Mastering MTR: Install, Run, and Decode Network Paths

What Is MTR?

MTR (My Traceroute) combines the functions of ping and traceroute into a single network‑diagnostic tool. Unlike the traditional traceroute which uses UDP packets, MTR sends ICMP packets by default, which may be deprioritized on some routers, causing measured values to appear lower than the actual performance.

Installation Methods

Windows

Download https://cdn.ipip.net/17mon/besttrace.exe (BestTrace) and install.

Alternatively, obtain the portable WinMTR binary from https://github.com/oott123/WinMTR/releases and run without installation.

Linux

Debian/Ubuntu: sudo apt install mtr RedHat/CentOS: sudo yum install mtr macOS

Search for “Best NetTools” in the App Store and install.

Android

Download TracePing from Google Play (or from https://dwz.cn/KCdNPH4c when Play Store is inaccessible).

Basic Usage

Run mtr qq.com to view the route and connection status from the local machine to qq.com. The output includes several columns whose meanings are explained below.

Key Parameters

-r or --report : Generates a static report by sending 10 ICMP packets (default). Use mtr -r qq.com for a one‑time snapshot.

-s or --packetsize : Sets the size of each ICMP packet. Example: mtr -s 100 qq.com sends 100‑byte packets.

-c : Specifies the number of packets to send. Example: mtr -c 100 qq.com sends 100 packets.

-n : Disables hostname resolution, showing only IP addresses. Example: mtr -n qq.com.

Understanding the Output Columns

IP address – first column.

Loss – packet loss percentage.

Snt – number of packets sent.

Last – latency of the most recent packet.

Avg – average latency.

Best – minimum latency.

Wrst – maximum latency.

StDev – standard deviation, indicating latency stability.

Analyzing Packet Loss

Identify any hop with a non‑zero loss percentage. If loss appears only on a single hop but the next hop shows 0 % loss, the loss is likely due to ICMP rate limiting by that router. If consecutive hops all show loss, the problem is probably upstream.

Analyzing Latency

Latency generally increases with hop count due to physical distance and line quality. Large jumps in the Avg column (e.g., from hop 11 to hop 12) may indicate a long‑distance link, misconfigured router, or congestion. High StDev values suggest unstable latency, and using both best and worst values can give a more realistic picture than the average alone.

Troubleshooting Steps

Run a forward MTR test to locate loss or high latency.

If loss is observed, check the next hop to determine whether it is a true network issue or an ICMP rate‑limit artifact.

Perform a reverse MTR test (from the destination back to the source) to see if the return path differs.

Collect the MTR report and share it with the ISP or network provider when contacting support.

Conclusion

Most routing anomalies are temporary and resolve within 24 hours. Persistent loss or latency usually stems from ISP‑level problems, long physical distances, or peak‑time congestion. Using MTR’s detailed statistics helps pinpoint where the issue lies and what information to provide to service providers for faster resolution.