1. Introduction
When troubleshooting network issues, the first tools that come to mind are usually ping and traceroute. While they are useful, they provide only static snapshots and limited details.
MTR (My Traceroute) combines both tools into one, offering continuous, real-time network diagnostics. This article explains what MTR is, why it is useful, how it works, how to read its results, and how to use it on different operating systems.
2. What is MTR?
MTR (My Traceroute) is a powerful network diagnostic tool that:
- Combines the functionality of ping and traceroute.
- Provides real-time statistics on network latency and packet loss.
- Continuously monitors the path that packets take between your system and a destination host.
System administrators, network engineers, and support teams use MTR to identify connectivity issues quickly and accurately.
3. Why is MTR Useful?
MTR is particularly valuable because it:
- Detects where packet loss occurs in the network path.
- Identifies latency issues at different hops.
- Helps confirm whether a problem lies:
- Within your local network,
- With your ISP,
- In a transit provider’s network, or
- At the destination server.
- Provides evidence for escalation when working with ISPs or vendors.
4. How Does MTR Work?
MTR works by combining:
- Traceroute functionality → Maps the route packets take to the destination.
- Ping functionality → Sends multiple ICMP (or UDP/TCP) packets to each hop to measure round-trip time and loss.
Unlike a single-use traceroute, MTR continuously sends packets and updates results, allowing you to observe changes in real time.
5. How to Read MTR Results
When you run mtr <hostname>, you’ll see output similar to this:
HOST: test-server Loss% Snt Last Avg Best Wrst StDev
1. 192.168.1.1 0.0% 100 1.1 1.2 0.9 2.0 0.2
2. 10.10.20.1 0.0% 100 5.6 5.4 5.1 6.0 0.3
3. 203.0.113.5 2.0% 100 20.3 19.8 18.5 25.0 1.0
4. 198.51.100.2 0.0% 100 45.0 45.2 44.7 46.0 0.5
5. destination.com 0.0% 100 46.5 46.7 45.9 47.5 0.4
Key Columns:
- Host → Router or hop in the path.
- Loss% → Percentage of packets lost at that hop.
- Snt → Number of packets sent.
- Last → Latest round-trip time (ms).
- Avg → Average latency.
- Best → Minimum latency.
- Wrst → Maximum latency.
- StDev → Variability in latency (consistency).
How to Interpret:
- High packet loss at the first hop → Local network issue.
- Loss starting at one hop and continuing forward → Likely provider issue.
- A hop showing loss but later hops are fine → That hop may be rate-limiting ICMP (not a real problem).
- Latency spikes persisting across all later hops → True congestion point.
6. How to Use MTR on Different Operating Systems
On Linux
Install MTR:
# Debian/Ubuntu
sudo apt update
sudo apt install mtr
# CentOS/RHEL
sudo yum install mtr
Run MTR:
mtr google.com
Run in report mode (100 packets):
mtr -r -c 100 google.com
On macOS
Install using Homebrew:
brew install mtr
Run with sudo for full results:
sudo mtr google.com
On Windows
- Use WinMTR (a GUI version of MTR).
- Download from the official WinMTR project page.
- Run
WinMTR.exe, enter the hostname (e.g.,google.com), and start the test.
7. Practical Use Cases of MTR
- Troubleshooting slow website connections.
- Identifying ISP-related packet loss.
- Monitoring connectivity between data centers.
- Providing evidence for support tickets to vendors or ISPs.
8. Best Practices
- Run with enough packets (
-c 100) for accuracy. - Always test in both directions (source → destination and vice versa).
- Do not assume intermediate hop loss is an issue unless it continues beyond.
- Combine MTR results with other tools (
ping,iperf,netstat) for full context.
9. Conclusion
MTR is a versatile, real-time network diagnostic tool that simplifies troubleshooting by combining traceroute and ping. By providing continuous statistics on latency and packet loss, it helps IT teams quickly locate the root cause of network problems and provides evidence for escalation.
If you manage servers, applications, or networks, MTR should be one of your go-to diagnostic tools.