On Tue, Sep 06, 2005 at 01:16:59PM -0400, andrew2@one.net wrote:
owner-nanog@merit.edu wrote:
Best Practices of wide-area diagnosis, anyone?
I'd be interested in a discussion of this as well. To answer a slightly different question, I usually point the "ping and traceroute" geeks to Karl's wonderful treatise on the subject: http://www.iwl.com/Resources/Papers/icmp-echo_print.html.
i've found it useful to use a simple udp probe tool to test networks in the past. You can test end-to-end loss and get something reasonable. The following expects you to know: 1) GCC/Makefiles 2) how to insure you link in your resolver and socket/nsl functions 3) tweak your cpu compile options for your host.. but.. ftp://puck.nether.net/pub/jared/rtt-0.12.tar.gz If your clocks are accurately synced, you can even get unidirectional delay. I usually run it like this: ./rtt -v <host> you will need to run ./rtt_resp on the far end host. You can also use iperf or similer tools to help customers diagnose network problems, but a easy/lightweight daemon on a few hosts is always fairly easy to play with in a quick-and-dirty way... - jared -- Jared Mauch | pgp key available via finger from jared@puck.nether.net clue++; | http://puck.nether.net/~jared/ My statements are only mine.