With Cisco recently acquiring ThousandEyes, I had to take a look at this product to see what the hype was about. I got the hook up for a trial account and have been playing with it all day. While there is a TON more to this product, I wanted to detail out my first impressions and experience.
What is the use case?
Have you have had a customer/user call in about a issue only to discover theres a network outage somewhere out there on the internet? Yah me too… Almost anyone in the network world has experienced a downstream issue out there on the big sea that is the internet. We can try to trace the issue by having the customer run trace routes or calling our ISP.. however this hardly ever works. Most of the time this takes longer than the actual outage. Meanwhile we fly blind waiting for the phone… So in this modern world of networking there has to be a better way? right? Enter ThousandEyes!!!
What does it do?
ThousandEyes uses hundreds of sensors all over the world to give you thousands view points of your application. Hence the name! (See what they did their.. ok maybe not so funny). These Sensors can provide connectivity and latency reports from their given region. For example you could test that customers in Chicago can connect to your service. But heres where the magic happens. These sensor nodes can also track the entire BGP path to your service. Every single node between you and your customer. Latency and up down levels are checked and the IP/Provider of the node is shown. This allows you to quickly identify a internet outage impacting your customers and provides the details so you know who to contact and what node is the issue.
Lets take a look!
Alright lets dig in and see what this application can do. When you login you will be presented with your dashboards page (Figure 1). This dashboard will show all the active tests that are running. Tests are essentially SLA probes on steroids. We will dig into one of these next.
Next we will create our own test. To do this we will click “create your own test” under the tests section. Then click “Add a test” From here we will be asked for some basic information like test name, description, how often to run the probe (2 min default). See Figure 2 for my inputs. for this post I will be creating a test for my blog website “netevolve.io”.
Now things get interesting with the Agents selection. This menu allows you to select all the locations you want your app to be tested from. You can select cloud hosted nodes or physical nodes. For this test we will use 4 Chicago located nodes that are cloud hosted. See Figure 3
Now that the basic setup is done, we click on create test to save our config. If we navigate back to the dashboards page we should see our test there. See Figure 4.
If we open up the test, we can quickly see our up/down state over time all the back 14 days. See figure 5. We will see that all 4 agents currently show 100% uptime. This is no surprise since we just created the test! Using the metric drop down we can also see response time and throughput metrics over time. Figures 6 and 7 respectively.
Now lets get into the secret magic of ThousangEyes by selecting Path Visualization. This will show you the full node by node (AS by AS) path from the monitoring nodes we selected on the test setup your service (my blog for this test). Green nodes have no issues with there path to your service. If there is a issue, it will change to red. The node in the path with the issue will be identified and provider/IP information is given. See Figure 8. Wow how cool is that!? Imagine finding that info in the past? Now imagine the ability to proactively monitor all of these paths from any country consistently. Pretty sweet!
Before going into this I honestly didn’t know much about ThousandEyes other than it was one more monitoring tool. After playing around with it and creating some tests… Wow am I impressed. While the up/down and throughput numbers are great on there own, for me what makes this product truly unique is the provider path monitoring and the number of nodes we can select from. Th his allow the business to consistently monitor critical apps from various locations and provide immediate identification of where a issue occurs. For any SaaS or app provider, I cant think of a better way to improve visibility and response for customer experience. Without ThousandEyes you are blind to issues that occur the closest to your customer and outside of your control. 10/10 ThousandEyes!