Most administrators monitoring a network-connected server like to have live statistics about the machine's bandwidth utilization -- how much is being used and by what functions. For instance, if you have a server hosted at a remote NOC, the host usually provides a network statistics package. But if you're responsible for your own network, you'll have to devise statistics entirely on your own.
MRTG requires Perl to run under Windows. The Perl components are strictly for the sake of assembling and formatting the reports; routines written in C for speed and portability do the actual gathering of SNMP stats. Versions are also available for Unix and Novell NetWare machines. The target device being polled must have SNMP enabled and available (such as a Cisco router).
Over time, features have been added to MRTG, which include the use of SNMPv2c 64-bit counters (so counters don't roll over prematurely); the ability to identify the target interface by multiple criteria, not just interface number; and the ability to report back bandwidth usage by service and port. This would allow you to determine how much of your bandwidth is being used by e-mail versus Web services, for instance. Incidentally, this tool can monitor any SNMP variable, not just bandwidth; bandwidth is just one of the possible variables.
If you're curious, here are some examples of MRTG in use in various environments.
Serdar Yegulalp is editor of the Windows Power Users Newsletter. Check it out for the latest advice and musings on the world of Windows network administrators -- and please share your thoughts as well!
More information from SearchWinSystems.com
- Ask the Expert: How can I monitor SMTP traffic coming from all machines?
- Ask the Expert: How much NetBIOS traffic is normal?
- RSS: Sign up for our RSS feed to receive expert advice every day
This was first published in December 2005