor's side.
A traceroute shows one path only, from where it runs to your server, and each checkpoint reaches your provider through different networks. If the IP column differs between locations, your host name resolves to different servers by region (a CDN or geographic DNS), so each row traces a different server; compare paths only between rows that share an IP. Replies can also return by a different route (asymmetric routing), so a slow return path can show up as a jump on a hop whose forward link is fine.
They do the same job with different probes. Windows tracert sends ICMP echo requests, while Linux and macOS traceroute sends UDP probes by default, so the path and the silent hops can differ between them. Both send three probes per hop and give up after 30 hops by default. The probe type matters when a firewall filters one kind and not another.
ICMP. It sends ICMP echo requests, like Windows tracert, so its path and its silent hops can differ from a Linux or macOS traceroute, which sends UDP probes by default. Each hop gets three timed echo requests. If a firewall on the way drops ICMP, run a TCP trace from your own machine.
Pick a region in the location menu next to the address field instead of All world, and the trace runs only from HostTracker checkpoints there. Each row names its checkpoint's location, so you can compare that region's path with the rest of the world. With All world, the default, the trace runs from many checkpoints at once.
Trace from your own machine to see the path your office or users actually take:
  • Linux and macOS, UDP probes by default: traceroute example.com
  • ICMP probes like this tool, may need sudo: traceroute -I example.com
  • TCP to port 443, often gets past firewalls that drop ICMP and UDP (Linux, root): sudo traceroute -T -p 443 example.com
  • Windows, ICMP; -d skips reverse DNS: tracert -d example.com
  • mtr, repeats the trace and reports loss per hop (100 rounds): mtr -rwc 100 example.com
  • mtr over TCP: mtr --tcp --port 443 -rwc 100 example.com
Up to 30. The trace stops as soon as the destination answers, so a nearby server shows fewer hops, and a trace that never reaches it ends at the last router that replied. Silent hops in the middle stay in the list and count toward the Hops column.
Only roughly. This tool sends three timed echo requests to each hop, so a * among them is a lost reply, but three samples cannot give a loss rate. For a percentage, run mtr with 100 or more rounds from your own machine. Loss on a middle hop that does not continue to the final hop is rate limiting; loss that starts at one hop and stays to the end is real.
Yes, it needs no signup and no credit card. Without an account the table shows six of the locations in full; sign in with a free account to see every location. Each row names the location of the checkpoint that ran it, so the full list lets you compare paths across regions.

Part of HostTracker's website monitoring software.