Some of the best troubleshooting tools aren’t in your toolbag. In this video, you’ll learn about speed test sites and looking glass sites.
<< Previous: Physical Testing ToolsNext: Protocol Analyzers >>
A tool I use quite often are bandwidth speed test sites. These are able to test how much traffic you’re able to push through a particular internet connection. It effectively is transferring a file and it’s checking to see how long it takes to transfer that file through your particular internet link. The speed test sites can also be useful if you’re planning to implement some infrastructure equipment.
Maybe you want to know what the difference will be once you install a new firewall or a new package shaper, so you would run a speed test, you would install that device, and then run the same speed test again to see if you can see any differences in throughput. It might also be useful to measure at different times of the day. You may want to know if there is a difference in throughput in the mornings versus the afternoons versus the evenings, and based on that, you may want to decide how you want to implement some of your file transfers.
Not all of these speed test sites work the same. There are a number of different servers located in different places. We call those a point of presence or a POP. There may be different amount of bandwidth available at that POP, and the way that it tests may be a little bit different between different providers. Here’s a list at some of the more popular speed test sites.
I like speedof.me, this is my speedof.me results. There’s speedtest.net, testmy.net. Sometimes your ISP themselves have sites you can go to test your speed. For instance, speedtest.comcast.net and AT&T has a site– att.com/speedtest.
If you’re an ISP, one of your biggest troubleshooting challenges revolves around routing tables. Routing tables, especially at an ISP, are remarkably complex. There are a very large geographical areas, there’s new technologies being used for these routes and routing tables all the time, and there’s all kinds of different equipment that’s used between different data centers, and all of those can have an impact on the routing table. Another challenge is that there are many different ISPs with many different connections and many different people supporting all of these links.
If you’re trying to troubleshoot a routing issue, trying to find the right person to talk to is nearly impossible. That’s why many ISPs have created Looking Glass servers that allow you to get access to their routing table so that you can begin the troubleshooting process without needing to contact anyone directly. Here’s Sprint’s Looking Glass site. You can visit this yourself at sprint.net/lg for Looking Glass. They’ve got a number of options available.
This first option is to choose the source where this query will be made from. Let’s do an IPv4 query, and let’s say that this query is being hosted out of Orlando, Florida. Then you can choose which command you’d like to run. Would you like to run a ping, a trace route, or a number of BGP options? Well, let’s show some BGP routes from the Orlando, Florida location.
And let’s choose just a generic 18.104.22.168/8 for the subnet, and we’ll submit this command. Now this command is submitted to the Orlando, Florida site. It’s going to perform that show route and then you can see all of the information associated with the BGP route between Orlando and the network that was specified in the Looking Glass. As you can see, using a Looking Glass front end allows me to run a number of different troubleshooting commands from many different locations all without contacting the ISP.