One of my networking classes included the basics of troubleshooting a network connection. It included basic command line tools that really were very easy to use and understand. Most of the students thought they were fun to use.

Knowing these commands before having to deal with the frustration of a network connection problem is not a bad idea. A working connection shows you what is supposed to happen. If you see something different, you will know something is up. When problems strike, at the least you can get an idea of what is going on.

Step 0: Check the Cords & Power

The first thing you should always do is check to make sure everything is plugged in: your computer, router, device, etc. Many laptops have a button to turn off the wireless connection; the icon looks like a signal tower. When in doubt, read the manual (which you can often download from the manufacturer’s website). Bear in mind, we all forget this step at some point, so do not feel too silly when it happens to you.

Step 1: Ping Yourself

You want to test that your machine is working properly. To do this, you want to ping yourself. You use the loop-back address (127.0.0.1) to do this. Pinging the loop-back address tests to make sure software on your computer is working properly. Typically, if something is not working at this stage, you may just need to restart your computer.

ping
http://www.ghacks.net/2011/05/12/network-troubleshooting-basics-the-ping-command/

Step 2: Ping Your Router (AKA: the Default Gateway)

The next step would be to ping your router. You can find your router’s IP address with ipconfig as well (it should be on the bottom of the unit and listed in the manual too). Remember that ipconfig lists your router as the “Default Gateway.” It is very likely to be 192.168.1.1 or a similar number.

This is done to test if your router is responding. If it is not, and you have already checked to make sure it is on, then it may need to be turned off and turned on. Every once in a while it may need a refresh. If the problem continues, contact you ISP for assistance to see if they can help.

Note: Routers, like anything else in the IT world, will someday fail. There is a chance that this has happened. Your ISP help desk should be able to help you determine this. If the router is one of theirs, they will likely send you out a new one.

Step 3: Ping Yourself with Your IP Address

We want to test to make sure everything is working correctly between your router and your computer. To do this, ping your IP address. It is listed in the ipconfig command at the same time the router IP number is. If this works, you can be pretty confident that a problem is outside your home (or maybe small office) network.

Step 4: Ping and Tracert Outside Your Network

From here, you want to test something outside your network. In a medium or larger network setting, a server on another branch of the network will do. For a home network, the Internet is often your only option. Since chances are the problem is that one or more websites are (or seem) down, this is a logical thing to check.

You can use a few different tools. First try the ping command because it is the fastest. It will only tell you if the site is working or not. For more detailed information, use tracert and pathping. They can give a better idea of what is going on. For instance, if you can reach your router, but no further, the node that connects you to the Internet may be down: an ISP issue. If you can reach only a couple (one or two) steps past your router, then it still is probably an ISP issue. Your Internet is down. If you can reach a lot further, the site may be having issues (see is Facebook down, ways to check if a website is really down, for pointers) There even is a chance that other sites, either from the same proximity or even hosting service, are having the same problem. If it is an e-commerce site, there may be a number to call, but there probably is not much to do besides wait. At least you know where the problem is.

Parting Thought

On a parting note, one other benefit of trying these commands out on a nice day like today is that it creates a positive mental association with the tools. That might change with enough problems, but they are worthy of a good first impression. Having the skills to deal with a problem ahead of time is always a good idea.

Both of these commands are designed to give you more detailed information than the ping command. They will tell you about steps along the way. This can help you narrow down where a problem is.

The Tracert Command Basics

The tracert command is also called traceroute on other systems, such as on a Mac or in Cisco’s IOS (used in its routers and switches). It is used to find out what other devices are on the path to a destination. It works by sending out a number of signals. Each signal has an amount of locations it can jump to, called a time to live (TTL). When it reaches that number, the device it reaches sends back an error message. For example, the first packet of data sent out would just have one hop, so it would stop at your home router. Your router sends back a signal, and the tracert command records it. The next signal would have to go two hops and so on, until the destination is reached.

Using the command is simple. You type in tracert and an IP number or website’s name after it. You can see extra options by typing tracert /? in the command line.

To get the the command line, go to Start, Run, type cmd in, and press the Enter key.

After that, type tracert google.com in and hit Enter again. You should get something like this:

In Linux, the equivalent command is tracepath, but traceroute is available. Both commands perform similar actions, but you have to install traceroute on some systems (e.g. Ubuntu). Type in “sudo apt-get install traceroute” from the terminal to grab it. Enter the name of the command and a IP number or a website’s name to trace the path.

The Pathping Command Basics

The pathping command is more similar to tracert than a simple ping. Just like the tracert command, pathping will give you information about all the steps along the path to the website you want to test. It does this by sending out full ping commands along the way, gathering more information. This is done over a period of time, yielding a detailed report. The extra information is good to have, but the time required is a disadvantage. It can take a few minutes. Typically, tracert is the preferred command.

