Forum OpenACS Q&A: Network performance monitoring tools - tcpdump?

Collapse
1: Network performance monitoring tools - tcpdump?
Posted by Brian Fenton on 08/26/04 01:22 PM
Hi,
I have a client who is reporting regular poor performance on a system. Whenever I look at the system, the performance is fine for me. I suspect it's a networking issue and would like to verify that before proceeding any further. I've heard people mention tcpdump here before - is this the best tool to use? The clients are all on Windows PCs, so I guess I'll need to use WinDump.

Any good tcpdump references or tips?

Any other good tools available?

many thanks,
Brian

Collapse
2: Re: Network performance monitoring tools - tcpdump? (response to 1)
Posted by Bruno Mattarollo on 08/26/04 03:07 PM

Brian,

tcpdump is more a network analysis tool that network performance tool. It really depends on what you want to analyze or where you think the problem is. tcpdump could indeed help you gather information about what's happening network-wise when the user sees the performance problems but won't tell you too much about network performance more than what's going through the wire (and what the computer can listen to). For a good GUI to tcpdump, look at Ethereal.

If you have a router in the network that can output SNMP or a linux machine, then you could gather information from there and plot it with mrtg, it will tell you more about the amount of traffic that you have there.

Just my two cents, hope this helps.

Collapse
3: Re: Network performance monitoring tools - tcpdump? (response to 1)
Posted by Patrick Giagnocavo on 08/26/04 03:40 PM
Brian, I would suggest you set up "sar", a unix tool available on all kinds of Unices, and have them write down the exact time(s) when there is a problem.  Under Linux, I believe there is a "sadc" daemon which gathers the info.

You can then run sar, which gathers its stats every 15 minutes, and try to track down the problem.  Sar has stats for CPU, disk, network packets among other things; so simply looking at the data samples for the affected times will tell you if there was a jump in disk, CPU, or network usage.

Aside from tcpdump, MRTG would definitely be recommended.

Another thing to check might be DNS settings on both server and clients - if the DNS is slow to respond then the server will "seem" slow.  On a small internal network you can add the server to the hosts file on each client and see a small speed boost.