A traceroute detects and captures the path that a network packet will travel from a given starting point (a computer that has the PerformanceGuard agent installed) to a specific destination. While PerformanceGuard detects the route, it will also record the network latency between different routers along the route until reaching the final destination.

This requires that your organization has set up application pings. See Monitor Availability, Latency and Response Times with Application Ping for details about how to create and activate an application ping.

Latency Is One Several Factors

It's important to know that network latency is one factor that can contribute to "long waits" experienced by users, but it's not the only factor. For example, network routes may work fine, but users may still experience long waits if the server that they communicate with responds slowly because it's busy.

PerformanceGuard helps you identify the roots of problems, but it can also help you quickly eliminate potential suspects. That's why the traceroute feature can be really useful: It only takes a few seconds to view traceroute information, but it can save you hours of troubleshooting time if you can quickly rule out network latency issues.

View Traceroute Output

To view traceroute output, select ANALYZE > Overview > Trace Route and then either Trace Route Graph or the simple table-based Trace Route Overview.

These are the settings with which you can set up and view a traceroute graph (the settings for the table-based traceroute overview are really simple):

• Trace route: Select the traceroute that you want to view
• From Network: Select the network group that should be the starting point of your traceroute
• Interval: Select the period of time that you want the traceroute to cover. If the predefined intervals don't suit you, select Custom to specify your own interval.

Additionally, you can select Advanced Options:

• List only networks with at least ...: Located next to the From Network list, this option lets you filter the list so that it only contains networks with the required number of reporting computers.
• Format:
  • Show the trace in ... format: Select how to show the traceroute: Image or table.
• Extra Options:
  • Show Max and Show Min: Show the maximum and minimum network round trip times respectively.
  • Collapse network groups: By default, the starting point on a traceroute graph is formed by the network groups of the computers that performed the traceroute request, but this option lets you collapse all network groups into a single group.
  • Show Edge Weight: Show the percentage of traceroute requests that followed a specific path.
  • Hide non-replying routers.
• Included links:
  • Include ... traveled links: Select which links between routers should be shown: Often traveled links, all traveled links, or rarely traveled links.

When ready, click the Show button.

Traceroute Graph Colors and Line Thickness

The line colors in a traceroute graph (click thumbnail to view image in full size) mean the following:

1. Green: Responsible for a latency of up to 20 milliseconds
2. Yellow: Responsible for a latency of 21-50 milliseconds
3. Red: Responsible for a latency of more than 50 milliseconds
4. Blue: Not reachable

The line thickness means the following:

• Dashed: Less than 5 % of packets traveled this way
• 1 pixel: 5-19 % of packets traveled this way
•  2 pixels: 20-39 % of packets traveled this way
• 3 pixels: 40-59 % of packets traveled this way
• 4 pixels: 60-79 % of packets traveled this way
• 5 pixels: 80-100 % of packets traveled this way
• Blue dashed: Node isn't reachable

To further break down the latency between two routers, click an arrowhead to produce a graph of the latency over time between the two routers:

How Does a Traceroute Work?

A traceroute measurement works by sending a series of data packets to the required destination (defined as an IP address or hostname) with increasing TTL (Time-To-Live) values starting from 1.

As a packet travels across the network, routing devices will decrement the TTL value of the packet by 1 every time the packet is routed. Eventually, the packet will either reach its destination or the TTL value will become zero. If a router detects a packet with a TTL value of zero, it will return a Time Exceeded packet to the sender.

By comparing records of when you sent the packet with TTL=x with records of when you received a Time Exceeded response from a router, you'll know the roundtrip time to router number x on the route.

By repeating this step with increasing TTL values, you can record the roundtrip time to each router along the route. You can then subtract these roundtrip values from each other to get the latency between any two routers along the route.