Submit a ticketCall us

Get a crash course on Network Monitoring delivered right to your inbox
This free 7-day email course provides a primer to the philosophy, theory, and fundamental concepts involved in IT monitoring. Lessons will explain not only how to perform various monitoring tasks, but why and when you should use them. Sign up now.

Home > Success Center > Network Performance Monitor (NPM) > NetPath™ - Video

NetPath™ - Video

Updated 6-1-2016

Overview

This video (5:11) will demonstrate how to configure and use NetPath™, a feature available in Network Performance Monitor.

 

 

 

Environment

  • NPM 12 and above

Video Transcription

This is NetPath, a whole new way to find the problems that are affecting your users, and solve them quickly.

NetPath’s advanced probing allows you to troubleshoot network issues and performance whether they are on-premises, in a hybrid network, or in the cloud.

NetPath™ detects the network path from a source computer all the way to the destination service. Here, we see how our path to Google.com goes through our network, then our ISP, through some Internet transit providers, and finally through Google's own network before reaching the final destination server.

Up until now, traceroute was the only solution for identifying the paths that packets used to cross the internet. Unfortunately traceroute is usually blocked, cannot account for multiple paths or redundancy, and does not continuously monitor.

If we run traceroute to Google.com, you’ll quickly see that once we leave our internal network we have no visibility into the hop-by-hop performance.

NetPath™, on the other hand, continuously monitors complex multi-path networks and understands redundancy, keeps track of past results, alerts you to changes, and can show you the health of multiple paths at the same time.

Building these paths is easy. Navigate to the NetPath Services page and click Create New Service. Fill in a hostname or IP, any TCP port, optionally a nickname, and a probing interval. We'll talk about how to pick the right probing interval in just a moment.

Next, we can select to use the probe on our main polling engine, or deploy a probe to a remote location. The probe is a small piece of software that runs the probing operations. The probe is always the source of your path.

Click create, and in a few minutes, you'll have your path. That's it!

Let's talk about selecting the right probing interval.

Longer probing intervals give NetPath™ more time to find paths and measure their performance. Simple LAN paths can be done very quickly while longer, complex paths, like those over the Internet, take longer.

Shorter probing intervals give a more granular view of your network, in terms of time. If a problem occurred during a 5 minute window, a 5 minute probing interval will find it where a 30 minute probing interval may not.

Shorter intervals also take up more storage, and cause more strain on your probes, pollers, and the rest of your SolarWinds infrastructure. You can always get more hardware, but good Engineers use only the resources they need.

If you're looking for the fastest results, our recommendation is start with 5 minutes for LAN paths and 10 minutes with Internet paths.

Now that it's configured, let's take a look at how to troubleshoot using NetPath.

After discovering the path, NetPath measures the performance characteristics of each network node and each network link, making it easy to spot slow downs.

Let's go to a path that we have already built - salesforce.com. Here we can see this router has two incoming links. Only one link is experiencing increased delay. By correlating real transit impact to the wealth of information from SolarWinds Network Performance Monitor, NetPath is able to highlight the root cause of the problem.

If you have other products such as Network Traffic Analyzer and Network Configuration Manager installed, their data will show up in the graph when relevant as well.

In this example, we see a config change occurred at the same time as the problem started. Selecting the issue shows the config comparison from Network Configuration Manager. If we scroll down, we will see that a traffic shaper has been inserted that is probably the cause of the slow-down.

Selecting any node allows you to navigate to Node Details for that node. The same is true for interfaces.

NetPath also provides visibility into Internet paths. In our example path to salesforce.com, we can see what companies we're depending on for SalesForce to work properly for us. By clicking on their network, we'll get an expanded view of the routers and links inside.

If the ISP were experiencing any issues, we can click on the node to obtain their contact information.  Now I can call them and provide precise, actionable information about the problem we're seeing in their network.

Ever have a user tell you service was down earlier, but it seems fine now? With NetPath monitoring your services, you can travel back in time to see what the network behavior was at the time of the outage.

The time line across the bottom gives you the end to end performance for the path. If you notice an anomaly, you can click the interval on the chart to see what the path looked like and how it was performing at that time. You can also use the forward and back arrows to move through intervals. Holding down the arrow will play through the intervals so you can see how the network changes over time.

NetPath can also help you understand your network's performance, but in a different way than you're used to.

Traditional monitoring tools test from a single point in your network to the various infrastructure components, but the most important thing your network is doing is delivering services to users. NetPath monitors connectivity from your users to the services they care about.

NetPath then determines what infrastructure is in the path, where traffic slow-downs are occurring, and provides additional infrastructure data only when it appears to be related to a real problem.

So, how do you configure NetPath to show a representation of the entire network's performance, focusing on delivery of services to users?

First, think about where you have large groups of users. Using SolarWinds as an example again, we have our Headquarters in Austin, Texas, and we have branches in Lehi, Singapore, Cork, Ireland, and Brno Czech Republic. We'd want to deploy a probe to all of these locations. At particularly large locations with segmented networks, for example our Headquarters, we may want to deploy a probe in each segment of the network that we think may likely receive a different network experience.

Next, think about the services that are important for our users. At SolarWinds for instance, SalesForce, JIRA, Exchange, NetSuite, Confluence, and file shares are some of the most important services. We'd want to add all of these to NetPath. Remember, NetPath can monitor the network connectivity to any TCP based service. What are the most important services for your users?

Finally, we assign our probes to our services.

With all of this done, we can start to monitor and understand how well our network is doing at delivering services to users. We can see service level issues or issues with a specific group of users. Anytime we are having a problem, we can jump into the graph and start solving.

 

Last modified
03:06, 19 Apr 2017

Tags

Classifications

Public