WinMTR in Windows: quick analysis of network problems

WinMTR — graphical version of MTR for Windows. The tool combines capabilities traceroute And ping: shows the chain of nodes to the goal and simultaneously counts delays/losses on each hop. Below are completely rethought instructions for running and interpreting the results.

1) Preparation

1. Download WinMTR and run the utility on behalf of administrator.
2. Open Options and uncheck Resolve names, if you need to output IP addresses specifically (without reverse DNS). If desired, leave the polling interval at 1 second and the echo packet size at 64 bytes. Click OK.

2) Start of measurements

1. In the field Host indicate the domain or IP of the node to which access is being checked (for example, 8.8.8.8 or the domain of your website).
2. Click Start and leave the test running: the table will begin to be filled with data for each hop of the route.

3) What do the columns mean?

• Hostname — host name or IP. The message often indicates ICMP filtering on this hop.
• Nr — serial number of the node on the path.
• Loss% — the proportion of lost responses from this node.
• Sent — how many requests were sent.
• Recv — how many responses were received.
• Best — minimum recorded delay (ms).
• Avrg — average delay (ms) for the session.
• Worst — longest delay (ms).
• Last — time of the last received response (ms).

4) How long to take the test and when to sound the alarm

• Duration guideline: minimum 100 requests (at an interval of 1 second this is approximately 100 seconds). You can stop it at any time, but the more packets collected, the more reliable the statistics.
• Reason for in-depth check — repeatable losses 5%+ or average latency ~350 ms and above, and noticeable on several nodes at once, and not on just one.

5) How to prepare a report

Click Export TEXTto download the result to a text file and pass it to the support specialist/network team along with a description of the problem (time, resource, where you tested from).

These steps are enough to localize the bottleneck along the path to the service: see where delays are increasing or losses appear, and prepare a correct report for escalation.

CONTENT:

Similar

All news