[168704] in North American Network Operators' Group
Re: Do network diagnostic tools need upgrade?
daemon@ATHENA.MIT.EDU (Jared Mauch)
Mon Feb 3 14:15:25 2014
From: Jared Mauch <jared@puck.nether.net>
In-Reply-To: <52EFE702.2030103@alvarezp.ods.org>
Date: Mon, 3 Feb 2014 14:15:02 -0500
To: Octavio Alvarez <alvarezp@alvarezp.ods.org>
Cc: nanog list <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
On Feb 3, 2014, at 1:59 PM, Octavio Alvarez <alvarezp@alvarezp.ods.org> =
wrote:
> On 02/03/2014 05:33 AM, Ammar Salih wrote:
>> Hello NANOG list members,
>>=20
>> I have a question for you, are you happy with the current network
>> diagnostic tools, like ping, trace route .. etc,
>=20
> What tools are you referring to by "..."? There are many others. I =
like
> tcptraceroute (there are two variants of it) and mtr.
There are lesser known options that are used by folks, eg: ping =
record-route.
One could certainly use those available tools, but most folks have a =
hard
enough time interpreting traceroute output. I've seen customers =
complain
about performance to have us show them it's on their network, or their =
firewall
modules, etc..
Having statistics on network usage/errors/drops is incredible useful in
isolating the performance limitations. Knowing that a firewall maxes at =
350Mb/s
is as equally useful as having protocol extensions to collect the data.
One of my early experiences with a sysadmin who only cared about the =
application/OS
was "the router is a black box that gets my packets there". Knowing the =
behavior beyond
there is also important (how latency/loss impacts tcp/udp/application =
performance for
example).
Most importantly, keeping an open mind when troubleshooting is helpful. =
Sometimes
you find something unexpected. (eg: uRPF drops when responding IP is =
mapped-v4-in-v6
from within 6PE network).
- Jared=
