Technology news and Jobs arrow Information Technology News arrow Diagnose and manage your network for free with Linux and open source tools
Diagnose and manage your network for free with Linux and open source tools E-mail
by David M Williams   
Monday, 30 July 2007
traceroute


As widely installed as ping, but less well-known, is traceroute ("tracert" under Windows). This shows all the routers that data passes through on its way to a specified destination, passed in as the argument (e.g. traceroute www.cisco.com).

Although useful on your own local network, traceroute really comes into its own when trying to diagnose problems out of your control - where your traffic goes out onto the Internet. I've lost count of the number of times people have asked me for advice when they've believed their ISP was having a problem, only to be told by the provider's helpdesk to turn off and on their router, and to check their username and password. By using traceroute it can be proven that traffic is going out of their house and onto the ISP's network and then encountering a problem. Passing on the traceroute output to the ISP's technical team helps them work out where the issue is, and speeding its resolution.

Don't be surprised, though, if periodically traceroute can't identify a system it passes through. The program depends on networks allowing low-level IP responses to be returned and these can sometimes be turned off.

It's also important to understand that data won't always go the same way that traceroute's attempts figure out. This is because the whole basis of the Internet Protocol was to come up with a routing mechanism that could deliver data over a raft of different paths even if some nodes were lost due to conflict or other matters.

Nevertheless, as a basic - and always available - network fault finding tool, traceroute is invaluable. You might also like to check out mtr - Matt's TraceRoute - which combines ping and traceroute together as well as giving graphical output.

netstat


Once you have connectivity issues out of the way, netstat (network statistics) shows all active connections, both incoming and outgoing as well as other useful pieces of information.

Firstly, running netstat by itself lists hosts that are presently communicating with your computer and the port they are using. Couple this with the -p flag (-b on Windows) to see the name of the local process (or binary file) being used to perform the communication. You should see expected entries, like e-mail programs but you may also discover surprising processes that are chewing up bandwidth.

Next, run netstat with the -r flag; this gives entirely different output and shows the local computer's routing tables. This helps understand where the computer is sending packets to, based on the destination IP address.

There are many other useful statistics to be gained from netstat; it can give extended information. It can show Ethernet statistics. It can give figures for each network interface. It can print a one-off list or it can continue running, refreshing values each second. Unfortunately, the way in which these options are invoked - as well as how the results are formatted - varies from implementation to implementation. It is worth becoming familiar with the man page for your local flavour of netstat but always keep it in mind as an essential tool that shows the paths your computer uses for data transfer as well as all the network activity running on your machine right now.



 
< Next story in category   Previous story in the category >
iTWire user statistics Visitors last 30 days
694,279
Subscribers 15,210
#1 independent technology news advertise here
  •   *  
  • Search
  • AdvSeach
  • Login
  • Events
  • FreeStuff

- Advertisement -

Featured Whitepapers

Follow iTWire on Twitter

About iTWire

iTWire is all about technology news, information, jobs and community for the IT and telecommunications industry professional. Subscribe to our free ICT daily newsletter