The command works the same way that tracert or ping do, where you type in the name of the command and the name of the site you want to test. There are some option, which you can see by typing “pathping /?” into the command line. For something similar in Linux, use the mtr command.

Assuming you did not close the command line window from befor, type “pathping google.com” in and hit Enter. You should get a report similar to tracert, but it will tell you to wait for the extra results. There is a lot of information, so you may have to scroll a little to see it all.

Controlling Pathping

Waiting roughly 6 minutes for a report is a bit much for one command. You can shorten the time to create the report with a command line option. Instead of sending 100 pings, “pathping google.com -q 10″ will send 10 per hop, or “pathping google.com -q 3″ will send 3 per hop.

Cleaning-Up Tracert or Pathping Output

Tracert command is notorious for lines that spill past the width of the screen; the pathping command never fits in a screen. By saving the output to a file, it will remove line breaks automatically. There is no manual editing involved, and the output is nicer.

To do this for tracert, type in “tracert google.com > file.txt” into the command line. If you are not familiar with Dos, or you are not sure where you put the file, you can use “notepad file.txt” (or even just “file.txt”) in the command line to easily retrieve the file.

There is one problem with this method: it looks nothing is happening. This is even more evident with the pathping command, which can take several minutes to run (hence why controlling the pathping is covered above). Be patient; this is normal.

Side note: This is call “redirecting,” and it is commonly done at the command line. Use one greater-than sign (>) to overwrite a file; use two greater-than signs (>>) to append data to a file.

This can also be done in Linux, but it is not necessary. If the tracepath command spills over the edge, you can re-size the terminal and run the command again. The Dos prompt has a fixed character width, so resizing the window will not help.

What You Can Infer

Tracert and pathping are useful in seeing where a problem is. If a command recognizes your router, but not much further, something is probably wrong with your ISP. If a command trails off, but cannot find the destination, the website probably is having a problem. Typically, it is better to start testing with the ping command because it faster than its counterparts.

Closing

There is going to be one last article in this series that puts all three commands together. Each one is a piece that you need to understand in order to use it. Combining them can make finding a problem easier.

The command has a few disadvantages. It is not displaying additional information about each hop and it is kinda difficulty to process the information. Web Hopper makes use of a graphical user interface and several additional features that might make it worth the download. The country of each server on the route to the destination is shown. It is sometimes quite fascinating to see that connections to a server “next door” will make a trip around the world before they reach the destination. This can be used to analysis connection problems like bad pings to game servers.

Another advantage of Web Hopper is that you can also traceroute emails which the traceroute command in Windows cannot.

Web Hopper displays the country of the server, the server name and the time it takes to reach it. A trace is updated automatically and results can be saved in text format for further processing. The Internet traceroute tool comes with two shortcuts that can grab the active browser url to process it automatically and the option to view a url in the browser.

Just Traceroute (via Digital Inspiration) takes this a step further by issuing a traceroute command from multiple locations in the world. Different locations, especially when they are from different parts of the world, use different “ways” to reach their target which could give additional information about the cause of the connection problem.

The services is using servers located in the United States, Netherlands, Australia and Singapore to run the traceroute command.

The results can be forwarded to an email address. The very same window offers a way to copy the all four traceroutes to the clipboard directly.

I usually restart my router to be on the safe side but that’s it. Sometimes you can make a connection to the Internet but that connection feels slow, you get disconnections and websites that are not loading properly.

One way to find out if the problem is located on your side is to perform a tracert command to the target server. If that tracert times out you pretty much know the IP address of the server responsible for that along with information about the country the server is in.

This is also the basic approach of Hubble, a service that monitors Internet reachability in real-time. Instead of using one tracert at a time they send out more than 100k every 15 minutes to monitor reachability problems throughout the world using a Google Maps mashup to provide a map of so called black holes, meaning servers that are unresponsive or have a reduced reachability.

The interesting information for the users are the IP addresses or ranges of those servers and the country column of the table. Sorting the column by country reveals problematic server at a glance which can help determining if the problem is located at your end.

Most operating systems have those tools on board and it is normally not required to use an external tool. Many users however do not like to work with the command line in Windows and prefer a graphical user interface instead. Many also have difficulties issuing the correct commands, e.g. finding out how to save the result in a text document automatically.

eToolz is a collection of network and Internet tools that provides a graphical interface for several common commands. This includes ping and tracert but also DNS and whois lookups, mail checks, the retrieval of http headers, time synchronization, pagerank check and list with useful information (default ports, domain extensions).

Most functions come with several options. It is for instance possible to select one of six whois databases, define timeouts, ping counts or the maximum amount of hops of a tracert. Every echo in the program can be copied normally. Five languages are supported: German, English, French, Portuguese and Dutch although a few of the information are still not translated and therefor in German.

The developer of the product uses Google Translate to provide an English version of the homepage, ever heard of Nazi lookups ? Yeah, me neither.